Regional Tissue Perfusion Index (RTPI): A New Optical-Based Metric for Quantifying Regional Tissue Perfusion

Regional Tissue Perfusion Index (RTPI): A New Optical-Based Metric for Quantifying Regional Tissue Perfusion | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Regional Tissue Perfusion Index (RTPI): A New Optical-Based Metric for Quantifying Regional Tissue Perfusion Babak Shadgan, Iman Amani Tehrani, Sadra Khosravi, Zahra Askari, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7403210/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 8 You are reading this latest preprint version Abstract Purpose Accurate, continuous assessment of regional tissue perfusion remains a significant clinical challenge, as most existing modalities are invasive, indirect, or impractical for routine monitoring. Near-infrared spectroscopy (NIRS) has been widely adopted to assess tissue oxygenation; however, conventional NIRS-derived indices are insufficient surrogates for true perfusion and often fail to capture rapid hemodynamic changes. This study aimed to introduce and validate the Regional Tissue Perfusion Index (RTPI), a novel NIRS-derived metric that integrates multiple features of the NIRS signal to provide continuous, non-invasive, and physiologically relevant assessment of tissue perfusion. Methods RTPI was developed using principal component analysis (PCA) of multiple NIRS-derived parameters, including pulse amplitude ratio, signal derivatives, and area under the curve. Its performance was evaluated in healthy volunteers during controlled ischemia–reperfusion protocols and compared with established reference standards, including laser Doppler flowmetry (LDF) and photoplethysmography (PPG). Partial least squares (PLS) regression was also applied to test the robustness of the approach. Results RTPI showed strong correlations with LDF and PPG during dynamic perfusion changes. Unlike conventional NIRS-derived oxygenation and hemodynamic indices, which often exhibited delayed or paradoxical responses, RTPI demonstrated immediate and significant sensitivity to both complete and partial ischemia–reperfusion episodes across all cases. Intraclass correlation and error analyses confirmed high test–retest reliability and low measurement error. Comparable performance between PCA- and PLS-derived indices further supported robustness and generalizability. Conclusion RTPI represents a multiparametric, physiologically meaningful, and computationally efficient metric for real-time tissue perfusion monitoring. Its ability to detect perfusion compromise independently of oxygenation indices highlights its translational potential for bedside implementation in critical care, trauma, perioperative, and vascular medicine, where improved diagnostic accuracy could significantly impact patient outcomes. Blood Flow Microvascular Monitoring Near-Infrared Spectroscopy Principal Component Analysis Tissue Perfusion Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 INTRODUCTION Tissue perfusion, the delivery of blood through capillaries to biological tissues, is indispensable for supplying oxygen and nutrients and removing metabolic waste, thereby sustaining cellular metabolism and homeostasis [ 1 ]. Adequate perfusion is a critical determinant of tissue viability and organ function across all organ systems [ 2 ]. Conversely, perfusion deficits underlie a broad spectrum of disorders, including tissue ischemia and hypoxia, circulatory shock [ 3 ], ischemia-reperfusion injury [ 4 ], compartment syndrome [ 5 ], pressure ulcer formation [ 6 ], and impaired wound healing [ 7 ]. Hypoperfusion compromises cellular integrity and can escalate to multiple-organ dysfunction and systemic deterioration [ 8 ]. Accordingly, a reliable assessment of tissue perfusion is a central objective in critical care [ 9 ], postoperative monitoring [ 10 ], emergency medicine [ 11 ], vascular surgery [ 12 ] and rehabilitation [ 13 ]. Despite its clinical importance, accurate assessment and monitoring of tissue perfusion, especially at the microvascular level, remains a longstanding challenge in modern medicine. Current methods are often invasive, intermittent, or provide only indirect surrogates of perfusion. Techniques such as arterial catheterization, thermodilution, and radiolabeled tracers, although valuable in specific clinical settings, involve procedural risks and are not suitable for localized or continuous monitoring. Thermal-diffusion probes (e.g., Hemedex QFlow™) provide absolute, real-time, quantitative perfusion measurements directly within tissue, an invaluable tool in neuro-ICU care; however, a catheter must be implanted into the brain or soft tissue, limiting its use to operating rooms or ICUs [ 14 , 15 ]. Near-infrared spectroscopy with indocyanine green (NIRS-ICG) applies the Fick principle to yield site-specific blood-flow values; nevertheless, it still requires repeated intravenous dye injections, preventing truly continuous bedside use [ 16 , 17 ]. Collectively, these modalities are precise yet impractical for routine, multi-site, or prolonged monitoring because of their procedural burden. Non-invasive techniques that avoid vascular access are safer and more widely available; however, each method sacrifices depth, stability, absolute quantification, and accurate regional perfusion measurement. Conventional systemic surrogates, such as standard hemodynamic measurements (e.g., heart rate, blood pressure, and blood oxygen saturation), are widespread; however, they fail to detect spatial or temporal variations in tissue blood flow. Furthermore, they are designed for systemic evaluations rather than localized perfusion monitoring, which limits their utility in detecting subtle regional hemodynamic changes [ 18 , 19 ]. Laser Doppler flowmetry (LDF), a widely used optical technique for assessing microvascular perfusion, has provided valuable insights into skin and superficial tissue blood flow by detecting Doppler shifts caused by the movement of red blood cells. While LDF offers excellent sensitivity to changes in perfusion, it suffers from several important limitations. These include a shallow and variable penetration depth (typically ~ 1–2 mm), limited spatial resolution, susceptibility to motion artifacts, and difficulty in standardizing absolute perfusion values across subjects or tissue sites. Consequently, LDF is primarily used in research or niche clinical applications rather than routine monitoring [ 20 , 21 ]. The photoplethysmography (PPG)-derived perfusion index (PI), defined as the ratio of the pulsatile (AC) to non-pulsatile (DC) components of the pulse oximeter waveform, is designed to assess tissue perfusion at the pulse oximeter site [ 22 ]. The PI is easily obtainable at the bedside; however, it shows significant variability between healthy and critically ill individuals. Furthermore, differences in PI between fingers within the same person reduce its reliability as a stand-alone measure of tissue perfusion [ 23 , 24 ]. Diffuse correlation spectroscopy (DCS) can non-invasively probe centimetre-deep microvascular flow and shows good correlation with invasive cerebral flow probes; yet, current instruments remain bulky and costly, limiting their clinical uptake [ 22 ]. Hence, even the best non-invasive tools struggle with depth, motion or practicality constraints, underscoring the need for a dye-free, motion-tolerant, continuous index of microvascular perfusion. The ability to continuously and non-invasively monitor local tissue perfusion in a practical clinical setting remains an unmet need. Such a capability would allow for earlier detection of tissue damage, help customize hemodynamic treatments, and enable real-time monitoring of therapy effectiveness. It would also broaden perfusion monitoring to include outpatient, post-surgery, and rehabilitative settings, leading to better outcomes for a wide range of patients. This gap in clinical practice highlights the need to develop new physiologic perfusion indices that are accurate, reliable, and suitable for scalable, non-invasive platforms. Near-Infrared Spectroscopy (NIRS) has emerged as a valuable non-invasive modality for assessing localized tissue oxygenation and hemodynamics. To date, the majority of NIRS applications have focused on parameters related to tissue oxygenation, including oxygenated hemoglobin (O₂Hb), deoxygenated hemoglobin (HHb), hemoglobin difference (Hb-diff), and tissue oxygen saturation (StO 2 ) or tissue oxygenation index (TOI). However, using these parameters, particularly TOI, as surrogates for tissue perfusion is not accurate. Although TOI is often interpreted as an indirect marker of perfusion, this assumption is flawed, as clinical scenarios exist in which changes in TOI do not correspond to actual alterations in tissue perfusion [ 23 ]. Similarly, some studies have proposed total hemoglobin (THb) as a surrogate for perfusion; however, this is also a misconception. THb primarily reflects local blood volume, and its fluctuations indicate blood pooling or volume shifts rather than perfusion dynamics [ 24 , 25 ]. Currently, conventional NIRS methods lack a reliable and specific index for accurate measurement or monitoring of regional tissue perfusion [ 26 , 27 ]. To address this gap, we have developed a novel physiological metric, the Regional Tissue Perfusion Index (RTPI), derived from NIRS signals and designed to quantify tissue perfusion changes in real time. Unlike oxygenation-based indices, RTPI directly reflects dynamic changes in blood flow, independent of tissue oxygenation indices. Our approach extracts physiologically meaningful features and combines them using unsupervised principal component analysis (PCA). Comprehensive statistical analysis confirms both the clinical relevance and statistical significance of our results. In this manuscript, we present the conceptual foundation, methodological framework, and validation results of the NIRS-derived RTPI measure. We evaluated the real-time performance of RTPI against established references, including LDF and PPG, under experimentally controlled ischemia–reperfusion conditions in healthy participants. Our findings highlight the clinical potential of RTPI as a novel, noise-resistant measure of tissue perfusion and outline future directions for its development and translational application in both research and medical practice. Ongoing and future studies will focus on incorporating advanced feature extraction techniques and more sophisticated modelling approaches to further enhance the accuracy and reliability of perfusion assessment. METHODS Participants’ Recruitment Twenty healthy adult volunteers (14 males, 6 females) were enrolled in the study. Each participant completed two recording sessions, spaced one week apart, yielding a total of 40 datasets. This inter-session interval was selected to allow for full hemodynamic washout and to minimize potential carryover effects. All study procedures were approved by the institutional research ethics board and conducted in accordance with national guidelines for research involving human subjects. Written informed consent was obtained from all participants prior to data collection. Experimental Setup The complete hardware layout is illustrated in Fig. 1 . In this experiment, three fingertips on the right hand were used to monitor three adjacent similar tissues with NIRS, along with LDF and PPG pulse oximeter reference sensors (Fig. 1 B). A high-power LDF probe (Moor VMS-LDF1-HP; Moor Instruments, Axminster, United Kingdom) was affixed to the index fingertip, providing a continuous, high-temporal-resolution measurement of capillary blood flux at a 40 Hz sampling rate. To obtain arterial reference parameters, a finger-clip pulse oximeter sensor (MightySat; Masimo Corporation, Irvine, California, USA) was secured to the middle fingertip, delivering the manufacturer’s proprietary PI at 0.5 Hz. A custom-built, continuous-wave (CW) NIRS sensor in a spatially resolved configuration, developed to monitor free tissue transfer (FTT) surgical flaps [ 28 ], was positioned on the thumb fingertip to record tissue oxygenation and hemodynamics at a 64 Hz sampling rate. Experimental Protocol Each recording session was structured to impose two well-controlled ischaemic challenges on the participant’s right upper limb, thereby permitting a detailed characterization of the resulting perfusion dynamics. The protocol began with a five-minute baseline measurement. Participants lay on their backs and relaxed during this period, allowing cardiovascular variables to stabilize. Immediately after baseline, a phase of complete ischemia was induced. A digital pneumatic tourniquet (Delfi PTS, Delfi Medical, Vancouver, Canada) encircling the upper arm was inflated to 200 mmHg and maintained at that pressure for two minutes, thereby fully occluding arterial inflow. Once the cuff was released, a ten-minute period of reperfusion followed. This interval captured the expected reactive hyperaemic response and provided a window in which to quantify flow-mediated recovery. The second challenge started with partial ischemia. Here, the tourniquet was reinflated to 100 mmHg, a level sufficient to restrict venous return but not to stop arterial inflow, thereby causing venous congestion. This partial occlusion was sustained for three minutes, after which the cuff was deflated for a further ten-minute reperfusion period. The final reperfusion phase allowed us to observe how perfusion parameters returned toward baseline following a less severe but longer blockage. This sequence, comprising baseline, full occlusion, maximal reperfusion, partial occlusion, and a second low-level reperfusion, established distinct hemodynamic states that enabled a comprehensive assessment of microvascular responsiveness. Data Processing and Analysis To derive the RTPI, we began by analyzing the THb signal captured on the near channel of our NIRS sensor (source-detector separation of 10 mm), which records changes in hemoglobin concentration and derived tissue oxygenation indices within a superficial vascular bed that roughly matches the penetration depths of both the LDF and PPG sensors. Aligning penetration depths eliminates confounding differences in sampling volume and establishes a rigorous basis for subsequent comparisons. From the THb waveform, we extracted three physiologically meaningful features, each chosen to emphasize a distinct physiological aspect of microvascular perfusion: 1. Pulse-Amplitude Ratio (PAR) PAR quantifies the relationship between the pulsatile component of the hemodynamic signal and its slower baseline drift, providing a sensitive measure of microvascular perfusion dynamics. The THb signal was decomposed into alternating current (AC) and direct current (DC) components using frequency-domain filtering: the pulsatile AC component (cardiac-cycle oscillations) was isolated using a second-order Butterworth band-pass filter (0.5–5 Hz), while the slowly-varying DC baseline was extracted using a second-order Butterworth low-pass filter (0.5 Hz cutoff). The Butterworth filter, which is the default in many recent NIRS pipelines, offers a maximally ripple-free response pass-band, thereby preserving both the low-frequency baseline and the high-frequency cardiac morphology of the NIRS waveform while minimizing phase and amplitude distortion when implemented in a zero-phase configuration [ 29 ]. Figure 2 A illustrates the decomposition of the THb signal into its constituent components, displaying both the combined signal and the extracted DC component, while Fig. 2 B presents the isolated AC component. Peaks and valleys in the AC component were identified using physiologically constrained detection parameters, including a minimum interpeak distance of 0.25 seconds and a maximum distance of 2 seconds (heart rate boundaries), with height, width, and prominence thresholds empirically optimized for reliable feature detection. For each peak, the vertical distance to its immediately subsequent valley was measured and normalized by the DC value interpolated at the temporal midpoint between the peak and valley, yielding an AC/DC (pulse-amplitude) ratio. In PPG, an optical technique that shares the fundamental principles of optical physics with NIRS, the same AC/DC metric (PI) has been validated as a bedside marker of peripheral blood flow and vasomotor tone [ 30 ]. PAR was calculated within non-overlapping 2-second windows to ensure sufficient cardiac cycles for reliable estimation while maintaining temporal resolution. Within each window, all normalized peak-to-valley amplitudes were averaged to produce a single PAR value, with windows containing no detectable peaks assigned zero values to maintain temporal continuity. 2. Area Under the AC Curve (AUAC) While PAR quantifies the relative amplitude of cardiac pulsations, AUAC measures the absolute hemodynamic volume displaced during each cardiac cycle. To ensure physiologically meaningful values and reduce susceptibility to noise and artifacts, AUAC was calculated exclusively for validated cardiac peaks. Within each 2-second analysis window, peaks were identified using the same detection criteria as for PAR. For each confirmed peak, the area under the positive portion of the AC waveform was computed, excluding negative deflections to avoid spurious contributions from signal asymmetry. The trapezoidal rule was applied within the temporal bounds of each peak to yield a robust estimate of cumulative blood volume changes associated with individual cardiac cycles. AUAC metrics are physiologically meaningful, as the area under the NIRS parenchymal pulse (computed relative to an arterial reference) has been employed to quantify cerebral blood volume and hence regional cerebral perfusion in brain-injured patients [ 31 ], whereas the area under the PPG waveform has been shown to correlate with stroke volume and therefore peripheral perfusion in healthy volunteers [ 32 ]. 