Non-Invasive Prediction of Blood Lactate During Incremental Exercise via Heart Rate, Core Body Temperature, and Sweat-Derived Indices

Non-Invasive Prediction of Blood Lactate During Incremental Exercise via Heart Rate, Core Body Temperature, and Sweat-Derived Indices | 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 Article Non-Invasive Prediction of Blood Lactate During Incremental Exercise via Heart Rate, Core Body Temperature, and Sweat-Derived Indices Jaesung Lee, Jihye Moon, Youngim Kim, Hyeonmin Kim, Eunbi Kim, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8498078/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 07 Apr, 2026 Read the published version in Scientific Reports → Version 1 posted 10 You are reading this latest preprint version Abstract Blood lactate concentration (BLa) is a key marker of metabolic stress, but invasive sampling limits real-time monitoring. We developed a non-invasive model to estimate BLa during incremental exercise using heart rate (HR), core body temperature (CBT), and sweat-derived indices. Thirty-one healthy adult males performed a graded treadmill test. HR and CBT were monitored continuously. Sweat was sampled from the forehead, chest, and back to quantify sweat lactate concentration ([La−]sw) and lactate excretion rate (LER = [La−]sw × sweat rate). Linear mixed-effects models (LMMs) were fitted with log-transformed BLa (Log[BLa]) and participant-level random effects. BLa increased with exercise intensity (p < 0.001), accompanied by increases in HR, CBT and LER (both p < 0.001). LMMs combining HR, CBT, and sweat indices showed strong performance for Log[BLa]. The best model (HR + CBT + forehead LER) achieved conditional R²=0.939 and RMSE = 0.229 (log units), and forehead-based models outperformed chest and back. Combined cardiovascular, thermoregulatory, and sweat-derived measures enable accurate, non-invasive estimation of BLa during graded exercise, supporting wearable-based metabolic monitoring and individualized exercise prescription. Health sciences/Biomarkers Health sciences/Cardiology Health sciences/Health care Health sciences/Medical research Biological sciences/Physiology Blood lactate heart rate core body temperature sweat lactate non-invasive monitoring wearable biosensor Figures Figure 1 Figure 2 Figure 3 Figure 4 Introduction Blood lactate concentration (BLa) increases with exercise intensity and serves as a key indicator of aerobic capacity and endurance performance; accordingly, it is widely used in sports science and clinical rehabilitation[1,2]. However, conventional BLa assessment during exercise requires invasive blood sampling, which limits continuous and real-time monitoring due to discomfort, procedural complexity, and potential infection risk[3]. Recent advances in wearable devices for sweat lactate measurement have enabled non-invasive approaches to estimating BLa[4-7]. Importantly, sweat lactate concentration ([La⁻]ₛw) does not reflect passive diffusion from blood. It can also be locally produced through glycolytic activity within sweat gland cells, and is influenced by multiple physiological and environmental factors, including sweat rate, skin blood flow, and thermal stress[8,9]. As sweat rate increases, dilution effects may attenuate the association between [La⁻]ₛw and BLa[9]. To partially account for dilution, lactate excretion rate (LER; [La⁻]ₛw × sweat rate) has been proposed as a more informative index[10]. Nevertheless, regional differences in sweat gland density and sweating responses, as well as variation in sensor attachment and skin–sensor contact, may limit the representativeness of a single local measurement[11]. This suggests that additional physiological variables may be necessary to improve the robustness of non-invasive BLa estimation. Heart rate (HR) and core body temperature (CBT) are potential physiological signals that may reflect metabolic strain beyond cardiovascular or sweat-derived indices[12,13]. HR typically increases with exercise intensity and shows moderate-to-strong associations with BLa, supporting its use for lactate estimation[12]. However, inter-individual variability in cardiovascular responses, training status, and fatigue resistance can reduce predictive