3. Derivative (DRV) DRV quantifies the temporal rate of change in perfusion by measuring how rapidly the total hemoglobin (THb) signal varies over time. The THb waveform was segmented into short, non-overlapping windows, and within each window, the first derivative was computed as the point-to-point difference normalized by the sampling interval. These derivative values were then averaged within each 2-second window to reduce high variability and noise while preserving physiologically meaningful trends in perfusion dynamics. First-derivative (slope) metrics are perfusion-relevant because the upslope of the NIRS THb trace shows a strong correlation with Doppler-measured peak muscle blood flow during post-exercise reperfusion [ 33 ], and moving-slope features extracted from the first and second derivatives of the PPG waveform correlate directly with mean arterial pressure, which is an established bedside index of systemic perfusion [ 34 ]. Together, PAR, AUAC, and DRV create a multidimensional representation that comprehensively characterizes tissue perfusion dynamics. During post-processing, the THb, PI, and flux signals were first aligned to their respective software-derived timestamps. To ensure temporal coherence across modalities, all three signals were then cross-aligned using a physiologically distinct landmark, the abrupt onset of reperfusion following complete ischemia. While the temporal offset between flux and THb was minimal (≤ 0.5 s in most cases), likely reflecting slight discrepancies in manual acquisition timing, the PI signal consistently lagged behind by at least 16 seconds. PI delay, previously reported in the literature [ 35 ], is presumed to result from the signal-enhancement algorithm applied to the PPG waveform. Each windowed feature vector was standardized using a robust scaling approach, which centers the data by subtracting the median and scales it by the interquartile range. Unlike traditional z-score normalization, this method down-weights the influence of outliers, thereby reducing the impact of motion artifacts and transient sensor dropouts. Importantly, it preserves the relative amplitude relationships among the three features, maintaining the integrity of their physiological contributions to the composite index. The scaled triplet entered a principal-component analysis (PCA). Because PCA creates orthogonal axes that capture descending proportions of the total variance, the first principal component (PC1) represents the single direction in feature space that explains the greatest amount of shared fluctuation among PAR, AUAC, and DRV. We treat this PC1 score as an unsupervised composite and refer to it throughout as the RTPI derived by PCA (PCA-RTPI). For each visit, we computed two accuracy metrics: the Pearson correlation coefficient (r) and the root-mean-squared error (RMSE). Correlation coefficients were first converted to their Fisher-transformed values (z = arctanh r), and these z-scores were used for all group-level statistics. The Fisher transform stabilizes the variance of r and produces an approximately normal sampling distribution, which justifies parametric pooling and the construction of symmetric 95% confidence intervals [ 36 ]. To quantify session-to-session change, we applied a two-tailed paired-sample Student t-test to the vector of within-participant differences (visit 2 minus visit 1). A Shapiro–Wilk test confirmed that these difference scores were normal, so no non-parametric substitute was required. Reliability across visits was evaluated using the intraclass correlation coefficient (ICC) model 3,1, a two-way mixed-effects, single-measurement coefficient. This approach is recommended when the same fixed sessions, such as one baseline and one follow-up for each participant, are compared and when the focus is on consistency rather than absolute agreement [ 37 ]. Moreover, individual Fisher transformed correlation coefficients (z scores) were combined across participants using the weighted inverse normal method, also called Stouffer’s method [ 38 , 39 ]. Each participant’s z score was weighted by the square root of that participant’s effective sample size, with weights equal to √(n − 3) to reflect the approximate precision of Fisher z. The weighted sum was then divided by the square root of the sum of squared weights to yield a Stouffer Z that is standard normal under the null hypothesis [ 40 ]. This procedure produces a single omnibus p-value that reflects the aggregate evidence for association across the cohort while accounting for differences in window count between sessions. Finally, to test whether the PCA-derived index correlates more strongly with each reference signal than each individual feature, we compared pairs of dependent correlations that share a common variable using the Williams test for overlapping correlations [ 41 , 42 ]. We use the term Williams test to follow standard nomenclature and documentation commonly used in practice, for example, the materials that catalogue and implement the Williams and Steiger procedures in the cocor package in R software [ 43 ]. For each visit, we calculated one-sided p-values to test if the correlation between PCA and references exceeded the correlation between features and references, after accounting for the intercorrelation between PCA and the feature. Because inference was carried out per visit, we combined the resulting one-sided p values across visits using a weighted Stouffer Z with weights √(n − 3), as above, to obtain a single directional group p value for each feature to reference comparison. Figure 3 summarizes the pipeline from data acquisition to RTPI construction and statistical evaluation. RESULTS All 20 participants completed the ischemia–reperfusion protocol without adverse events. As anticipated, NIRS-derived perfusion features responded predictably: both the 200-mmHg complete occlusion and the 100-mmHg partial occlusion elicited immediate, measurable reductions across all modalities. Subsequent cuff release induced a transient reactive hyperemic response during the recovery phase. The composite RTPI, which integrates all three hemodynamic features, outperformed each individual feature. Table 1 presents a statistical validation of NIRS-derived perfusion parameters against PI and flux. Although the features (PAR, AUAC, and DRV) are derived using distinct mathematical approaches, they exhibited broadly similar overall trends. However, closer examination revealed distinct temporal dynamics for each parameter (Fig. 4 ). Specifically, during recovery from partial ischemia, DRV demonstrated the most pronounced overshoot, whereas after complete ischemia, PAR and AUAC reached higher peaks during reperfusion, while DRV increased more gradually and to a lower maximum. Table 1 Group-level correlation coefficients, intraclass-correlation coefficients, and error metrics for all NIRS-derived parameters versus the two reference modalities. Parameter Reference Test-Retest p ICC (95% CI) Pooled r (95% CI) Meta p RMSE PAR PI 0.73 0.04 (-0.43, 0.50) 0.85 (0.84, 0.85) 1.17 × 10⁻¹³ 0.60 PAR Flux 0.55 0.66 (0.28, 0.86) 0.73 (0.73, 0.74) 2.16 × 10⁻⁸ 0.83 AUAC PI 0.66 0.29 (-0.21, 0.66) 0.83 (0.82, 0.83) 6.52 × 10⁻¹² 0.63 AUAC Flux 0.85 0.76 (0.45, 0.91) 0.71 (0.71, 0.72) 2.88 × 10⁻⁸ 0.87 DRV PI 0.56 0.06 (-0.42, 0.51) 0.80 (0.80, 0.81) 1.13 × 10⁻¹⁰ 0.68 DRV Flux 0.81 0.54 (0.10, 0.80) 0.70 (0.69, 0.70) 3.22 × 10⁻⁷ 0.88 PCA PI 0.62 0.10 (-0.39, 0.54) 0.85 (0.84, 0.85) 3.72 × 10⁻¹³ 0.60 PCA Flux 0.94 0.70 (0.34, 0.88) 0.75 (0.75, 0.76) 1.39 × 10⁻⁸ 0.79 PI Flux 0.63 0.51 (0.06, 0.79) 0.68 (0.68, 0.69) 1.20 × 10⁻⁶ 0.93 Notably, RTPI demonstrated enhanced sensitivity for perfusion monitoring, detecting the onset of all ischemia and reperfusion episodes markedly earlier than either the tissue oxygenation index (TOI) or the raw THb signal (Fig. 5 ). Although most datasets demonstrated physiologically consistent responses in both RTPI and the reference measurements (e.g., Fig. 6 ), complete datasets from three participants (across both visits) were excluded from analysis. Two male participants exhibited inverted laser Doppler flux responses, with heartbeat waveforms reversed in polarity and a trend in which flux increased when both PI and RTPI decreased, resulting in very low or negative correlations. These two participants also had the highest body mass indices in the cohort (28 kg/m² and 31 kg/m²), raising the possibility that increased soft tissue thickness positioned the Doppler probe beyond its adequate penetration depth, though this explanation remains unconfirmed. One female participant’s data was excluded due to severe motion artifacts caused by frequent hand movement, which affected all recorded channels. Among the retained recordings, several sessions showed limitations in one or both reference modalities. In two sessions (from different participants), the PI signal decreased only after a marked delay following cuff inflation to 100 mmHg. In two other sessions, laser Doppler flux showed reduced sensitivity: one session exhibited a several-second delay, and in another, a small signal drop occurred during the partial occlusion phase. Paired t-tests on Fisher-transformed correlations showed no significant differences between visits for any parameter-reference combination (all p > 0.55), indicating stable performance across repeated measurements. Moreover, Intraclass correlation coefficients (ICC 3,1) revealed varying degrees of reliability across parameters. When compared against flux, all measures demonstrated good test-retest reliability. However, when validated against PI, most parameters showed poor to fair reliability, with ICC values ranging from 0.04 to 0.29, suggesting greater measurement variability when using PI as the reference standard. Direct comparison between the two reference standards (PI vs. flux) yielded a lower correlation (r = 0.68, 95% CI [0.68, 0.69]) with relatively high error (RMSE = 0.93) compared to NIRS-derived parameters. Moreover, Stouffer's Z meta-analysis confirmed highly significant correlations for all parameter-reference pairs (all p < 0.05). The strongest meta-analytic evidence was observed for the PCA-RTPI and PI (p = 3.72 × 10 –13 ) combination. Additionally, RMSEs were consistently lower when parameters were validated against PI (range: 0.60–0.68) compared to flux (range: 0.79–0.91). Although the raw correlations suggest that PCA generally aligns most strongly with the references, group-level results from the Williams test for dependent correlations reveal a more nuanced pattern of advantages and exceptions, as summarized in Table 2 . Table 2 Group-level comparisons of PCA versus individual features using dependent-correlation tests. Feature Reference \(\:\varDelta\:\varvec{r}\) (PCA minus feature) Group p-value (one-sided) PAR PI -0.02 1 AUAC PI 0.07 1.29 × 10⁻ 262 DRV PI 0.14 0 PAR Flux 0.03 1.57 × 10⁻ 15 AUAC Flux 0.05 8.32 × 10⁻ 35 DRV Flux 0.10 3.55 × 10⁻ 294 When PI was used as the reference, PCA outperformed AUAC and DRV (Δr = 0.07 and 0.14, respectively; p < 0.05), whereas no gain was observed relative to PAR (Δr = − 0.02, p = 1). In contrast, when Flux served as the reference, PCA consistently outperformed all individual features, with gains ranging from Δr = 0.03 to 0.10 and minimal group p values. These findings indicate that the composite PCA-based index provides an advantage over single features in capturing reference-related variance, with the sole exception being the case where PI is compared against PAR. DISCUSSION This study introduces a novel, non-invasive near-infrared spectroscopy framework for quantifying and continuously monitoring regional tissue perfusion. By extracting three physiologically meaningful features, pulse PAR, AUAC, and DRV, from the NIRS-derived THb signal and integrating them using principal component analysis, we derived a composite Regional Tissue Perfusion Index (RTPI). The RTPI demonstrated strong concurrent validity with established reference standards, yielding a correlation of r = 0.85 with the photoplethysmography-based perfusion index and r = 0.75 with laser Doppler flux. These findings support the feasibility of NIRS-based perfusion tracking with clinically relevant fidelity. Furthermore, each individual feature showed statistically significant correlations with both references, reinforcing their standalone value as interpretable perfusion surrogates. Notably, the strong association between PAR and PI is consistent with their mutual reliance on pulsatile-to-baseline amplitude ratios (AC/DC), a well-characterized parameter in optical hemodynamic monitoring [ 44 , 45 ]. Although the three NIRS-derived features exhibited concordant trends during hemodynamic transitions, subtle differences during ischemia and reperfusion phases highlighted their physiological complementarity. PAR and AUAC primarily reflect the reappearance of pulsatile flow, whereas DRV quantifies the instantaneous rate of hemoglobin concentration change, offering temporal resolution independent of cardiac cycles. By integrating these features via PCA, we derived the composite RTPI that slightly outperformed individual metrics in its agreement with both PI and laser Doppler flux, while maintaining a fully reference-free architecture. Beyond incremental performance improvements, PCA confers several methodological advantages. First, it addresses a persistent challenge in pulse peak detection. Situations such as motion artifacts, low-amplitude waveforms during hypoperfusion [ 46 ], clipped peaks from sensor saturation [ 47 ], or waveform variability due to arrhythmia or hemodynamic instability can result in zero or missing peak-based features [ 48 ]. PCA mitigates this issue by redistributing signal variance across components, preserving signal continuity even when individual feature channels degrade. In our experiment, substantial inter-participant differences in pulse amplitude required continuous adjustment of the findpeaks algorithm's parameters, making it challenging to define a single range set that reliably captured true peaks while rejecting noise. This difficulty likely explains the development of advanced peak-detection solutions in prior work (e.g., using an ensemble of algorithms [ 49 ], Masimo SET [ 50 ] and underscores that, while peak-based features (e.g., PAR) can provide higher accuracy, their reliable extraction often demands sophisticated algorithms and intensive processing. Second, PCA has a well-established track record in enhancing NIRS signal fidelity. By emphasizing high-variance components, PCA effectively filters out noise and physiological confounders, often outperforming advanced filtering techniques such as Kalman filters and wavelet-based methods [ 51 , 52 ]. Third, PCA’s efficiency in computation, due to its minimal processing demands, and its unsupervised approach make it particularly suitable for real-time use in clinical environments, where signal characteristics can vary unpredictably and reference modalities are often absent. Finally, the unsupervised nature of PCA also allows it to extract orthogonal components purely from the covariance structure of the NIRS features, eliminating reliance on external reference signals and thus avoiding the propagation of motion artifacts and site-specific variability that frequently degrade laser-Doppler flux measurements [ 53 – 55 ]. These characteristics make PCA particularly well-suited for real-time clinical applications, where signal properties may vary unpredictably across individuals or conditions. The stronger correlation observed between NIRS-derived parameters and PI, relative to laser Doppler flux, likely reflects fundamental differences in the underlying measurement principles. Both NIRS and PPG rely on optical absorption to quantify changes in hemoglobin concentration and thus share a common sensitivity to pulsatile blood volume. In contrast, laser LDF measures a composite signal that includes both the concentration and velocity of moving red blood cells, derived from Doppler frequency shifts. This added velocity component enables a more comprehensive, yet fundamentally distinct, assessment of microvascular perfusion. Notably, because NIRS and PPG lack this velocity dimension, they may not capture certain dynamic aspects of flow that are accessible to LDF-based measurements. The observed variability in measurements, particularly the low intraclass correlation coefficients (ICCs) when validated against PI, suggests systematic differences between visits that likely stem from the intrinsic instability of PPG-based peripheral perfusion assessments, rather than deficiencies in the NIRS methodology itself. In contrast, the higher test–retest reliability seen when validating against laser Doppler flux supports the reproducibility of NIRS-derived perfusion metrics when benchmarked against a more physiologically stable reference standard. The correlation between the two reference standards themselves (r = 0.68) provides crucial context for interpreting our validation results. This inter-reference disagreement inherently limits the achievable correlation for any derived parameter, as no measurement can simultaneously achieve perfect agreement with two references that disagree with each other. This mathematical constraint establishes both a performance ceiling and realistic expectations for NIRS-derived parameters. Remarkably, the NIRS-derived RTPI achieved correlations with individual references that exceeded the correlation between the references themselves. This finding not only validates our approach but also suggests that RTPI captures distinct perfusion information that aligns more closely with each reference modality than the references align with one another. The signal irregularities observed in a subset of sessions underscore known limitations of the two clinical reference standards. Specifically, the failure of LDF to respond appropriately in some cases, including one instance with minimal decrease during partial occlusion, highlights the potential inconsistency of this method under certain physiological or anatomical conditions, even when using clinically approved, high-performance equipment. Likewise, delayed responses in PPG-derived PI suggest susceptibility to algorithmic smoothing or signal processing delays. These findings reinforce the need for more consistent and robust measures of tissue perfusion that are less vulnerable to motion artifacts, tissue heterogeneity, sensor placement variability and delayed response. RTPI demonstrated sensitivity to both ischemic events across all cases and exhibited immediate and significant declines at the onset of both occlusions. Its multiparametric design and real-time responsiveness constitute a progressive advancement toward overcoming these limitations; however, further technical refinements and clinical validation are required. The superior sensitivity of RTPI relative to conventional NIRS parameters was especially evident during partial ischemia. RTPI exhibited an immediate decrease upon cuff inflation, whereas the TOI paradoxically increased before eventually declining. This divergent response highlights the distinct physiological processes associated with venous outflow obstruction. Specifically, when venous drainage is impaired but arterial inflow persists, oxygenated blood accumulates in the capillary bed, transiently elevating local oxygenation levels and producing a short-lived increase in TOI, a phenomenon sometimes referred to as the “vascular trap.” As venous congestion progresses, rising hydrostatic pressure within the microvasculature reduces the arteriovenous pressure gradient, restricting subsequent arterial inflow. Meanwhile, tissue metabolism continues to consume the trapped oxygen, increasing HHb and leading to a delayed decrease in TOI. Similar hemodynamic mechanisms underlie the THb response. Under baseline conditions, the THb signal in a supine, healthy participant remains relatively stable, exhibiting only minor respiratory or vasomotor oscillations. When the cuff is inflated above venous but below arterial pressure, venous outflow is occluded while a reduced arterial inflow continues, leading to blood pooling and a progressive rise in THb. Each incoming pulse further distends the compliant venous reservoir, gradually increasing venous pressure toward the level of the cuff and diminishing the arteriovenous pressure gradient. As this gradient approaches zero, the rate of accumulation slows and THb reaches a plateau, signalling the onset of functional ischemia. Although cuff inflation was limited to three minutes for participant comfort, it is plausible that a longer occlusion period would have revealed a subsequent decline in THb following the peak, analogous to the delayed TOI decrease observed during sustained venous congestion. The immediate drop in RTPI, contrasting sharply with the simultaneous rise in both TOI and THb, demonstrates that perfusion-based parameters provide physiological information beyond conventional oxygenation metrics. This independence of NIRS-derived RTPI from tissue oxygenation values confirms its utility for real-time detection and management of tissue hypoperfusion across diverse clinical scenarios, including trauma, sepsis, and the entire postoperative continuum. While this study successfully validated the NIRS-derived RTPI, several methodological considerations warrant discussion. First, although the sample size provided sufficient statistical power for initial validation, larger and more diverse cohorts are necessary to evaluate generalizability, establish normative thresholds, and assess inter-individual variability across different physiological and pathological states. Second, there were instrumentation-specific limitations affecting both references. LDF exhibited some sensitivity to probe placement, with signal quality in some cases improving only after minor repositioning. In addition, for a few participants, there was a delayed response in reflecting partial ischemia in both references, or in one case, a minimal change during partial ischemia for LDF. Consequently, analysis was restricted to three physiologically meaningful features previously supported in the literature, and feature selection pipelines (e.g., forward or backward selection, ablation tests) were not applied, as the reference itself could not be assumed to provide error-free ground truth for such procedures. Third, although our use of two discrete occlusion pressures (100 and 200 mmHg) effectively demonstrated differential parameter responses, future studies using graded or continuous pressure increments could provide more detailed resolution of the perfusion–pressure relationship. Additional limitations should also be acknowledged. All measurements were conducted at a single anatomical site under resting, supine conditions with minimal autonomic variability; thus, RTPI performance under dynamic cardiovascular states needs to be investigated. Furthermore, the use of a fixed optode configuration and single-wavelength sensor also constrains spatial resolution and tissue-layer specificity; advanced NIRS systems with multi-distance or time-resolved capabilities may enhance signal depth discrimination. While test–retest reliability was assessed, the potential impact of sensor repositioning between sessions was not systematically evaluated. Lastly, although RTPI incorporates artifact-tolerant feature fusion, its robustness against motion remains to be tested under ambulatory conditions, which are critical for real-world wearable applications. Several avenues for future investigation emerge from this work. First, the strong performance of RTPI in healthy participants supports its extension to clinical populations requiring tissue perfusion assessment and monitoring, such as individuals with peripheral arterial disease, diabetes-related microvascular complications, or patients undergoing reconstructive surgery where sustained flap viability monitoring is essential. In particular, the prefrontal cortex represents a promising target for cerebral applications: combining RTPI and TOI, both derived from one prefrontal NIRS sensor, may enable a multimodal approach for real-time monitoring of cortical perfusion and oxygen delivery during surgery, 23 sedation, or critical care. Second, applying dynamic physiological stressors, including exercise, orthostatic maneuvers, and paced respiration, will facilitate assessment of RTPI stability under varying autonomic and cardiovascular conditions. Third, expanding the anatomical scope of RTPI measurement to include the extremities, surgical flaps, and internal organs using implantable NIRS sensors will enhance its clinical versatility [ 56 ]. Finally, implementing graded occlusion protocols with stepwise increases in pressure will enable high-resolution characterization of pressure–perfusion relationships and support the definition of physiologically and clinically relevant ischemic thresholds. This capability may open new avenues for the early diagnosis of acute compartment syndrome conditions, an urgent need that remains unmet in current clinical practice [ 57 , 58 ]. While the computational efficiency of PCA supports real-time RTPI computation, further optimization is needed to refine signal processing algorithms and ensure seamless integration into existing patient monitoring systems. Despite the intrinsic limitations of optical concentration-based techniques, specifically their inability to capture flow velocity, our findings demonstrate that unsupervised dimensionality reduction can extract clinically meaningful perfusion dynamics. Looking ahead, machine learning models trained on laser Doppler flowmetry data may enable the identification of NIRS-derived features that approximate flow velocity, potentially bridging the gap between concentration-based and velocity-sensitive perfusion assessments. However, such approaches must also address key limitations of LDF itself, including shallow tissue penetration and vulnerability to motion artifacts, to ensure robust, generalizable performance across clinical scenarios. CONCLUSION This study introduced and validated the Regional Tissue Perfusion Index (RTPI), a novel NIRS-derived metric for continuous, non-invasive monitoring of regional tissue perfusion. By integrating multiple physiologically meaningful NIRS signal features through principal component analysis, RTPI achieved strong correlations with established reference standards while maintaining computational efficiency suitable for real-time applications. Importantly, RTPI demonstrated superior sensitivity compared with conventional NIRS-derived oxygenation indices, detecting immediate perfusion changes during both complete and partial ischemia, whereas oxygenation metrics exhibited delayed or paradoxical responses. The multiparametric design of RTPI enables a more comprehensive characterization of microvascular dynamics than any single feature alone, establishing it as a robust and physiologically relevant alternative for bedside monitoring. The comparable performance between unsupervised PCA and supervised partial least squares (PLS) methods underscores the robustness and generalizability of this approach. These findings position RTPI as a promising tool for clinical applications across critical care, trauma medicine, perioperative and postoperative monitoring, vascular diagnostics, and rehabilitation. Future work should aim to validate RTPI across diverse patient populations, extend measurements to multiple anatomical sites, and integrate the index into clinical monitoring platforms. Its ability to detect perfusion compromise independently of oxygenation indices highlights its potential value in addressing current diagnostic gaps, particularly in the early recognition of microvascular dysfunction. Finally, combining RTPI with advanced signal processing and machine learning techniques may further enhance its accuracy and scalability, paving the way toward reliable and clinically actionable monitoring of tissue health. Declarations This work was supported by a Translational Research Award from the United States Department of Defence, Spinal Cord Injury Research Program (Grant Number 13704264). COMPETING INTERESTS The authors have no relevant financial or non-financial interests to disclose. ETHICS APPROVAL This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Clinical Research Ethics Board of the University of British Columbia (2024 - #H24-00312-A004) and the U.S. Army Medical Research and Materiel Command (USAMRMC) Office of Human and Animal Research Oversight (OHARO) (2024 - #E04773.1a). CONSENT TO PARTICIPATE Informed consent was obtained from all individual participants included in the study. CONSENT TO PUBLISH The authors affirm that a human research participant provided informed consent for publication of the images in Fig. 1. Author Contribution BS, IT, and AB contributed to the study conception. Experimental design, material preparation, data collection, and analysis were performed by BS, IT, SK, ZA, and APR. The first draft of the manuscript was written by BS and IT, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. Acknowledgement The authors gratefully acknowledge the technical expertise and contributions of Dr. Shadi Momtahen, Dr. Shahbaz Askari, and Jocelyn Begin in the study and design of the RTPI calculating concept. References Phua TJ. Hallmarks of aging: middle-aging hypovascularity, tissue perfusion and nitric oxide perspective on healthspan. Front Aging Front. 2025;5. 10.3389/fragi.2024.1526230 . Guven G, Hilty MP, Ince C, Microcirculation. Physiology, Pathophysiology, and Clinical Application. Blood Purif. 2020;49(1–2):143–50. 10.1159/000503775 . Haseer Koya H, Paul M. Shock. In: StatPearls. [Internet]. Treasure Island (FL): StatPearls Publishing; 2025 [cited 2025 Jul 1]. Available from: http://www.ncbi.nlm.nih.gov/books/NBK531492/ Soares ROS, Losada DM, Jordani MC, Évora P, Castro-e-Silva O. Ischemia/Reperfusion Injury Revisited: An Overview of the Latest Pharmacological Strategies. Int J Mol Sci. 2019;20(20):5034. 10.3390/ijms20205034 . Lawendy A-R, Sanders DW, Bihari A, Parry N, Gray D, Badhwar A. Compartment syndrome–induced microvascular dysfunction: an experimental rodent model. Can J Surg. 2011;54(3):194–200. 10.1503/cjs.048309 . Tsuji S, Ichioka S, Sekiya N, Nakatsuka T. Analysis of ischemia-reperfusion injury in a microcirculatory model of pressure ulcers. Wound Repair Regen. 2005;13(2):209–15. 10.1111/j.1067-1927.2005.130213.x . Wernick B, Nahirniak P, Stawicki SP. Impaired Wound Healing. In: StatPearls. [Internet]. Treasure Island (FL): StatPearls Publishing; 2025 [cited 2025 Jul 1]. Available from: http://www.ncbi.nlm.nih.gov/books/NBK482254/ [Internet]. Multiple Organ Dysfunction Syndrome - Nicholas M. Gourd, Nikitas Nikitas, 2020 [cited 2025 Jul 1]. Available from: https://journals.sagepub.com/doi/abs/10.1177/0885066619871452 van Genderen ME, van Bommel J, Lima A. Monitoring peripheral perfusion in critically ill patients at the bedside. Curr Opin Crit Care. 2012;18(3):273–9. 10.1097/MCC.0b013e3283533924 . Li B, Dai Y, Cai W, Sun M, Sun J. Monitoring of perioperative tissue perfusion and impact on patient outcomes. J Cardiothorac Surg. 2025;20(1):100. 10.1186/s13019-025-03353-6 . Levins TT, Shock. Early Recognition and Management. Journal of Emergency Nursing. Volume 36. Elsevier; 2010. pp. 300–1. 410.1016/j.jen.2010.04.008. Gerken ALH, Keese M, Weiss C, Krücken H-S, Pecher KAP, Ministro A, et al. Investigation of Different Methods of Intraoperative Graft Perfusion Assessment during Kidney Transplantation for the Prediction of Delayed Graft Function: A Prospective Pilot Trial. J Pers Med. 2022;12(10):1749. 10.3390/jpm12101749 . Mazzucco G, Pilón L, Torres-Castro R, Lista-Paz A, López S, Chichizola N, et al. Effects of Cardiovascular Rehabilitation on Myocardial Perfusion and Functional Exercise Capacity in Patients With Stable Coronary Artery Disease and Myocardial Ischemia. J Cardiopulm Rehabil Prev. 2025;45(2):132–8. 10.1097/HCR.0000000000000924 . Mathieu F, Khellaf A, Thelin EP, Zeiler FA. Continuous Thermal Diffusion-Based Cerebral Blood Flow Monitoring in Adult Traumatic Brain Injury: A Scoping Systematic Review. J Neurotrauma. 2019;36(11):1707–23. 10.1089/neu.2018.6309 . Li C, Narayan RK, Wang P, Hartings JA. Regional temperature and quantitative cerebral blood flow responses to cortical spreading depolarization in the rat. J Cereb Blood Flow Metab. Volume 37. SAGE Publications Ltd STM; 2017. pp. 1634–40. 510.1177/0271678X16667131. Guenette JA, Henderson WR, Dominelli PB, Querido JS, Brasher PM, Griesdale DEG, et al. Blood flow index using near-infrared spectroscopy and indocyanine green as a minimally invasive tool to assess respiratory muscle blood flow in humans. Am J Physiol Regul Integr Comp Physiol. 2011;300(4):R984–992. 10.1152/ajpregu.00739.2010 . Habazettl H, Athanasopoulos D, Kuebler WM, Wagner H, Roussos C, Wagner PD, et al. Near-infrared spectroscopy and indocyanine green derived blood flow index for noninvasive measurement of muscle perfusion during exercise. J Appl Physiol (1985). 2010;108(4):962–7. 10.1152/japplphysiol.01269.2009 . Ince C. Hemodynamic coherence and the rationale for monitoring the microcirculation. Crit Care. 2015;19(3):S8. 10.1186/cc14726 . Arnold RC, Dellinger RP, Parrillo JE, Chansky ME, Lotano VE, McCoy JV, et al. Discordance between microcirculatory alterations and arterial pressure in patients with hemodynamic instability. J Crit Care. 2012;27(5):531. 10.1016/j.jcrc.2012.02.007 . Rajan V, Varghese B, van Leeuwen TG, Steenbergen W. Review of methodological developments in laser Doppler flowmetry. Lasers Med Sci. 2009;24(2):269–83. 10.1007/s10103-007-0524-0 . Wårdell K, Richter J, Zsigmond P. Cerebral Microcirculation: Progress and Outlook of Laser Doppler Flowmetry in Neurosurgery and Neurointensive Care. Microcirculation. 2024;31(8):e12884. 10.1111/micc.12884 . Wang Q, Pan M, Kreiss L, Samaei S, Carp SA, Johansson JD, et al. A comprehensive overview of diffuse correlation spectroscopy: Theoretical framework, recent advances in hardware, analysis, and applications. NeuroImage. 2024;298:120793. 10.1016/j.neuroimage.2024.120793 . Shadgan B, Butskiy O. Intraoperative Near-infrared Spectroscopy Can Predict Skin Flap Necrosis. Plast Reconstr Surg Glob Open. 2024;12(9):e6155. 10.1097/GOX.0000000000006155 . Seddone S, Ermini L, Policastro P, Mesin L, Roatta S. Evidence that large vessels do affect near infrared spectroscopy. Sci Rep Nat Publishing Group. 2022;12(1):2155. 10.1038/s41598-022-05863-y . Myers D, McGraw M, George M, Mulier K, Beilman G. Tissue hemoglobin index: a non-invasive optical measure of total tissue hemoglobin. Crit Care. 2009;13(Suppl 5):S2. 10.1186/cc8000 . Seddone S, Ermini L, Policastro P, Mesin L, Roatta S. Evidence that large vessels do affect near infrared spectroscopy. Sci Rep Nat Publishing Group. 2022;12(1):2155. 10.1038/s41598-022-05863-y . GOMEZ M. MONTALVO S, GUROVICH AN. Near Infrared Spectroscopy is not a Surrogate of Venous Occlusion Plethysmography to Assess Microvascular Resting Blood Flow and Function. Int J Exerc Sci [Internet]. 2022 [cited 2025 Mar 14];15(2):1616–26. Available from: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9762160/ Mah A, Anderson D, Askari S, Khosravi S, Butskiy O, Shadgan B. Optical monitoring of transplanted free flaps using an implantable near-infrared spectroscopy sensor. Biophotonics in Exercise Science, Sports Medicine, Health Monitoring Technologies, and Wearables IV. SPIE; 2023. pp. 95–102. Park J, Seok HS, Kim S-S, Shin H. Photoplethysmogram Analysis and Applications: An Integrative Review. Front Physiol Front. 2022;12. 10.3389/fphys.2021.808451 . Coutrot M, Dudoignon E, Joachim J, Gayat E, Vallée F, Dépret F. Perfusion index: Physical principles, physiological meanings and clinical implications in anaesthesia and critical care. Anaesth Crit Care Pain Med. 2021;40(6):100964. 10.1016/j.accpm.2021.100964 . [Internet]. Validation of frontal near-infrared spectroscopy as noninvasive bedside monitoring for regional cerebral blood flow in brain-injured patients - PubMed [cited 2025 Aug 6]. Available from: https://pubmed.ncbi.nlm.nih.gov/22296679/ Bartels K, Thiele RH. Advances in photoplethysmography: beyond arterial oxygen saturation. Can J Anesth/J Can Anesth. 2015;62(12):1313–28. 10.1007/s12630-015-0458-0 . Alvares TS, de Oliveira GV, Soares R, Murias JM. Near-infrared spectroscopy-derived total haemoglobin as an indicator of changes in muscle blood flow during exercise-induced hyperaemia. J Sports Sci. 2020;38(7):751–8. 10.1080/02640414.2020.1733774 . Gupta S, Singh A, Sharma A. Exploiting moving slope features of PPG derivatives for estimation of mean arterial pressure. Biomed Eng Lett. 2023;13(1):1–9. 10.1007/s13534-022-00247-7 . Wall PL, Buising CM, Nelms D, Grulke L, Renner CH. Masimo Perfusion Index Versus Doppler for Tourniquet Effectiveness Monitoring. J Spec Oper Med. 2019;19(1):44–6. DOI: 10.55460/HOAU-RLAW. Fisher RA. Frequency Distribution of the Values of the Correlation Coefficient in Samples from an Indefinitely Large Population., Biometrika. [Oxford University Press, Biometrika Trust]; 1915;10(4):507–21. 10.2307/2331838 Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychol Bull. 1979;86(2):420–8. 10.1037//0033-2909.86.2.420 . Stouffer SA, Suchman EA, Devinney LC, Star SA, Williams RM Jr.. The American soldier: Adjustment during army life. (Studies in social psychology in World War II), Vol. 1. Oxford, England: Princeton Univ. Press; 1949. p. xii, 599. (The American soldier: Adjustment during army life. (Studies in social psychology in World War II), Vol. 1). Whitlock MC. Combining probability from independent tests: the weighted Z-method is superior to Fisher’s approach. j evol Biol. 2005;18(5):1368–73. 10.1111/j.1420-9101.2005.00917.x . Zaykin DV. Optimally weighted Z-test is a powerful method for combining probabilities in meta-analysis. J Evol Biol. 2011;24(8):1836–41. 10.1111/j.1420-9101.2011.02297.x . Williams EJ. The Comparison of Regression Variables. Royal Stat Soc J Ser B: Methodological. 1959;21(2):396–9. 10.1111/j.2517-6161.1959.tb00346.x . Steiger JH. Tests for comparing elements of a correlation matrix. Psychological Bulletin. Volume 87. US: American Psychological Association; 1980. pp. 245–51. 