reliability[13]. CBT also rises progressively during exercise due to metabolic heat accumulation, and elevations in body temperature have been linked to increased glycolytic activation and lactate accumulation[14,15]. Despite this physiological rationale, CBT has rarely been incorporated as a predictor of lactate dynamics, as prior work has primarily focused on thermoregulatory function or heat-related illness risk[16]. Accordingly, the aim of this study was to develop and validate multivariable models to predict BLa during incremental exercise using only non-invasive physiological signals. We constructed regression models incorporating HR, CBT, and sweat-derived lactate indices. Based on the optimal combination of explanatory variables identified using a linear mixed-effects model, we additionally applied a random forest algorithm using the same inputs to evaluate predictive performance. Results Stage-dependent Responses of Physiological Variables . As shown in Figure 1, blood lactate concentration (BLa) increased progressively with increasing exercise intensity, showing marked elevation from stage 4 and peaking at stage 7 (p < 0.001; Figure 1A). Heart rate (HR) and core body temperature (CBT) significantly increased with workload (p < 0.001; Figure 1B), with HR increasing steadily from ~100 bpm to ~180 bpm and plateauing near stage 7, and CBT rising sharply between stages 6 and 8 to ~39 °C. Lactate excretion rate (LER) at the forehead, chest, and back significantly increased with exercise intensity (p < 0.001; Figure 1D), with the forehead showing the largest rise before stabilizing at higher stages. In contrast, sweat lactate concentration ([La⁻]ₛw) significantly decreased across stages (p < 0.001; Figure 1C), peaking at stage 0 and a progressive decline that stabilized from stage 5 onward. Correlations Between BLa and Non-Invasive Physiological Variables. Figure 2 shows the correlations between BLa and non-invasive physiological variables. HR strongly correlated with BLa (r = 0.814, p < 0.001; Figure 2A), and CBT showed a moderate positive correlation (r = 0.563, p < 0.001; Figure 2B). [La⁻]ₛw exhibited weak negative correlations across measurement sites—forehead (r = –0.242, p = 0.002; Figure 2C) and back (r = –0.250, p < 0.001; Figure 2E). LER showed significant positive correlations with BLa—forehead (r = 0.391, p < 0.001; Figure 2F), chest (r = 0.254, p < 0.001; Figure 2G), and back (r = 0.224, p < 0.001; Figure 2H). Regression Models for Predicting Blood Lactate Concentration. Table 1 summarizes the regression performance for predicting BLa using various physiological signal combinations. Conditional and marginal R² values and corresponding RMSE were used to evaluate each model’s explanatory power and prediction accuracy. Models that included HR demonstrated the highest predictive capability, with conditional R² values consistently > 0.90. Among these, combining HR, CBT, and forehead LER yielded the best performance (conditional R² = 0.939, marginal R² = 0.788, conditional RMSE = 0.2288). Models using CBT alone showed moderate predictive accuracy (conditional R² = 0.799, marginal R² = 0.481). Predictive performance improved when CBT was combined with sweat-derived parameters such as LER or [La⁻]ₛw. The CBT + forehead LER model achieved a conditional R² of 0.901 and a conditional RMSE of 0.3662. Overall, regression models integrating cardiovascular (HR) and thermoregulatory (CBT) variables with sweat-derived parameters (LER or [La⁻]ₛw) exhibited conditional R² values ranging from 0.921 to 0.939. Table 1. Regression Performance (R², RMSE) by physiological signal Combination Factor Conditional R² Marginal R² Conditional RMSE Marginal RMSE HR 0.908*** 0.795*** 0.2973 0.4678 HR+CBT 0.917*** 0.807*** 0.2818 0.4554 HR+Forehead[La-]sw 0.926*** 0.807*** 0.2648 0.4618 HR+Chest[La-]sw 0.918*** 0.808*** 0.2795 0.4510 HR+Back[La-]sw 0.916*** 0.791*** 0.2839 0.4681 HR+Forehead LER 0.921*** 0.785*** 0.2596 0.4561 HR+Chest LER 0.915*** 0.797*** 0.2779 0.4556 HR+Back LER 0.912*** 0.789*** 0.2829 0.4614 HR+CBT+Forehead[La-]sw 0.933*** 0.818*** 0.2429 0.4248 HR+CBT+Chest[La-]sw 0.921*** 0.828*** 0.2698 0.4231 HR+CBT+Back[La-]sw 0.926*** 0.799*** 0.2675 0.4634 HR+CBT+Forehead LER 0.939*** 