210.1037/0033-2909.87.2.245. Diedenhofen B, Musch J. cocor: A Comprehensive Solution for the Statistical Comparison of Correlations. PLOS ONE. Public Libr Sci. 2015;10(4):e0121945. 10.1371/journal.pone.0121945 . Al-Ali A. Perfusion index smoother. [Internet]. US11006867B2, 2021 May 18 [cited 2025 Aug 4]. Lima AP, Beelen P, Bakker J. Use of a peripheral perfusion index derived from the pulse oximetry signal as a noninvasive indicator of perfusion. Crit Care Med. 2002;30(6):1210–3. 10.1097/00003246-200206000-00006 . Argüello Prada EJ, Serna Maldonado RD. A novel and low-complexity peak detection algorithm for heart rate estimation from low-amplitude photoplethysmographic (PPG) signals. J Med Eng Technol. 2018;42(8):569–77. 10.1080/03091902.2019.1572237 . Supelnic MN, Ferreira AF, Bota PJ, Brás-Rosário L, Plácido da Silva H. Benchmarking of Sensor Configurations and Measurement Sites for Out-of-the-Lab Photoplethysmography. Sens (Basel). 2023;24(1):214. 10.3390/s24010214 . Pereira T, Gadhoumi K, Xiao R, Editorial. Hemodynamic parameters and cardiovascular changes. Front Physiol. 2024;15:1538859. 10.3389/fphys.2024.1538859 . Han D, Bashar SK, Lázaro J, Mohagheghian F, Peitzsch A, Nishita N, et al. A Real-Time PPG Peak Detection Method for Accurate Determination of Heart Rate during Sinus Rhythm and Cardiac Arrhythmia. Biosens (Basel). 2022;12(2):82. 10.3390/bios12020082 . Barker SJ, Wilson WC. Racial effects on Masimo pulse oximetry: a laboratory study. J Clin Monit Comput. 2023;37(2):567–74. 10.1007/s10877-022-00927-w . Zhou X, Sobczak G, McKay CM, Litovsky RY. Comparing fNIRS signal qualities between approaches with and without short channels. PLoS ONE. 2020;15(12):e0244186. 10.1371/journal.pone.0244186 . Cooper R, Selb J, Gagnon L, Phillip D, Schytz HW, Iversen HK, et al. A Systematic Comparison of Motion Artifact Correction Techniques for Functional Near-Infrared Spectroscopy. Front Neurosci Front. 2012;6. 10.3389/fnins.2012.00147 . Aström M. Laser Doppler flowmetry in the assessment of tendon blood flow. Scand J Med Sci Sports. 2000;10(6):365–7. 10.1034/j.1600-0838.2000.010006365.x . Pemp B, Maar N, Weigert G, Luksch A, Resch H, Garhofer G, et al. Strategies for reducing variance in laser Doppler flowmetry measurements. Graefes Arch Clin Exp Ophthalmol. 2009;247(1):67–71. 10.1007/s00417-008-0942-0 . Steinmeier R, Bondar I, Bauhuf C, Fahlbusch R. Laser Doppler flowmetry mapping of cerebrocortical microflow: characteristics and limitations. NeuroImage. 2002;15(1):107–19. 10.1006/nimg.2001.0943 . Cheung A, Tu L, Manouchehri N, Kim K-T, So K, Webster M, et al. Continuous Optical Monitoring of Spinal Cord Oxygenation and Hemodynamics during the First Seven Days Post-Injury in a Porcine Model of Acute Spinal Cord Injury. J Neurotrauma. 2020;37(21):2292–301. 10.1089/neu.2020.7086 . Novak M, Penhaker M, Raska P, Pleva L, Schmidt M. Extremity compartment syndrome: A review with a focus on non-invasive methods of diagnosis. Front Bioeng Biotechnol. 2022;10:801586. 10.3389/fbioe.2022.801586 . PMID: 35923576; PMCID: PMC9340208. Shadgan B, Menon M, O'Brien PJ, Reid WD. Diagnostic techniques in acute compartment syndrome of the leg. J Orthop Trauma. 2008;22(8):581-7. 10.1097/BOT.0b013e318183136d . PMID: 18758292. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 26 Nov, 2025 Reviews received at journal 23 Sep, 2025 Reviewers agreed at journal 04 Sep, 2025 Reviewers agreed at journal 03 Sep, 2025 Reviewers invited by journal 03 Sep, 2025 Editor assigned by journal 21 Aug, 2025 Submission checks completed at journal 21 Aug, 2025 First submitted to journal 18 Aug, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7403210","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":511948060,"identity":"3daf5202-b0dc-422b-ad25-691a74e0edf0","order_by":0,"name":"Babak Shadgan","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA60lEQVRIiWNgGAWjYBACAwYeBsYGBoYEMGKoAJMkaTlDshbGNiK0mLOfPfhxBkNdHn97+sXPhfMO5/E3MD/8gE+LZU9esuQGhsPFEmfeFEvP3AZkHGAzlsDrsAM5BpIPGA4kNtzISZDm3XY4cQPQqfi1nH9j/PMBQ13i/Bs5yb9554C1MP/Aq+VGjhnQYcyJG26kH5PmbQBrYcNri+WMd2mWMwwOJ24884bNmudYerHEYTYzC3xazPlzD9/sqahLnHc8/fFtnhprYNA1P76BTwvUeSCCxwDCYSasHgbYHxCvdhSMglEwCkYUAADx81Ac09Dk+wAAAABJRU5ErkJggg==","orcid":"","institution":"University of British Columbia","correspondingAuthor":true,"prefix":"","firstName":"Babak","middleName":"","lastName":"Shadgan","suffix":""},{"id":511948061,"identity":"59386aee-2196-4d50-a889-63390736d3c6","order_by":1,"name":"Iman Amani Tehrani","email":"","orcid":"","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Iman","middleName":"Amani","lastName":"Tehrani","suffix":""},{"id":511948062,"identity":"49798e85-9c03-4124-931d-f3bea97a4598","order_by":2,"name":"Sadra Khosravi","email":"","orcid":"","institution":"International Collaboration on Repair Discoveries","correspondingAuthor":false,"prefix":"","firstName":"Sadra","middleName":"","lastName":"Khosravi","suffix":""},{"id":511948063,"identity":"fca158e8-5400-4da5-b267-cbdc30e6900a","order_by":3,"name":"Zahra Askari","email":"","orcid":"","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Zahra","middleName":"","lastName":"Askari","suffix":""},{"id":511948064,"identity":"b8631ea1-ff3c-43c1-8930-d69e840e7284","order_by":4,"name":"Amir Parham Pirhadi Rad","email":"","orcid":"","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Amir","middleName":"Parham Pirhadi","lastName":"Rad","suffix":""},{"id":511948065,"identity":"d4ed4863-dab8-47a0-aa48-f6ca1e2ec237","order_by":5,"name":"Ali Bashashati","email":"","orcid":"","institution":"University of British Columbia","correspondingAuthor":false,"prefix":"","firstName":"Ali","middleName":"","lastName":"Bashashati","suffix":""}],"badges":[],"createdAt":"2025-08-19 00:23:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7403210/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7403210/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":90983239,"identity":"e5136fea-a8eb-4e07-b322-842690e1de54","added_by":"auto","created_at":"2025-09-10 09:37:04","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":208389,"visible":true,"origin":"","legend":"\u003cp\u003eMultisensory configuration for tissue perfusion monitoring. \u003cstrong\u003e(A)\u003c/strong\u003e Experimental setup during participant monitoring. \u003cstrong\u003e(B)\u003c/strong\u003e Placement of sensors: the NIRS sensor (1) on the thumb, the LDF sensor (2) on the index finger, and the PPG pulse oximeter sensor (3) on the middle finger. This arrangement enables simultaneous acquisition of NIRS-derived hemodynamic signals, microvascular blood flux, and arterial oxygen saturation from anatomically adjacent sites.\u003c/p\u003e","description":"","filename":"Picture1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7403210/v1/5a8dbe8ca16007581459a731.jpg"},{"id":90982231,"identity":"bc865085-b3db-4bf9-8e06-189f30bac6c3","added_by":"auto","created_at":"2025-09-10 09:29:04","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":169485,"visible":true,"origin":"","legend":"\u003cp\u003eDecomposition of total hemoglobin (THb) signal into AC and DC components. (A) Time-series plot showing the raw THb signal (green line) containing both AC and DC components, and the extracted DC component (blue line) obtained using a second-order Butterworth low-pass filter with a cutoff frequency of 0.5 Hz. The DC component represents the slowly varying baseline of the signal. (B) The isolated AC component (blue line) was extracted using a second-order Butterworth bandpass filter (0.5-5 Hz), with the red shaded area representing the area under the AC curve. Data shown for a 20-second window from a representative measurement.\u003c/p\u003e","description":"","filename":"Picture2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7403210/v1/d42972c405b8d92309e1c05a.jpg"},{"id":90982234,"identity":"073157f5-ff3a-45ab-b0ea-66b7b355c464","added_by":"auto","created_at":"2025-09-10 09:29:04","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":77238,"visible":true,"origin":"","legend":"\u003cp\u003ePipeline from acquisition to analysis for RTPI and reference comparisons.\u003c/p\u003e","description":"","filename":"Picture3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7403210/v1/5eb26add8cf6bf4abfa71611.jpg"},{"id":90982235,"identity":"6481db47-ea9f-4bb5-a5f0-21bf87538177","added_by":"auto","created_at":"2025-09-10 09:29:04","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":158581,"visible":true,"origin":"","legend":"\u003cp\u003eTemporal dynamics of all three features during the entire protocol. Time-series plot showing robust-scaled values of NIRS-derived features: pulse amplitude ratio (PAR, blue), area under the AC curve (AUAC, green), and first derivative (DRV, red) over the full 30-minute experimental protocol.\u003c/p\u003e","description":"","filename":"Picture4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7403210/v1/f448b26ff0b50b94b87b3ab2.jpg"},{"id":90982240,"identity":"46ec8eb7-40ad-4593-8b45-0e87455a6aa0","added_by":"auto","created_at":"2025-09-10 09:29:04","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":171130,"visible":true,"origin":"","legend":"\u003cp\u003eComparative hemodynamic responses of NIRS tissue perfusion index (RTPI-PCA), THb, and TOI during complete and partial ischemia challenges. Time-series plot showing robust-scaled values during a complete experimental protocol. The red shaded region indicates complete ischemia induced by 200 mmHg cuff inflation, producing immediate and synchronized decreases in all three parameters. The blue shaded area represents partial ischemia at 100 mmHg cuff pressure, revealing divergent responses: RTPI (red) demonstrates immediate perfusion compromise, while TOI (black) initially increases due to venous congestion before eventually declining. THb (green) shows pronounced accumulation during partial ischemia, reflecting blood pooling from obstructed venous return.\u003c/p\u003e","description":"","filename":"Picture5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7403210/v1/393a187de3ab8c23a8ce8660.jpg"},{"id":90984687,"identity":"4f480918-6b13-43df-9913-a7d8487fdd20","added_by":"auto","created_at":"2025-09-10 09:45:04","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":199867,"visible":true,"origin":"","legend":"\u003cp\u003eTemporal dynamics of NIRS tissue perfusion index (RTPI-PCA) compared with reference perfusion measurements throughout the ischemia-reperfusion protocol. Time-series plot showing robust-scaled values of RTPI derived from principal component analysis (green), photoplethysmography perfusion index (PI, red), and laser Doppler flux (blue) over the complete 30-minute experimental protocol. All three measures demonstrate roughly synchronized responses during complete ischemia and partial ischemia periods, with concordant perfusion cessation and reactive hyperemia upon reperfusion.\u003c/p\u003e","description":"","filename":"Picture6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7403210/v1/651e3143fa10162d33cb6ed2.jpg"},{"id":90985905,"identity":"cae11387-eb5b-4a07-8fdf-b85e25a18e94","added_by":"auto","created_at":"2025-09-10 10:01:05","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1685255,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7403210/v1/f48718d2-acf1-483f-9258-96cb3f230c45.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Regional Tissue Perfusion Index (RTPI): A New Optical-Based Metric for Quantifying Regional Tissue Perfusion","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eTissue perfusion, the delivery of blood through capillaries to biological tissues, is indispensable for supplying oxygen and nutrients and removing metabolic waste, thereby sustaining cellular metabolism and homeostasis [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Adequate perfusion is a critical determinant of tissue viability and organ function across all organ systems [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Conversely, perfusion deficits underlie a broad spectrum of disorders, including tissue ischemia and hypoxia, circulatory shock [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e], ischemia-reperfusion injury [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e], compartment syndrome [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e], pressure ulcer formation [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e], and impaired wound healing [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Hypoperfusion compromises cellular integrity and can escalate to multiple-organ dysfunction and systemic deterioration [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Accordingly, a reliable assessment of tissue perfusion is a central objective in critical care [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e], postoperative monitoring [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e], emergency medicine [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e], vascular surgery [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e] and rehabilitation [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eDespite its clinical importance, accurate assessment and monitoring of tissue perfusion, especially at the microvascular level, remains a longstanding challenge in modern medicine. Current methods are often invasive, intermittent, or provide only indirect surrogates of perfusion. Techniques such as arterial catheterization, thermodilution, and radiolabeled tracers, although valuable in specific clinical settings, involve procedural risks and are not suitable for localized or continuous monitoring. Thermal-diffusion probes (e.g., Hemedex QFlow\u0026trade;) provide absolute, real-time, quantitative perfusion measurements directly within tissue, an invaluable tool in neuro-ICU care; however, a catheter must be implanted into the brain or soft tissue, limiting its use to operating rooms or ICUs [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. Near-infrared spectroscopy with indocyanine green (NIRS-ICG) applies the Fick principle to yield site-specific blood-flow values; nevertheless, it still requires repeated intravenous dye injections, preventing truly continuous bedside use [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. Collectively, these modalities are precise yet impractical for routine, multi-site, or prolonged monitoring because of their procedural burden.\u003c/p\u003e\u003cp\u003eNon-invasive techniques that avoid vascular access are safer and more widely available; however, each method sacrifices depth, stability, absolute quantification, and accurate regional perfusion measurement. Conventional systemic surrogates, such as standard hemodynamic measurements (e.g., heart rate, blood pressure, and blood oxygen saturation), are widespread; however, they fail to detect spatial or temporal variations in tissue blood flow. Furthermore, they are designed for systemic evaluations rather than localized perfusion monitoring, which limits their utility in detecting subtle regional hemodynamic changes [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. Laser Doppler flowmetry (LDF), a widely used optical technique for assessing microvascular perfusion, has provided valuable insights into skin and superficial tissue blood flow by detecting Doppler shifts caused by the movement of red blood cells. While LDF offers excellent sensitivity to changes in perfusion, it suffers from several important limitations. These include a shallow and variable penetration depth (typically\u0026thinsp;~\u0026thinsp;1\u0026ndash;2 mm), limited spatial resolution, susceptibility to motion artifacts, and difficulty in standardizing absolute perfusion values across subjects or tissue sites. Consequently, LDF is primarily used in research or niche clinical applications rather than routine monitoring [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. The photoplethysmography (PPG)-derived perfusion index (PI), defined as the ratio of the pulsatile (AC) to non-pulsatile (DC) components of the pulse oximeter waveform, is designed to assess tissue perfusion at the pulse oximeter site [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. The PI is easily obtainable at the bedside; however, it shows significant variability between healthy and critically ill individuals. Furthermore, differences in PI between fingers within the same person reduce its reliability as a stand-alone measure of tissue perfusion [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e, \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. Diffuse correlation spectroscopy (DCS) can non-invasively probe centimetre-deep microvascular flow and shows good correlation with invasive cerebral flow probes; yet, current instruments remain bulky and costly, limiting their clinical uptake [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. Hence, even the best non-invasive tools struggle with depth, motion or practicality constraints, underscoring the need for a dye-free, motion-tolerant, continuous index of microvascular perfusion.\u003c/p\u003e\u003cp\u003eThe ability to continuously and non-invasively monitor local tissue perfusion in a practical clinical setting remains an unmet need. Such a capability would allow for earlier detection of tissue damage, help customize hemodynamic treatments, and enable real-time monitoring of therapy effectiveness. It would also broaden perfusion monitoring to include outpatient, post-surgery, and rehabilitative settings, leading to better outcomes for a wide range of patients. This gap in clinical practice highlights the need to develop new physiologic perfusion indices that are accurate, reliable, and suitable for scalable, non-invasive platforms.\u003c/p\u003e\u003cp\u003eNear-Infrared Spectroscopy (NIRS) has emerged as a valuable non-invasive modality for assessing localized tissue oxygenation and hemodynamics. To date, the majority of NIRS applications have focused on parameters related to tissue oxygenation, including oxygenated hemoglobin (O₂Hb), deoxygenated hemoglobin (HHb), hemoglobin difference (Hb-diff), and tissue oxygen saturation (StO\u003csub\u003e2\u003c/sub\u003e) or tissue oxygenation index (TOI). However, using these parameters, particularly TOI, as surrogates for tissue perfusion is not accurate. Although TOI is often interpreted as an indirect marker of perfusion, this assumption is flawed, as clinical scenarios exist in which changes in TOI do not correspond to actual alterations in tissue perfusion [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. Similarly, some studies have proposed total hemoglobin (THb) as a surrogate for perfusion; however, this is also a misconception. THb primarily reflects local blood volume, and its fluctuations indicate blood pooling or volume shifts rather than perfusion dynamics [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. Currently, conventional NIRS methods lack a reliable and specific index for accurate measurement or monitoring of regional tissue perfusion [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eTo address this gap, we have developed a novel physiological metric, the Regional Tissue Perfusion Index (RTPI), derived from NIRS signals and designed to quantify tissue perfusion changes in real time. Unlike oxygenation-based indices, RTPI directly reflects dynamic changes in blood flow, independent of tissue oxygenation indices. Our approach extracts physiologically meaningful features and combines them using unsupervised principal component analysis (PCA). Comprehensive statistical analysis confirms both the clinical relevance and statistical significance of our results.