0.788*** 0.2288 0.4572 HR+CBT+Chest LER 0.921*** 0.806*** 0.2684 0.4502 HR+CBT+Back LER 0.933*** 0.792*** 0.2668 0.4671 CBT 0.799*** 0.481*** 0.5549 0.9517 CBT+Forehead[La-]sw 0.871*** 0.460*** 0.4283 0.9407 CBT+Chest[La-]sw 0.854*** 0.503*** 0.4691 0.9873 CBT+Back[La-]sw 0.838*** 0.522*** 0.4816 0.9244 CBT+Forehead LER 0.901*** 0.512*** 0.3662 0.8807 CBT+Chest LER 0.854*** 0.503*** 0.4679 0.9804 CBT+Back LER 0.877*** 0.505*** 0.4507 1.0101 *** p <0.001, Note: RMSE values are reported on the log-transformed BLa scale (log units). Agreement Between Predicted and Observed BLa. Figure 3 presents the agreement between predicted and observed log-transformed blood lactate concentration (Log[BLa]) values across different body regions, based on conditional predictions from the linear mixed-effects model (LMM). Log-scale values can be back-transformed to the original scale using the exponential function (BLa = exp(Log[BLa])). Therefore, differences on the log scale correspond to multiplicative (ratio) differences in BLa. Predictive models incorporating HR, CBT, and sweat-derived variables showed strong consistency between predicted and measured values. Among [La⁻]ₛw-based models, combining forehead [La⁻]ₛw, HR, and CBT yielded the highest predictive accuracy (R² = 0.933, RMSE = 0.4204; Figure 3A), followed by chest [La⁻]ₛw (R² = 0.921, RMSE = 0.4109; Figure 3B) and back [La⁻]ₛw (R² = 0.926, RMSE = 0.4289; Figure 3C). When LER was used instead of concentration, similar predictive performance was observed. The forehead LER model showed the strongest agreement (R² = 0.938, RMSE = 0.4637; Figure 3D). The chest and back LER models also exhibited high predictive accuracy (R² = 0.921, RMSE = 0.4432; Figure 3E; R² = 0.923, RMSE = 0.4544; Figure 3F). The LMM models achieved high predictive accuracy for Log(BLa) across all body regions, indicating consistently strong relative agreement in (multiplicative) terms after back-transformation to the original BLa scale. Discussion This study demonstrated that blood lactate concentration during graded exercise can be reliably estimated by integrating non-invasive physiological signals. Specifically, HR and CBT exhibited strong and moderate positive correlations with BLa, whereas the lactate excretion rate (LER) showed a significant yet modest correlation. Regression analyses revealed that combined models incorporating HR, CBT, and sweat-derived parameters—particularly forehead LER—achieved the highest predictive accuracy, with conditional R² values > 0.93. Linear mixed-effects modeling enhanced interpretability and stability of predictions across participants. Collectively, these findings highlight the feasibility of integrated cardiovascular, thermoregulatory, and sweat-based indices for non-invasive monitoring of metabolic stress, with applications in exercise testing, performance optimization, and clinical rehabilitation. In this study, all measured physiological variables exhibited characteristic response patterns across progressive exercise stages, reflecting coordinated cardiovascular, thermoregulatory, and metabolic system activation. BLa increased progressively with exercise intensity, consistent with established lactate accumulation patterns during incremental workloads[1]. HR increased in a near-linear pattern before plateauing at maximal exertion (117.3 ± 14.1 → 182.1 ± 16.0 bpm; Figure 1), aligning with previous reports showing that HR increases steadily with exercise intensity and levels off as sympathetic activation reaches maximum capacity[13,18]. CBT rose continuously throughout all stages, indicating cumulative heat storage as workload increased—a physiological response attributed to metabolic heat production-dissipation imbalance)[9,19]. In contrast, [La⁻]ₛw significantly decreased with increasing workload, whereas LER increased. This inverse relationship reflects dilutional effects associated with rising sweat rates at higher exercise intensities[8, 9]. Collectively, these results represented typical physiological adaptations to incremental exercise and confirmed that the variables measured appropriately characterize metabolic, cardiovascular, and thermoregulatory stress. Correlation analyses were conducted to evaluate the physiological