\u003c/p\u003e\u003cp\u003eIn this manuscript, we present the conceptual foundation, methodological framework, and validation results of the NIRS-derived RTPI measure. We evaluated the real-time performance of RTPI against established references, including LDF and PPG, under experimentally controlled ischemia\u0026ndash;reperfusion conditions in healthy participants. Our findings highlight the clinical potential of RTPI as a novel, noise-resistant measure of tissue perfusion and outline future directions for its development and translational application in both research and medical practice. Ongoing and future studies will focus on incorporating advanced feature extraction techniques and more sophisticated modelling approaches to further enhance the accuracy and reliability of perfusion assessment.\u003c/p\u003e"},{"header":"METHODS","content":"\u003cp\u003e\u003cb\u003eParticipants\u0026rsquo; Recruitment\u003c/b\u003e\u003c/p\u003e\u003cp\u003eTwenty healthy adult volunteers (14 males, 6 females) were enrolled in the study. Each participant completed two recording sessions, spaced one week apart, yielding a total of 40 datasets. This inter-session interval was selected to allow for full hemodynamic washout and to minimize potential carryover effects. All study procedures were approved by the institutional research ethics board and conducted in accordance with national guidelines for research involving human subjects. Written informed consent was obtained from all participants prior to data collection.\u003c/p\u003e\u003cp\u003e\u003cb\u003eExperimental Setup\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe complete hardware layout is illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. In this experiment, three fingertips on the right hand were used to monitor three adjacent similar tissues with NIRS, along with LDF and PPG pulse oximeter reference sensors (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). A high-power LDF probe (Moor VMS-LDF1-HP; Moor Instruments, Axminster, United Kingdom) was affixed to the index fingertip, providing a continuous, high-temporal-resolution measurement of capillary blood flux at a 40 Hz sampling rate. To obtain arterial reference parameters, a finger-clip pulse oximeter sensor (MightySat; Masimo Corporation, Irvine, California, USA) was secured to the middle fingertip, delivering the manufacturer\u0026rsquo;s proprietary PI at 0.5 Hz. A custom-built, continuous-wave (CW) NIRS sensor in a spatially resolved configuration, developed to monitor free tissue transfer (FTT) surgical flaps [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e], was positioned on the thumb fingertip to record tissue oxygenation and hemodynamics at a 64 Hz sampling rate.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eExperimental Protocol\u003c/b\u003e\u003c/p\u003e\u003cp\u003e Each recording session was structured to impose two well-controlled ischaemic challenges on the participant\u0026rsquo;s right upper limb, thereby permitting a detailed characterization of the resulting perfusion dynamics. The protocol began with a five-minute baseline measurement. Participants lay on their backs and relaxed during this period, allowing cardiovascular variables to stabilize. Immediately after baseline, a phase of complete ischemia was induced. A digital pneumatic tourniquet (Delfi PTS, Delfi Medical, Vancouver, Canada) encircling the upper arm was inflated to 200 mmHg and maintained at that pressure for two minutes, thereby fully occluding arterial inflow. Once the cuff was released, a ten-minute period of reperfusion followed. This interval captured the expected reactive hyperaemic response and provided a window in which to quantify flow-mediated recovery. The second challenge started with partial ischemia. Here, the tourniquet was reinflated to 100 mmHg, a level sufficient to restrict venous return but not to stop arterial inflow, thereby causing venous congestion. This partial occlusion was sustained for three minutes, after which the cuff was deflated for a further ten-minute reperfusion period. The final reperfusion phase allowed us to observe how perfusion parameters returned toward baseline following a less severe but longer blockage.\u003c/p\u003e\u003cp\u003eThis sequence, comprising baseline, full occlusion, maximal reperfusion, partial occlusion, and a second low-level reperfusion, established distinct hemodynamic states that enabled a comprehensive assessment of microvascular responsiveness.\u003c/p\u003e\u003cp\u003e\u003cb\u003eData Processing and Analysis\u003c/b\u003e\u003c/p\u003e\u003cp\u003eTo derive the RTPI, we began by analyzing the THb signal captured on the near channel of our NIRS sensor (source-detector separation of 10 mm), which records changes in hemoglobin concentration and derived tissue oxygenation indices within a superficial vascular bed that roughly matches the penetration depths of both the LDF and PPG sensors. Aligning penetration depths eliminates confounding differences in sampling volume and establishes a rigorous basis for subsequent comparisons.\u003c/p\u003e\u003cp\u003eFrom the THb waveform, we extracted three physiologically meaningful features, each chosen to emphasize a distinct physiological aspect of microvascular perfusion:\u003c/p\u003e\n\u003ch3\u003e1. Pulse-Amplitude Ratio (PAR)\u003c/h3\u003e\n\u003cp\u003e\u003cdiv class=\"BlockQuote\"\u003e\u003cp\u003ePAR quantifies the relationship between the pulsatile component of the hemodynamic signal and its slower baseline drift, providing a sensitive measure of microvascular perfusion dynamics. The THb signal was decomposed into alternating current (AC) and direct current (DC) components using frequency-domain filtering: the pulsatile AC component (cardiac-cycle oscillations) was isolated using a second-order Butterworth band-pass filter (0.5\u0026ndash;5 Hz), while the slowly-varying DC baseline was extracted using a second-order Butterworth low-pass filter (0.5 Hz cutoff). The Butterworth filter, which is the default in many recent NIRS pipelines, offers a maximally ripple-free response pass-band, thereby preserving both the low-frequency baseline and the high-frequency cardiac morphology of the NIRS waveform while minimizing phase and amplitude distortion when implemented in a zero-phase configuration [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. Figure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA illustrates the decomposition of the THb signal into its constituent components, displaying both the combined signal and the extracted DC component, while Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB presents the isolated AC component.\u003c/p\u003e\u003cp\u003ePeaks and valleys in the AC component were identified using physiologically constrained detection parameters, including a minimum interpeak distance of 0.25 seconds and a maximum distance of 2 seconds (heart rate boundaries), with height, width, and prominence thresholds empirically optimized for reliable feature detection. For each peak, the vertical distance to its immediately subsequent valley was measured and normalized by the DC value interpolated at the temporal midpoint between the peak and valley, yielding an AC/DC (pulse-amplitude) ratio. In PPG, an optical technique that shares the fundamental principles of optical physics with NIRS, the same AC/DC metric (PI) has been validated as a bedside marker of peripheral blood flow and vasomotor tone [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e].\u003c/p\u003e\u003cp\u003ePAR was calculated within non-overlapping 2-second windows to ensure sufficient cardiac cycles for reliable estimation while maintaining temporal resolution. Within each window, all normalized peak-to-valley amplitudes were averaged to produce a single PAR value, with windows containing no detectable peaks assigned zero values to maintain temporal continuity.\u003c/p\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\n\u003ch3\u003e2. Area Under the AC Curve (AUAC)\u003c/h3\u003e\n\u003cp\u003e\u003cdiv class=\"BlockQuote\"\u003e\u003cp\u003eWhile PAR quantifies the relative amplitude of cardiac pulsations, AUAC measures the absolute hemodynamic volume displaced during each cardiac cycle. To ensure physiologically meaningful values and reduce susceptibility to noise and artifacts, AUAC was calculated exclusively for validated cardiac peaks. Within each 2-second analysis window, peaks were identified using the same detection criteria as for PAR. For each confirmed peak, the area under the positive portion of the AC waveform was computed, excluding negative deflections to avoid spurious contributions from signal asymmetry. The trapezoidal rule was applied within the temporal bounds of each peak to yield a robust estimate of cumulative blood volume changes associated with individual cardiac cycles. AUAC metrics are physiologically meaningful, as the area under the NIRS parenchymal pulse (computed relative to an arterial reference) has been employed to quantify cerebral blood volume and hence regional cerebral perfusion in brain-injured patients [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e], whereas the area under the PPG waveform has been shown to correlate with stroke volume and therefore peripheral perfusion in healthy volunteers [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e].\u003c/p\u003e\u003c/div\u003e\u003c/p\u003e\n\u003ch3\u003e3. Derivative (DRV)\u003c/h3\u003e\n\u003cp\u003e\u003cdiv class=\"BlockQuote\"\u003e\u003cp\u003eDRV quantifies the temporal rate of change in perfusion by measuring how rapidly the total hemoglobin (THb) signal varies over time. The THb waveform was segmented into short, non-overlapping windows, and within each window, the first derivative was computed as the point-to-point difference normalized by the sampling interval. These derivative values were then averaged within each 2-second window to reduce high variability and noise while preserving physiologically meaningful trends in perfusion dynamics. First-derivative (slope) metrics are perfusion-relevant because the upslope of the NIRS THb trace shows a strong correlation with Doppler-measured peak muscle blood flow during post-exercise reperfusion [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e], and moving-slope features extracted from the first and second derivatives of the PPG waveform correlate directly with mean arterial pressure, which is an established bedside index of systemic perfusion [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e].\u003c/p\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eTogether, PAR, AUAC, and DRV create a multidimensional representation that comprehensively characterizes tissue perfusion dynamics.\u003c/p\u003e\u003cp\u003eDuring post-processing, the THb, PI, and flux signals were first aligned to their respective software-derived timestamps. To ensure temporal coherence across modalities, all three signals were then cross-aligned using a physiologically distinct landmark, the abrupt onset of reperfusion following complete ischemia. While the temporal offset between flux and THb was minimal (\u0026le;\u0026thinsp;0.5 s in most cases), likely reflecting slight discrepancies in manual acquisition timing, the PI signal consistently lagged behind by at least 16 seconds. PI delay, previously reported in the literature [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e], is presumed to result from the signal-enhancement algorithm applied to the PPG waveform. Each windowed feature vector was standardized using a robust scaling approach, which centers the data by subtracting the median and scales it by the interquartile range. Unlike traditional z-score normalization, this method down-weights the influence of outliers, thereby reducing the impact of motion artifacts and transient sensor dropouts. Importantly, it preserves the relative amplitude relationships among the three features, maintaining the integrity of their physiological contributions to the composite index.\u003c/p\u003e\u003cp\u003eThe scaled triplet entered a principal-component analysis (PCA). Because PCA creates orthogonal axes that capture descending proportions of the total variance, the first principal component (PC1) represents the single direction in feature space that explains the greatest amount of shared fluctuation among PAR, AUAC, and DRV. We treat this PC1 score as an unsupervised composite and refer to it throughout as the RTPI derived by PCA (PCA-RTPI). For each visit, we computed two accuracy metrics: the Pearson correlation coefficient (r) and the root-mean-squared error (RMSE). Correlation coefficients were first converted to their Fisher-transformed values (z\u0026thinsp;=\u0026thinsp;arctanh r), and these z-scores were used for all group-level statistics. The Fisher transform stabilizes the variance of r and produces an approximately normal sampling distribution, which justifies parametric pooling and the construction of symmetric 95% confidence intervals [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eTo quantify session-to-session change, we applied a two-tailed paired-sample Student t-test to the vector of within-participant differences (visit 2 minus visit 1). A Shapiro\u0026ndash;Wilk test confirmed that these difference scores were normal, so no non-parametric substitute was required. Reliability across visits was evaluated using the intraclass correlation coefficient (ICC) model 3,1, a two-way mixed-effects, single-measurement coefficient. This approach is recommended when the same fixed sessions, such as one baseline and one follow-up for each participant, are compared and when the focus is on consistency rather than absolute agreement [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eMoreover, individual Fisher transformed correlation coefficients (z scores) were combined across participants using the weighted inverse normal method, also called Stouffer\u0026rsquo;s method [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. Each participant\u0026rsquo;s z score was weighted by the square root of that participant\u0026rsquo;s effective sample size, with weights equal to \u0026radic;(n\u0026thinsp;\u0026minus;\u0026thinsp;3) to reflect the approximate precision of Fisher z. The weighted sum was then divided by the square root of the sum of squared weights to yield a Stouffer Z that is standard normal under the null hypothesis [\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]. This procedure produces a single omnibus p-value that reflects the aggregate evidence for association across the cohort while accounting for differences in window count between sessions.\u003c/p\u003e\u003cp\u003eFinally, to test whether the PCA-derived index correlates more strongly with each reference signal than each individual feature, we compared pairs of dependent correlations that share a common variable using the Williams test for overlapping correlations [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. We use the term Williams test to follow standard nomenclature and documentation commonly used in practice, for example, the materials that catalogue and implement the Williams and Steiger procedures in the cocor package in R software [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]. For each visit, we calculated one-sided p-values to test if the correlation between PCA and references exceeded the correlation between features and references, after accounting for the intercorrelation between PCA and the feature. Because inference was carried out per visit, we combined the resulting one-sided p values across visits using a weighted Stouffer Z with weights \u0026radic;(n\u0026thinsp;\u0026minus;\u0026thinsp;3), as above, to obtain a single directional group p value for each feature to reference comparison. Figure\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e summarizes the pipeline from data acquisition to RTPI construction and statistical evaluation.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cp\u003eAll 20 participants completed the ischemia\u0026ndash;reperfusion protocol without adverse events. As anticipated, NIRS-derived perfusion features responded predictably: both the 200-mmHg complete occlusion and the 100-mmHg partial occlusion elicited immediate, measurable reductions across all modalities. Subsequent cuff release induced a transient reactive hyperemic response during the recovery phase.\u003c/p\u003e\u003cp\u003eThe composite RTPI, which integrates all three hemodynamic features, outperformed each individual feature. Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e presents a statistical validation of NIRS-derived perfusion parameters against PI and flux. Although the features (PAR, AUAC, and DRV) are derived using distinct mathematical approaches, they exhibited broadly similar overall trends. However, closer examination revealed distinct temporal dynamics for each parameter (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Specifically, during recovery from partial ischemia, DRV demonstrated the most pronounced overshoot, whereas after complete ischemia, PAR and AUAC reached higher peaks during reperfusion, while DRV increased more gradually and to a lower maximum.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eGroup-level correlation coefficients, intraclass-correlation coefficients, and error metrics for all NIRS-derived parameters versus the two reference modalities.