relationships among these variables during exercise. HR showed a strong positive correlation with BLa (r = 0.814, P <0.001), consistent with prior evidence linking HR to metabolic intensity and lactate kinetics during graded exercise[12,20]. CBT demonstrated a moderate positive association with BLa (r = 0.563, P <0.001), aligning with previous findings that elevated internal temperature parallels increased lactate accumulation owing to enhanced glycolytic flux under thermal load[10.14]. The weaker correlation of CBT relative to HR likely reflects the multifactorial nature of thermoregulation, which is influenced by metabolic heat production, heat dissipation mechanisms such as sweating, skin blood flow, and environmental conditions[21,22]. Prior studies have similarly reported progressive increases in CBT with workload; however, it shows greater variability due to environmental and individual thermoregulatory differences, weakening its direct association with lactate kinetics[19,23]. [La⁻]ₛw exhibited the weakest negative correlations with BLa (–0.135≤ r ≤–0.250) among the measured variables in this study. This weak association reflects the strong influence of local factors on [La⁻]ₛw rather than systemic metabolic status. Specifically, sweat composition depends on sweat gland activity, skin blood flow, evaporation rate, and local dilution effects, which vary across body sites and individuals. With increasing exercise intensity, total lactate excretion increases; however, sweat concentration decreases owing to higher sweat volume and dilution, obscuring its direct relationship with blood lactate[8,9].Regional variability (forehead vs. chest vs. back) and temporal lag between sweat and blood lactate changes further reduce the predictive value of sweat lactate concentation[24]. Regression analyses further demonstrated that models combining HR, CBT, and sweat-derived parameters yielded the most accurate predictions of BLa (Table 1). Conditional and marginal R² values and RMSE showed that integrated models consistently outperformed single-variable approaches, consistent with recent findings emphasizing multi-signal integration for physiological modeling[5]. The model combining HR, CBT, and forehead LER achieved the best performance (conditional R² = 0.939; marginal R² = 0.788; RMSE = 0.2288). This suggested that forehead-derived signals may offer practical advantages owing easier sensor attachment and enhanced signal stability in real-world monitoring. Similarly, models pairing CBT with sweat-derived variables improved prediction accuracy compared with CBT alone, consistent with prior studies reporting the additive predictive value of thermal and metabolic markers[10,25]. Integrating cardiovascular and thermoregulatory responses (HR and CBT) with peripheral sweat-based metrics enhances overall predictive capability. Log transformation of lactate data and application of the linear mixed-effects model (LMM) improved model fit and interpretability, as recommended for physiological datasets exhibiting non-linear accumulation patterns[26]. The LMM incorporated both fixed (HR, CBT, and sweat-derived indices) and random effects (participant variability), enabling accurate estimation of within-subject trends. Although results are presented on the log scale, model effects can be interpreted on the original BLa scale via back-transformation; specifically, exp(β) denotes a multiplicative change (ratio) in BLa for a one-unit change in the predictor, and log-scale predicted values can be converted to BLa using BLa = exp(Log[BLa]). Conditional R² values across all body regions ranged from 0.921 to 0.939, indicating consistent predictive accuracy between predicted and observed log-transformed BLa values. This important methodological approach improves model stability and interpretability, particularly during high-intensity exercise stages. Despite these promising findings, this study had several limitations. First, this study was conducted in a controlled laboratory setting with a relatively homogeneous group of healthy young adults. Future research should validate this model in more diverse populations, including athletes, clinical patients, and different environmental conditions. Second, manual and intermittent sweat sampling was performed may have limited temporal resolution. The use of wearable, real-time sweat biosensors could improve capture of dynamic lactate fluctuations with higher fidelity. Third, although HR, CBT, and sweat-derived parameters captured key physiological domains, integrating additional signals—such as skin temperature response—may further enhance prediction accuracy. Collectively, these findings demonstrated that integrating non-invasive physiological signals—specifically HR, CBT, and sweat-derived lactate measures—provides a robust and physiologically interpretable framework for estimating BLa during graded exercise. The strong agreement between predicted and observed outcomes confirms the method’s applicability for real-time, non-invasive metabolic monitoring, with potential implications for individualized exercise testing, performance optimization, and clinical rehabilitation. Methods Participants. Thirty-one healthy adult males aged 20–39 years were recruited. All participants engaged in ≥150 min of exercise per week for ≥3 months and maintained stable body weight (±10%) during the previous 6 months. Table 2 presents the physical characteristics of participants (n = 31). None used medication and all complied with a 12‑hour fast before testing. Exclusion criteria included metabolic, cardiovascular, or neurological disorders; Type 1 or Type 2 diabetes; orthopedic limitations; history of surgery; or non-obesity related cancer. This study was approved by the Institutional Review Board of Korea University (KUIRB‑2024‑0280‑01), and written informed consent was obtained from all participants. All methods were carried out in accordance with relevant guidelines and regulations and the Declaration of Helsinki. Table 2. Descriptive characteristics of participants (n=31) Mean±SD Minimum Maximum Age(yr) 26.2±5.82 18 37 Height(cm) 176±4.96 167 188 Weight(kg) 73.4±9.69 58 99.9 Body Fat(%) 14.7±4.5 6.5 23.3 Experimental p rocedure . Participants visited the laboratory on two occasions. The indoor environment was maintained at 29–31 °C and 50–70% relative humidity. At visit 1, written informed consent was obtained, followed by a preliminary assessment to confirm adequate sweating capacity. Body composition was assessed using a stadiometer and bioelectrical impedance analysis (InBody 270, Biospace, Seoul, Korea). While wearing a heart-rate monitor (Polar 810i, Polar Electro Oy, Finland), participants performed 30 min of treadmill exercise at 60% intensity based on previous protocols[7,10]. Using a gravimetric method[27], participants with a local sweat rate (LSR) ≤ 0.4 mg/cm2/min were excluded. For visit 2, participants were instructed to refrain from alcohol consumption, strenuous exercise, and smoking for at least 24 h before testing. They also fasted for ≥ 12 h, avoided caffeine-containing beverages, and maintained adequate hydration. These procedures were implemented to standardize physiological conditions and minimize potential confounding factors. Participants consumed the telemetric core-body temperature capsule with water approximately 2 h before testing. A licensed nurse collected venous blood samples immediately after each exercise stage. Thereafter, sweat samples were collected using double-film absorbent patches placed on the forehead, chest, and upper back. Exercise Protocol. The exercise test was performed on a motorized treadmill (Drex NR20 NR20X, Drex, Korea) according to a modified Bruce protocol (Figure 4). The test began with a 10-minute warm-up phase at 4.6 km/h and 0% incline, followed by a 2-minute rest period. The modified Bruce protocol was adopted to secure sufficient sweat volume over progressive workloads, enabling reliable sampling across both pre- and post-lactate threshold stages. This gradual intensity progression ensured stable thermoregulatory activation, facilitating continuous sweat excretion while delaying abrupt lactate accumulation. The protocol was suitable for this study because sustained sweat availability was essential for analyzing sweat-derived lactate biomarkers throughout dynamic metabolic transitions. The incremental treadmill subsequently commenced, consisting of seven mandatory 5-minute stages, with