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\"\u0026times;\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eParameter\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eReference\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eTest-Retest p\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eICC (95% CI)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003ePooled r (95% CI)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eMeta p\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eRMSE\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePAR\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003ePI\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.73\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.04 (-0.43, 0.50)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.85 (0.84, 0.85)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c6\"\u003e\u003cp\u003e1.17 \u0026times; 10⁻\u0026sup1;\u0026sup3;\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.60\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePAR\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFlux\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.55\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.66 (0.28, 0.86)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.73 (0.73, 0.74)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c6\"\u003e\u003cp\u003e2.16 \u0026times; 10⁻⁸\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.83\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAUAC\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003ePI\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.66\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.29 (-0.21, 0.66)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.83 (0.82, 0.83)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c6\"\u003e\u003cp\u003e6.52 \u0026times; 10⁻\u0026sup1;\u0026sup2;\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.63\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAUAC\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFlux\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.85\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.76 (0.45, 0.91)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.71 (0.71, 0.72)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c6\"\u003e\u003cp\u003e2.88 \u0026times; 10⁻⁸\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.87\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDRV\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003ePI\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.56\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.06 (-0.42, 0.51)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.80 (0.80, 0.81)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c6\"\u003e\u003cp\u003e1.13 \u0026times; 10⁻\u0026sup1;⁰\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.68\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDRV\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFlux\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.81\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.54 (0.10, 0.80)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.70 (0.69, 0.70)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c6\"\u003e\u003cp\u003e3.22 \u0026times; 10⁻⁷\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.88\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePCA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003ePI\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.62\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.10 (-0.39, 0.54)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.85 (0.84, 0.85)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c6\"\u003e\u003cp\u003e3.72 \u0026times; 10⁻\u0026sup1;\u0026sup3;\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.60\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePCA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFlux\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.94\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.70 (0.34, 0.88)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.75 (0.75, 0.76)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c6\"\u003e\u003cp\u003e1.39 \u0026times; 10⁻⁸\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.79\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePI\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFlux\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.63\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.51 (0.06, 0.79)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.68 (0.68, 0.69)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\"\u0026times;\" colname=\"c6\"\u003e\u003cp\u003e1.20 \u0026times; 10⁻⁶\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.93\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eNotably, RTPI demonstrated enhanced sensitivity for perfusion monitoring, detecting the onset of all ischemia and reperfusion episodes markedly earlier than either the tissue oxygenation index (TOI) or the raw THb signal (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eAlthough most datasets demonstrated physiologically consistent responses in both RTPI and the reference measurements (e.g., Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e), complete datasets from three participants (across both visits) were excluded from analysis. Two male participants exhibited inverted laser Doppler flux responses, with heartbeat waveforms reversed in polarity and a trend in which flux increased when both PI and RTPI decreased, resulting in very low or negative correlations. These two participants also had the highest body mass indices in the cohort (28 kg/m\u0026sup2; and 31 kg/m\u0026sup2;), raising the possibility that increased soft tissue thickness positioned the Doppler probe beyond its adequate penetration depth, though this explanation remains unconfirmed. One female participant\u0026rsquo;s data was excluded due to severe motion artifacts caused by frequent hand movement, which affected all recorded channels. Among the retained recordings, several sessions showed limitations in one or both reference modalities. In two sessions (from different participants), the PI signal decreased only after a marked delay following cuff inflation to 100 mmHg. In two other sessions, laser Doppler flux showed reduced sensitivity: one session exhibited a several-second delay, and in another, a small signal drop occurred during the partial occlusion phase.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003ePaired t-tests on Fisher-transformed correlations showed no significant differences between visits for any parameter-reference combination (all p\u0026thinsp;\u0026gt;\u0026thinsp;0.55), indicating stable performance across repeated measurements. Moreover, Intraclass correlation coefficients (ICC 3,1) revealed varying degrees of reliability across parameters. When compared against flux, all measures demonstrated good test-retest reliability. However, when validated against PI, most parameters showed poor to fair reliability, with ICC values ranging from 0.04 to 0.29, suggesting greater measurement variability when using PI as the reference standard.\u003c/p\u003e\u003cp\u003eDirect comparison between the two reference standards (PI vs. flux) yielded a lower correlation (r\u0026thinsp;=\u0026thinsp;0.68, 95% CI [0.68, 0.69]) with relatively high error (RMSE\u0026thinsp;=\u0026thinsp;0.93) compared to NIRS-derived parameters. Moreover, Stouffer's Z meta-analysis confirmed highly significant correlations for all parameter-reference pairs (all p\u0026thinsp;\u0026lt;\u0026thinsp;0.05). The strongest meta-analytic evidence was observed for the PCA-RTPI and PI (p\u0026thinsp;=\u0026thinsp;3.72 \u0026times; 10\u003csup\u003e\u0026ndash;13\u003c/sup\u003e) combination. Additionally, RMSEs were consistently lower when parameters were validated against PI (range: 0.60\u0026ndash;0.68) compared to flux (range: 0.79\u0026ndash;0.91). Although the raw correlations suggest that PCA generally aligns most strongly with the references, group-level results from the Williams test for dependent correlations reveal a more nuanced pattern of advantages and exceptions, as summarized in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eGroup-level comparisons of PCA versus individual features using dependent-correlation tests.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFeature\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eReference\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\varDelta\\:\\varvec{r}\\)\u003c/span\u003e\u003c/span\u003e (PCA minus feature)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eGroup p-value (one-sided)\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePAR\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003ePI\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e-0.02\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAUAC\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003ePI\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.07\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.29 \u0026times; 10⁻\u003csup\u003e262\u003c/sup\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDRV\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003ePI\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePAR\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFlux\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.03\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.57 \u0026times; 10⁻\u003csup\u003e15\u003c/sup\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAUAC\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFlux\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.05\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e8.32 \u0026times; 10⁻\u003csup\u003e35\u003c/sup\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDRV\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eFlux\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e3.55 \u0026times; 10⁻\u003csup\u003e294\u003c/sup\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eWhen PI was used as the reference, PCA outperformed AUAC and DRV (Δr\u0026thinsp;=\u0026thinsp;0.07 and 0.14, respectively; p\u0026thinsp;\u0026lt;\u0026thinsp;0.05), whereas no gain was observed relative to PAR (Δr\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.02, p\u0026thinsp;=\u0026thinsp;1). In contrast, when Flux served as the reference, PCA consistently outperformed all individual features, with gains ranging from Δr\u0026thinsp;=\u0026thinsp;0.03 to 0.10 and minimal group p values. These findings indicate that the composite PCA-based index provides an advantage over single features in capturing reference-related variance, with the sole exception being the case where PI is compared against PAR.\u003c/p\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eThis study introduces a novel, non-invasive near-infrared spectroscopy framework for quantifying and continuously monitoring regional tissue perfusion. By extracting three physiologically meaningful features, pulse PAR, AUAC, and DRV, from the NIRS-derived THb signal and integrating them using principal component analysis, we derived a composite Regional Tissue Perfusion Index (RTPI).\u003c/p\u003e\u003cp\u003eThe RTPI demonstrated strong concurrent validity with established reference standards, yielding a correlation of r\u0026thinsp;=\u0026thinsp;0.85 with the photoplethysmography-based perfusion index and r\u0026thinsp;=\u0026thinsp;0.75 with laser Doppler flux. These findings support the feasibility of NIRS-based perfusion tracking with clinically relevant fidelity. Furthermore, each individual feature showed statistically significant correlations with both references, reinforcing their standalone value as interpretable perfusion surrogates. Notably, the strong association between PAR and PI is consistent with their mutual reliance on pulsatile-to-baseline amplitude ratios (AC/DC), a well-characterized parameter in optical hemodynamic monitoring [\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eAlthough the three NIRS-derived features exhibited concordant trends during hemodynamic transitions, subtle differences during ischemia and reperfusion phases highlighted their physiological complementarity. PAR and AUAC primarily reflect the reappearance of pulsatile flow, whereas DRV quantifies the instantaneous rate of hemoglobin concentration change, offering temporal resolution independent of cardiac cycles. By integrating these features via PCA, we derived the composite RTPI that slightly outperformed individual metrics in its agreement with both PI and laser Doppler flux, while maintaining a fully reference-free architecture.\u003c/p\u003e\u003cp\u003eBeyond incremental performance improvements, PCA confers several methodological advantages. First, it addresses a persistent challenge in pulse peak detection. Situations such as motion artifacts, low-amplitude waveforms during hypoperfusion [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e], clipped peaks from sensor saturation [\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e], or waveform variability due to arrhythmia or hemodynamic instability can result in zero or missing peak-based features [\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e]. PCA mitigates this issue by redistributing signal variance across components, preserving signal continuity even when individual feature channels degrade. In our experiment, substantial inter-participant differences in pulse amplitude required continuous adjustment of the findpeaks algorithm's parameters, making it challenging to define a single range set that reliably captured true peaks while rejecting noise. This difficulty likely explains the development of advanced peak-detection solutions in prior work (e.g., using an ensemble of algorithms [\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e], Masimo SET [\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e] and underscores that, while peak-based features (e.g., PAR) can provide higher accuracy, their reliable extraction often demands sophisticated algorithms and intensive processing. Second, PCA has a well-established track record in enhancing NIRS signal fidelity. By emphasizing high-variance components, PCA effectively filters out noise and physiological confounders, often outperforming advanced filtering techniques such as Kalman filters and wavelet-based methods [\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e]. Third, PCA\u0026rsquo;s efficiency in computation, due to its minimal processing demands, and its unsupervised approach make it particularly suitable for real-time use in clinical environments, where signal characteristics can vary unpredictably and reference modalities are often absent. Finally, the unsupervised nature of PCA also allows it to extract orthogonal components purely from the covariance structure of the NIRS features, eliminating reliance on external reference signals and thus avoiding the propagation of motion artifacts and site-specific variability that frequently degrade laser-Doppler flux measurements [\u003cspan additionalcitationids=\"CR54\" citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e]. These characteristics make PCA particularly well-suited for real-time clinical applications, where signal properties may vary unpredictably across individuals or conditions.\u003c/p\u003e\u003cp\u003eThe stronger correlation observed between NIRS-derived parameters and PI, relative to laser Doppler flux, likely reflects fundamental differences in the underlying measurement principles. Both NIRS and PPG rely on optical absorption to quantify changes in hemoglobin concentration and thus share a common sensitivity to pulsatile blood volume. In contrast, laser LDF measures a composite signal that includes both the concentration and velocity of moving red blood cells, derived from Doppler frequency shifts. This added velocity component enables a more comprehensive, yet fundamentally distinct, assessment of microvascular perfusion. Notably, because NIRS and PPG lack this velocity dimension, they may not capture certain dynamic aspects of flow that are accessible to LDF-based measurements.\u003c/p\u003e\u003cp\u003eThe observed variability in measurements, particularly the low intraclass correlation coefficients (ICCs) when validated against PI, suggests systematic differences between visits that likely stem from the intrinsic instability of PPG-based peripheral perfusion assessments, rather than deficiencies in the NIRS methodology itself. In contrast, the higher test\u0026ndash;retest reliability seen when validating against laser Doppler flux supports the reproducibility of NIRS-derived perfusion metrics when benchmarked against a more physiologically stable reference standard.\u003c/p\u003e\u003cp\u003eThe correlation between the two reference standards themselves (r\u0026thinsp;=\u0026thinsp;0.68) provides crucial context for interpreting our validation results. This inter-reference disagreement inherently limits the achievable correlation for any derived parameter, as no measurement can simultaneously achieve perfect agreement with two references that disagree with each other. This mathematical constraint establishes both a performance ceiling and realistic expectations for NIRS-derived parameters. Remarkably, the NIRS-derived RTPI achieved correlations with individual references that exceeded the correlation between the references themselves. This finding not only validates our approach but also suggests that RTPI captures distinct perfusion information that aligns more closely with each reference modality than the references align with one another.\u003c/p\u003e\u003cp\u003eThe signal irregularities observed in a subset of sessions underscore known limitations of the two clinical reference standards. Specifically, the failure of LDF to respond appropriately in some cases, including one instance with minimal decrease during partial occlusion, highlights the potential inconsistency of this method under certain physiological or anatomical conditions, even when using clinically approved, high-performance equipment. Likewise, delayed responses in PPG-derived PI suggest susceptibility to algorithmic smoothing or signal processing delays. These findings reinforce the need for more consistent and robust measures of tissue perfusion that are less vulnerable to motion artifacts, tissue heterogeneity, sensor placement variability and delayed response. RTPI demonstrated sensitivity to both ischemic events across all cases and exhibited immediate and significant declines at the onset of both occlusions. Its multiparametric design and real-time responsiveness constitute a progressive advancement toward overcoming these limitations; however, further technical refinements and clinical validation are required.\u003c/p\u003e\u003cp\u003eThe superior sensitivity of RTPI relative to conventional NIRS parameters was especially evident during partial ischemia. RTPI exhibited an immediate decrease upon cuff inflation, whereas the TOI paradoxically increased before eventually declining. This divergent response highlights the distinct physiological processes associated with venous outflow obstruction. Specifically, when venous drainage is impaired but arterial inflow persists, oxygenated blood accumulates in the capillary bed, transiently elevating local oxygenation levels and producing a short-lived increase in TOI, a phenomenon sometimes referred to as the \u0026ldquo;vascular trap.\u0026rdquo; As venous congestion progresses, rising hydrostatic pressure within the microvasculature reduces the arteriovenous pressure gradient, restricting subsequent arterial inflow. Meanwhile, tissue metabolism continues to consume the trapped oxygen, increasing HHb and leading to a delayed decrease in TOI.\u003c/p\u003e\u003cp\u003eSimilar hemodynamic mechanisms underlie the THb response. Under baseline conditions, the THb signal in a supine, healthy participant remains relatively stable, exhibiting only minor respiratory or vasomotor oscillations. When the cuff is inflated above venous but below arterial pressure, venous outflow is occluded while a reduced arterial inflow continues, leading to blood pooling and a progressive rise in THb. Each incoming pulse further distends the compliant venous reservoir, gradually increasing venous pressure toward the level of the cuff and diminishing the arteriovenous pressure gradient. As this gradient approaches zero, the rate of accumulation slows and THb reaches a plateau, signalling the onset of functional ischemia. Although cuff inflation was limited to three minutes for participant comfort, it is plausible that a longer occlusion period would have revealed a subsequent decline in THb following the peak, analogous to the delayed TOI decrease observed during sustained venous congestion. The immediate drop in RTPI, contrasting sharply with the simultaneous rise in both TOI and THb, demonstrates that perfusion-based parameters provide physiological information beyond conventional oxygenation metrics. This independence of NIRS-derived RTPI from tissue oxygenation values confirms its utility for real-time detection and management of tissue hypoperfusion across diverse clinical scenarios, including trauma, sepsis, and the entire postoperative continuum.\u003c/p\u003e\u003cp\u003eWhile this study successfully validated the NIRS-derived RTPI, several methodological considerations warrant discussion. First, although the sample size provided sufficient statistical power for initial validation, larger and more diverse cohorts are necessary to evaluate generalizability, establish normative thresholds, and assess inter-individual variability across different physiological and pathological states. Second, there were instrumentation-specific limitations affecting both references. LDF exhibited some sensitivity to probe placement, with signal quality in some cases improving only after minor repositioning. In addition, for a few participants, there was a delayed response in reflecting partial ischemia in both references, or in one case, a minimal change during partial ischemia for LDF. Consequently, analysis was restricted to three physiologically meaningful features previously supported in the literature, and feature selection pipelines (e.g., forward or backward selection, ablation tests) were not applied, as the reference itself could not be assumed to provide error-free ground truth for such procedures. Third, although our use of two discrete occlusion pressures (100 and 200 mmHg) effectively demonstrated differential parameter responses, future studies using graded or continuous pressure increments could provide more detailed resolution of the perfusion\u0026ndash;pressure relationship. Additional limitations should also be acknowledged. All measurements were conducted at a single anatomical site under resting, supine conditions with minimal autonomic variability; thus, RTPI performance under dynamic cardiovascular states needs to be investigated. Furthermore, the use of a fixed optode configuration and single-wavelength sensor also constrains spatial resolution and tissue-layer specificity; advanced NIRS systems with multi-distance or time-resolved capabilities may enhance signal depth discrimination. While test\u0026ndash;retest reliability was assessed, the potential impact of sensor repositioning between sessions was not systematically evaluated. Lastly, although RTPI incorporates artifact-tolerant feature fusion, its robustness against motion remains to be tested under ambulatory conditions, which are critical for real-world wearable applications.\u003c/p\u003e\u003cp\u003eSeveral avenues for future investigation emerge from this work. First, the strong performance of RTPI in healthy participants supports its extension to clinical populations requiring tissue perfusion assessment and monitoring, such as individuals with peripheral arterial disease, diabetes-related microvascular complications, or patients undergoing reconstructive surgery where sustained flap viability monitoring is essential. In particular, the prefrontal cortex represents a promising target for cerebral applications: combining RTPI and TOI, both derived from one prefrontal NIRS sensor, may enable a multimodal approach for real-time monitoring of cortical perfusion and oxygen delivery during surgery,\u003csup\u003e23\u003c/sup\u003e sedation, or critical care. Second, applying dynamic physiological stressors, including exercise, orthostatic maneuvers, and paced respiration, will facilitate assessment of RTPI stability under varying autonomic and cardiovascular conditions. Third, expanding the anatomical scope of RTPI measurement to include the extremities, surgical flaps, and internal organs using implantable NIRS sensors will enhance its clinical versatility [\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e]. Finally, implementing graded occlusion protocols with stepwise increases in pressure will enable high-resolution characterization of pressure\u0026ndash;perfusion relationships and support the definition of physiologically and clinically relevant ischemic thresholds. This capability may open new avenues for the early diagnosis of acute compartment syndrome conditions, an urgent need that remains unmet in current clinical practice [\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e, \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eWhile the computational efficiency of PCA supports real-time RTPI computation, further optimization is needed to refine signal processing algorithms and ensure seamless integration into existing patient monitoring systems. Despite the intrinsic limitations of optical concentration-based techniques, specifically their inability to capture flow velocity, our findings demonstrate that unsupervised dimensionality reduction can extract clinically meaningful perfusion dynamics. Looking ahead, machine learning models trained on laser Doppler flowmetry data may enable the identification of NIRS-derived features that approximate flow velocity, potentially bridging the gap between concentration-based and velocity-sensitive perfusion assessments. However, such approaches must also address key limitations of LDF itself, including shallow tissue penetration and vulnerability to motion artifacts, to ensure robust, generalizable performance across clinical scenarios.\u003c/p\u003e"},{"header":"CONCLUSION","content":"\u003cp\u003eThis study introduced and validated the Regional Tissue Perfusion Index (RTPI), a novel NIRS-derived metric for continuous, non-invasive monitoring of regional tissue perfusion. By integrating multiple physiologically meaningful NIRS signal features through principal component analysis, RTPI achieved strong correlations with established reference standards while maintaining computational efficiency suitable for real-time applications. Importantly, RTPI demonstrated superior sensitivity compared with conventional NIRS-derived oxygenation indices, detecting immediate perfusion changes during both complete and partial ischemia, whereas oxygenation metrics exhibited delayed or paradoxical responses.\u003c/p\u003e\u003cp\u003eThe multiparametric design of RTPI enables a more comprehensive characterization of microvascular dynamics than any single feature alone, establishing it as a robust and physiologically relevant alternative for bedside monitoring. The comparable performance between unsupervised PCA and supervised partial least squares (PLS) methods underscores the robustness and generalizability of this approach. These findings position RTPI as a promising tool for clinical applications across critical care, trauma medicine, perioperative and postoperative monitoring, vascular diagnostics, and rehabilitation.\u003c/p\u003e\u003cp\u003eFuture work should aim to validate RTPI across diverse patient populations, extend measurements to multiple anatomical sites, and integrate the index into clinical monitoring platforms. Its ability to detect perfusion compromise independently of oxygenation indices highlights its potential value in addressing current diagnostic gaps, particularly in the early recognition of microvascular dysfunction. Finally, combining RTPI with advanced signal processing and machine learning techniques may further enhance its accuracy and scalability, paving the way toward reliable and clinically actionable monitoring of tissue health.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eThis work was supported by a Translational Research Award from the United States Department of Defence, Spinal Cord Injury Research Program (Grant Number 13704264).\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eCOMPETING INTERESTS\u003c/strong\u003e\u003cp\u003eThe authors have no relevant financial or non-financial interests to disclose.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003ch2\u003eETHICS APPROVAL\u003c/h2\u003e\u003cp\u003e This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Clinical Research Ethics Board of the University of British Columbia (2024 - #H24-00312-A004) and the U.S. Army Medical Research and Materiel Command (USAMRMC) Office of Human and Animal Research Oversight (OHARO) (2024 - #E04773.1a).\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eCONSENT TO PARTICIPATE\u003c/strong\u003e\u003cp\u003e Informed consent was obtained from all individual participants included in the study.\u003c/p\u003e\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eCONSENT TO PUBLISH\u003c/strong\u003e\u003cp\u003e The authors affirm that a human research participant provided informed consent for publication of the images in Fig.\u0026nbsp;1.\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eBS, IT, and AB contributed to the study conception. Experimental design, material preparation, data collection, and analysis were performed by BS, IT, SK, ZA, and APR. The first draft of the manuscript was written by BS and IT, and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eThe authors gratefully acknowledge the technical expertise and contributions of Dr. Shadi Momtahen, Dr. Shahbaz Askari, and Jocelyn Begin in the study and design of the RTPI calculating concept.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003ePhua TJ. Hallmarks of aging: middle-aging hypovascularity, tissue perfusion and nitric oxide perspective on healthspan. Front Aging Front. 2025;5. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fragi.2024.1526230\u003c/span\u003e\u003cspan address=\"10.3389/fragi.2024.1526230\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGuven G, Hilty MP, Ince C, Microcirculation. Physiology, Pathophysiology, and Clinical Application. Blood Purif. 2020;49(1\u0026ndash;2):143\u0026ndash;50. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1159/000503775\u003c/span\u003e\u003cspan address=\"10.1159/000503775\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHaseer Koya H, Paul M. Shock. In: StatPearls. [Internet]. Treasure Island (FL): StatPearls Publishing; 2025 [cited 2025 Jul 1]. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.ncbi.nlm.nih.gov/books/NBK531492/\u003c/span\u003e\u003cspan address=\"http://www.ncbi.nlm.nih.gov/books/NBK531492/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSoares ROS, Losada DM, Jordani MC, \u0026Eacute;vora P, Castro-e-Silva O. Ischemia/Reperfusion Injury Revisited: An Overview of the Latest Pharmacological Strategies. Int J Mol Sci. 2019;20(20):5034. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/ijms20205034\u003c/span\u003e\u003cspan address=\"10.3390/ijms20205034\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLawendy A-R, Sanders DW, Bihari A, Parry N, Gray D, Badhwar A. Compartment syndrome\u0026ndash;induced microvascular dysfunction: an experimental rodent model. Can J Surg. 2011;54(3):194\u0026ndash;200. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1503/cjs.048309\u003c/span\u003e\u003cspan address=\"10.1503/cjs.048309\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTsuji S, Ichioka S, Sekiya N, Nakatsuka T. Analysis of ischemia-reperfusion injury in a microcirculatory model of pressure ulcers. Wound Repair Regen. 2005;13(2):209\u0026ndash;15. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/j.1067-1927.2005.130213.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1067-1927.2005.130213.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWernick B, Nahirniak P, Stawicki SP. Impaired Wound Healing. In: StatPearls. [Internet]. Treasure Island (FL): StatPearls Publishing; 2025 [cited 2025 Jul 1]. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.ncbi.nlm.nih.gov/books/NBK482254/\u003c/span\u003e\u003cspan address=\"http://www.ncbi.nlm.nih.gov/books/NBK482254/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e[Internet]. Multiple Organ Dysfunction Syndrome - Nicholas M. Gourd, Nikitas Nikitas, 2020 [cited 2025 Jul 1]. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://journals.sagepub.com/doi/abs/10.1177/0885066619871452\u003c/span\u003e\u003cspan address=\"https://journals.sagepub.com/doi/abs/10.1177/0885066619871452\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003evan Genderen ME, van Bommel J, Lima A. Monitoring peripheral perfusion in critically ill patients at the bedside. Curr Opin Crit Care. 2012;18(3):273\u0026ndash;9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/MCC.0b013e3283533924\u003c/span\u003e\u003cspan address=\"10.1097/MCC.0b013e3283533924\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi B, Dai Y, Cai W, Sun M, Sun J. Monitoring of perioperative tissue perfusion and impact on patient outcomes. J Cardiothorac Surg. 2025;20(1):100. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/s13019-025-03353-6\u003c/span\u003e\u003cspan address=\"10.1186/s13019-025-03353-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLevins TT, Shock. Early Recognition and Management. Journal of Emergency Nursing. Volume 36. Elsevier; 2010. pp. 300\u0026ndash;1. 410.1016/j.jen.2010.04.008.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGerken ALH, Keese M, Weiss C, Kr\u0026uuml;cken H-S, Pecher KAP, Ministro A, et al. Investigation of Different Methods of Intraoperative Graft Perfusion Assessment during Kidney Transplantation for the Prediction of Delayed Graft Function: A Prospective Pilot Trial. J Pers Med. 2022;12(10):1749. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/jpm12101749\u003c/span\u003e\u003cspan address=\"10.3390/jpm12101749\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMazzucco G, Pil\u0026oacute;n L, Torres-Castro R, Lista-Paz A, L\u0026oacute;pez S, Chichizola N, et al. Effects of Cardiovascular Rehabilitation on Myocardial Perfusion and Functional Exercise Capacity in Patients With Stable Coronary Artery Disease and Myocardial Ischemia. J Cardiopulm Rehabil Prev. 2025;45(2):132\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/HCR.0000000000000924\u003c/span\u003e\u003cspan address=\"10.1097/HCR.0000000000000924\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMathieu F, Khellaf A, Thelin EP, Zeiler FA. Continuous Thermal Diffusion-Based Cerebral Blood Flow Monitoring in Adult Traumatic Brain Injury: A Scoping Systematic Review. J Neurotrauma. 2019;36(11):1707\u0026ndash;23. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1089/neu.2018.6309\u003c/span\u003e\u003cspan address=\"10.1089/neu.2018.6309\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi C, Narayan RK, Wang P, Hartings JA. Regional temperature and quantitative cerebral blood flow responses to cortical spreading depolarization in the rat. J Cereb Blood Flow Metab. Volume 37. SAGE Publications Ltd STM; 2017. pp. 1634\u0026ndash;40. 510.1177/0271678X16667131.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGuenette JA, Henderson WR, Dominelli PB, Querido JS, Brasher PM, Griesdale DEG, et al. Blood flow index using near-infrared spectroscopy and indocyanine green as a minimally invasive tool to assess respiratory muscle blood flow in humans. Am J Physiol Regul Integr Comp Physiol. 2011;300(4):R984\u0026ndash;992. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1152/ajpregu.00739.2010\u003c/span\u003e\u003cspan address=\"10.1152/ajpregu.00739.2010\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHabazettl H, Athanasopoulos D, Kuebler WM, Wagner H, Roussos C, Wagner PD, et al. Near-infrared spectroscopy and indocyanine green derived blood flow index for noninvasive measurement of muscle perfusion during exercise. J Appl Physiol (1985). 2010;108(4):962\u0026ndash;7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1152/japplphysiol.01269.2009\u003c/span\u003e\u003cspan address=\"10.1152/japplphysiol.01269.2009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eInce C. Hemodynamic coherence and the rationale for monitoring the microcirculation. Crit Care. 2015;19(3):S8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/cc14726\u003c/span\u003e\u003cspan address=\"10.1186/cc14726\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eArnold RC, Dellinger RP, Parrillo JE, Chansky ME, Lotano VE, McCoy JV, et al. Discordance between microcirculatory alterations and arterial pressure in patients with hemodynamic instability. J Crit Care. 2012;27(5):531. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.jcrc.2012.02.007\u003c/span\u003e\u003cspan address=\"10.1016/j.jcrc.2012.02.007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRajan V, Varghese B, van Leeuwen TG, Steenbergen W. Review of methodological developments in laser Doppler flowmetry. Lasers Med Sci. 2009;24(2):269\u0026ndash;83. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s10103-007-0524-0\u003c/span\u003e\u003cspan address=\"10.1007/s10103-007-0524-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eW\u0026aring;rdell K, Richter J, Zsigmond P. Cerebral Microcirculation: Progress and Outlook of Laser Doppler Flowmetry in Neurosurgery and Neurointensive Care. Microcirculation. 2024;31(8):e12884. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/micc.12884\u003c/span\u003e\u003cspan address=\"10.1111/micc.12884\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang Q, Pan M, Kreiss L, Samaei S, Carp SA, Johansson JD, et al. A comprehensive overview of diffuse correlation spectroscopy: Theoretical framework, recent advances in hardware, analysis, and applications. NeuroImage. 2024;298:120793. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.neuroimage.2024.120793\u003c/span\u003e\u003cspan address=\"10.1016/j.neuroimage.2024.120793\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eShadgan B, Butskiy O. Intraoperative Near-infrared Spectroscopy Can Predict Skin Flap Necrosis. Plast Reconstr Surg Glob Open. 2024;12(9):e6155. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/GOX.0000000000006155\u003c/span\u003e\u003cspan address=\"10.1097/GOX.0000000000006155\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSeddone S, Ermini L, Policastro P, Mesin L, Roatta S. Evidence that large vessels do affect near infrared spectroscopy. Sci Rep Nat Publishing Group. 2022;12(1):2155. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41598-022-05863-y\u003c/span\u003e\u003cspan address=\"10.1038/s41598-022-05863-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMyers D, McGraw M, George M, Mulier K, Beilman G. Tissue hemoglobin index: a non-invasive optical measure of total tissue hemoglobin. Crit Care. 2009;13(Suppl 5):S2. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1186/cc8000\u003c/span\u003e\u003cspan address=\"10.1186/cc8000\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSeddone S, Ermini L, Policastro P, Mesin L, Roatta S. Evidence that large vessels do affect near infrared spectroscopy. Sci Rep Nat Publishing Group. 2022;12(1):2155. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41598-022-05863-y\u003c/span\u003e\u003cspan address=\"10.1038/s41598-022-05863-y\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGOMEZ M. MONTALVO S, GUROVICH AN. Near Infrared Spectroscopy is not a Surrogate of Venous Occlusion Plethysmography to Assess Microvascular Resting Blood Flow and Function. Int J Exerc Sci [Internet]. 2022 [cited 2025 Mar 14];15(2):1616\u0026ndash;26. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ncbi.nlm.nih.gov/pmc/articles/PMC9762160/\u003c/span\u003e\u003cspan address=\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9762160/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMah A, Anderson D, Askari S, Khosravi S, Butskiy O, Shadgan B. Optical monitoring of transplanted free flaps using an implantable near-infrared spectroscopy sensor. Biophotonics in Exercise Science, Sports Medicine, Health Monitoring Technologies, and Wearables IV. SPIE; 2023. pp. 95\u0026ndash;102.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePark J, Seok HS, Kim S-S, Shin H. Photoplethysmogram Analysis and Applications: An Integrative Review. Front Physiol Front. 2022;12. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fphys.2021.808451\u003c/span\u003e\u003cspan address=\"10.3389/fphys.2021.808451\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCoutrot M, Dudoignon E, Joachim J, Gayat E, Vall\u0026eacute;e F, D\u0026eacute;pret F. Perfusion index: Physical principles, physiological meanings and clinical implications in anaesthesia and critical care. Anaesth Crit Care Pain Med. 2021;40(6):100964. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1016/j.accpm.2021.100964\u003c/span\u003e\u003cspan address=\"10.1016/j.accpm.2021.100964\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003e[Internet]. Validation of frontal near-infrared spectroscopy as noninvasive bedside monitoring for regional cerebral blood flow in brain-injured patients - PubMed [cited 2025 Aug 6]. Available from: \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://pubmed.ncbi.nlm.nih.gov/22296679/\u003c/span\u003e\u003cspan address=\"https://pubmed.ncbi.nlm.nih.gov/22296679/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBartels K, Thiele RH. Advances in photoplethysmography: beyond arterial oxygen saturation. Can J Anesth/J Can Anesth. 2015;62(12):1313\u0026ndash;28. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s12630-015-0458-0\u003c/span\u003e\u003cspan address=\"10.1007/s12630-015-0458-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAlvares TS, de Oliveira GV, Soares R, Murias JM. Near-infrared spectroscopy-derived total haemoglobin as an indicator of changes in muscle blood flow during exercise-induced hyperaemia. J Sports Sci. 2020;38(7):751\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1080/02640414.2020.1733774\u003c/span\u003e\u003cspan address=\"10.1080/02640414.2020.1733774\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGupta S, Singh A, Sharma A. Exploiting moving slope features of PPG derivatives for estimation of mean arterial pressure. Biomed Eng Lett. 2023;13(1):1\u0026ndash;9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s13534-022-00247-7\u003c/span\u003e\u003cspan address=\"10.1007/s13534-022-00247-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWall PL, Buising CM, Nelms D, Grulke L, Renner CH. Masimo Perfusion Index Versus Doppler for Tourniquet Effectiveness Monitoring. J Spec Oper Med. 2019;19(1):44\u0026ndash;6. DOI: 10.55460/HOAU-RLAW.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFisher RA. Frequency Distribution of the Values of the Correlation Coefficient in Samples from an Indefinitely Large Population., Biometrika. [Oxford University Press, Biometrika Trust]; 1915;10(4):507\u0026ndash;21. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.2307/2331838\u003c/span\u003e\u003cspan address=\"10.2307/2331838\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eShrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychol Bull. 1979;86(2):420\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1037//0033-2909.86.2.420\u003c/span\u003e\u003cspan address=\"10.1037//0033-2909.86.2.420\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eStouffer SA, Suchman EA, Devinney LC, Star SA, Williams RM Jr.. The American soldier: Adjustment during army life. (Studies in social psychology in World War II), Vol. 1. Oxford, England: Princeton Univ. Press; 1949. p. xii, 599. (The American soldier: Adjustment during army life. (Studies in social psychology in World War II), Vol. 1).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWhitlock MC. Combining probability from independent tests: the weighted Z-method is superior to Fisher\u0026rsquo;s approach. j evol Biol. 2005;18(5):1368\u0026ndash;73. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/j.1420-9101.2005.00917.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1420-9101.2005.00917.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZaykin DV. Optimally weighted Z-test is a powerful method for combining probabilities in meta-analysis. J Evol Biol. 2011;24(8):1836\u0026ndash;41. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/j.1420-9101.2011.02297.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1420-9101.2011.02297.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWilliams EJ. The Comparison of Regression Variables. Royal Stat Soc J Ser B: Methodological. 1959;21(2):396\u0026ndash;9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/j.2517-6161.1959.tb00346.x\u003c/span\u003e\u003cspan address=\"10.1111/j.2517-6161.1959.tb00346.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSteiger JH. Tests for comparing elements of a correlation matrix. Psychological Bulletin. Volume 87. US: American Psychological Association; 1980. pp. 245\u0026ndash;51. 210.1037/0033-2909.87.2.245.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDiedenhofen B, Musch J. cocor: A Comprehensive Solution for the Statistical Comparison of Correlations. PLOS ONE. Public Libr Sci. 2015;10(4):e0121945. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1371/journal.pone.0121945\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0121945\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAl-Ali A. Perfusion index smoother. [Internet]. US11006867B2, 2021 May 18 [cited 2025 Aug 4].\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLima AP, Beelen P, Bakker J. Use of a peripheral perfusion index derived from the pulse oximetry signal as a noninvasive indicator of perfusion. Crit Care Med. 2002;30(6):1210\u0026ndash;3. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/00003246-200206000-00006\u003c/span\u003e\u003cspan address=\"10.1097/00003246-200206000-00006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eArg\u0026uuml;ello Prada EJ, Serna Maldonado RD. A novel and low-complexity peak detection algorithm for heart rate estimation from low-amplitude photoplethysmographic (PPG) signals. J Med Eng Technol. 2018;42(8):569\u0026ndash;77. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1080/03091902.2019.1572237\u003c/span\u003e\u003cspan address=\"10.1080/03091902.2019.1572237\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSupelnic MN, Ferreira AF, Bota PJ, Br\u0026aacute;s-Ros\u0026aacute;rio L, Pl\u0026aacute;cido da Silva H. Benchmarking of Sensor Configurations and Measurement Sites for Out-of-the-Lab Photoplethysmography. Sens (Basel). 2023;24(1):214. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/s24010214\u003c/span\u003e\u003cspan address=\"10.3390/s24010214\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePereira T, Gadhoumi K, Xiao R, Editorial. Hemodynamic parameters and cardiovascular changes. Front Physiol. 2024;15:1538859. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fphys.2024.1538859\u003c/span\u003e\u003cspan address=\"10.3389/fphys.2024.1538859\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHan D, Bashar SK, L\u0026aacute;zaro J, Mohagheghian F, Peitzsch A, Nishita N, et al. A Real-Time PPG Peak Detection Method for Accurate Determination of Heart Rate during Sinus Rhythm and Cardiac Arrhythmia. Biosens (Basel). 2022;12(2):82. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3390/bios12020082\u003c/span\u003e\u003cspan address=\"10.3390/bios12020082\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBarker SJ, Wilson WC. Racial effects on Masimo pulse oximetry: a laboratory study. J Clin Monit Comput. 2023;37(2):567\u0026ndash;74. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s10877-022-00927-w\u003c/span\u003e\u003cspan address=\"10.1007/s10877-022-00927-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhou X, Sobczak G, McKay CM, Litovsky RY. Comparing fNIRS signal qualities between approaches with and without short channels. PLoS ONE. 2020;15(12):e0244186. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1371/journal.pone.0244186\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0244186\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCooper R, Selb J, Gagnon L, Phillip D, Schytz HW, Iversen HK, et al. A Systematic Comparison of Motion Artifact Correction Techniques for Functional Near-Infrared Spectroscopy. Front Neurosci Front. 2012;6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fnins.2012.00147\u003c/span\u003e\u003cspan address=\"10.3389/fnins.2012.00147\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAstr\u0026ouml;m M. Laser Doppler flowmetry in the assessment of tendon blood flow. Scand J Med Sci Sports. 2000;10(6):365\u0026ndash;7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1034/j.1600-0838.2000.010006365.x\u003c/span\u003e\u003cspan address=\"10.1034/j.1600-0838.2000.010006365.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePemp B, Maar N, Weigert G, Luksch A, Resch H, Garhofer G, et al. Strategies for reducing variance in laser Doppler flowmetry measurements. Graefes Arch Clin Exp Ophthalmol. 2009;247(1):67\u0026ndash;71. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s00417-008-0942-0\u003c/span\u003e\u003cspan address=\"10.1007/s00417-008-0942-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSteinmeier R, Bondar I, Bauhuf C, Fahlbusch R. Laser Doppler flowmetry mapping of cerebrocortical microflow: characteristics and limitations. NeuroImage. 2002;15(1):107\u0026ndash;19. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1006/nimg.2001.0943\u003c/span\u003e\u003cspan address=\"10.1006/nimg.2001.0943\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCheung A, Tu L, Manouchehri N, Kim K-T, So K, Webster M, et al. Continuous Optical Monitoring of Spinal Cord Oxygenation and Hemodynamics during the First Seven Days Post-Injury in a Porcine Model of Acute Spinal Cord Injury. J Neurotrauma. 2020;37(21):2292\u0026ndash;301. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1089/neu.2020.7086\u003c/span\u003e\u003cspan address=\"10.1089/neu.2020.7086\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eNovak M, Penhaker M, Raska P, Pleva L, Schmidt M. Extremity compartment syndrome: A review with a focus on non-invasive methods of diagnosis. Front Bioeng Biotechnol. 2022;10:801586. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.3389/fbioe.2022.801586\u003c/span\u003e\u003cspan address=\"10.3389/fbioe.2022.801586\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 35923576; PMCID: PMC9340208.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eShadgan B, Menon M, O'Brien PJ, Reid WD. Diagnostic techniques in acute compartment syndrome of the leg. J Orthop Trauma. 2008;22(8):581-7. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1097/BOT.0b013e318183136d\u003c/span\u003e\u003cspan address=\"10.1097/BOT.0b013e318183136d\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. PMID: 18758292.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"journal-of-clinical-monitoring-and-computing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [Journal of Clinical Monitoring and Computing](https://www.springer.com/journal/10877)","snPcode":"10877","submissionUrl":"https://submission.nature.com/new-submission/10877/3","title":"Journal of Clinical Monitoring and Computing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Blood Flow, Microvascular Monitoring, Near-Infrared Spectroscopy, Principal Component Analysis, Tissue Perfusion","lastPublishedDoi":"10.21203/rs.3.rs-7403210/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7403210/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003ePurpose\u003c/h2\u003e\u003cp\u003eAccurate, continuous assessment of regional tissue perfusion remains a significant clinical challenge, as most existing modalities are invasive, indirect, or impractical for routine monitoring. Near-infrared spectroscopy (NIRS) has been widely adopted to assess tissue oxygenation; however, conventional NIRS-derived indices are insufficient surrogates for true perfusion and often fail to capture rapid hemodynamic changes. This study aimed to introduce and validate the Regional Tissue Perfusion Index (RTPI), a novel NIRS-derived metric that integrates multiple features of the NIRS signal to provide continuous, non-invasive, and physiologically relevant assessment of tissue perfusion.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e\u003cp\u003eRTPI was developed using principal component analysis (PCA) of multiple NIRS-derived parameters, including pulse amplitude ratio, signal derivatives, and area under the curve. Its performance was evaluated in healthy volunteers during controlled ischemia\u0026ndash;reperfusion protocols and compared with established reference standards, including laser Doppler flowmetry (LDF) and photoplethysmography (PPG). Partial least squares (PLS) regression was also applied to test the robustness of the approach.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eRTPI showed strong correlations with LDF and PPG during dynamic perfusion changes. Unlike conventional NIRS-derived oxygenation and hemodynamic indices, which often exhibited delayed or paradoxical responses, RTPI demonstrated immediate and significant sensitivity to both complete and partial ischemia\u0026ndash;reperfusion episodes across all cases. Intraclass correlation and error analyses confirmed high test\u0026ndash;retest reliability and low measurement error. Comparable performance between PCA- and PLS-derived indices further supported robustness and generalizability.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e\u003cp\u003eRTPI represents a multiparametric, physiologically meaningful, and computationally efficient metric for real-time tissue perfusion monitoring. Its ability to detect perfusion compromise independently of oxygenation indices highlights its translational potential for bedside implementation in critical care, trauma, perioperative, and vascular medicine, where improved diagnostic accuracy could significantly impact patient outcomes.\u003c/p\u003e","manuscriptTitle":"Regional Tissue Perfusion Index (RTPI): A New Optical-Based Metric for Quantifying Regional Tissue Perfusion","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-10 09:28:59","doi":"10.21203/rs.3.rs-7403210/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-11-26T12:39:52+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-23T21:53:32+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"119208955318106277349901825099311639518","date":"2025-09-04T16:18:44+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"33875591995749622510856340180941961105","date":"2025-09-03T11:47:41+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-09-03T07:59:17+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-21T08:17:10+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-08-21T08:13:57+00:00","index":"","fulltext":""},{"type":"submitted","content":"Journal of Clinical Monitoring and Computing","date":"2025-08-19T00:10:30+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"journal-of-clinical-monitoring-and-computing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"Learn more about [Journal of Clinical Monitoring and Computing](https://www.springer.com/journal/10877)","snPcode":"10877","submissionUrl":"https://submission.nature.com/new-submission/10877/3","title":"Journal of Clinical Monitoring and Computing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"9a2143e3-ab18-4275-9cb6-d14a8d672efd","owner":[],"postedDate":"September 10th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-02-17T16:24:26+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-10 09:28:59","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7403210","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7403210","identity":"rs-7403210","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}