a 2-minute rests between stages. Speed increased by 0.5 km/h and incline by 1% per stage, as detailed in Figure 4. The test continued until volitional exhaustion, defined inability to maintain the required pace despite verbal encouragement. Blood Sampling and Analysis. A 22 G catheter (Kovax-Cath, Korea) was inserted into the antecubital vein of the non-dominant arm before exercise. Venous blood samples (~3 mL) were collected immediately after each exercise stage. Samples were stored at 4 °C and analyzed colorimetrically for blood lactate concentration at a certified clinical laboratory (Seegene Medical Foundation, Seoul, Korea). Measurements Heart Rate (HR). HR was continuously monitored using a chest-worn heart rate sensor. Resting HR was recorded during a seated rest period before exercise. HR values were recorded 10 s before each stage ended. Sweat Sampling and Analysis. Sweat collection patches were applied after the warm-up phase, after skin cleansing with 70% isopropyl alcohol and deionized water to remove debris and oils. Sweat samples were collected continuously using double-film patches placed on three the forehead (midline, 1 cm above the eyebrows), chest (lateral side, near the pectoral muscle), and upper back (most medial side of the scapula), based on an anatomical mapping method 15 . Each patch consisted of an absorbent cotton layer, a parafilm barrier ((Disposable consumable laboratory film, CNWTC, Chongqing, China), and a transparent waterproof dressing (Tegaderm Film, 3M, St. Paul, MN, USA). Cotton sizes were 16.8, 33.6, and 33.6 cm² for the forehead, chest, and back, respectively, with corresponding parafilm sizes of 19.2, 38.2, and 38.4 cm. Patches were secured during each 5-minute exercise stage and replaced afterward. Only sweat patches producing ≥2 µL of sweat were analyzed to ensure adequate sample size. Collected sweat samples were immediately weighed using an analytical balance (Hansung HS-103, Hansung Instrument, South Korea; precision: 0.001 g) to determine sweat volume, and subsequently centrifuged at 2,000 rpm for 5 min at 4 °C (LC-8S, JoanLab, China) to separate the supernatant for further analysis. Samples were stored at 4 °C prior to analysis. [La⁻]sw was quantified amperometrically at 0.1 V with a calibrated lactate sensor (Dongwoo Fine-chem Inc.) connected to a CHI-630 electrochemical workstation (CH Instruments, Inc.). The sensor was calibrated daily with standard lactate solutions (0–20 mM) to ensure measurement accuracy within ±5%. Core Body Temperature (CBT). CBT was continuously monitored using an ingestible telemetric capsule transmitting data wirelessly to an external receiver (e-Viewer® Performance). Data were recorded throughout the test and recovery period to assess thermoregulatory and metabolic responses. Statistical Analyses. Statistical analyses were performed using Jamovi version 2.3 for Mac (The Jamovi Project, Sydney, Australia). Statistical significance was set at α = 0.05, data are presented as the mean ± standard deviation (SD) unless otherwise stated. Linear mixed-effects models (LMMs) were assessed associations between blood lactate concentration (BLa) and non-invasive physiological variables (HR, CBT, [La⁻]ₛw, and LER). BLa data were log-transformed to improve normality, and participant-level random effects accounted for repeated measures across exercise stages. Pearson’s correlation analyses additionally assessed linear relationships between BLa and each non-invasive variable. To confirm the robustness of the LMM findings, a supplementary random forest regression was applied using the same input variables (HR, CBT, and sweat-derived lactate variables) to explore the predictive relationships under a non-parametric framework. This analysis served as a secondary reference, suggesting that the variable combination retained predictive relevance beyond linear assumptions. Perspective Increasing interest in wearable biosensors has renewed the need for valid, non-invasive indicators of metabolic strain during exercise. Our findings show that integrating HR and CBT with sweat-derived lactate indices, particularly forehead LER, enables accurate estimation of log-transformed blood lactate during incremental running. This supports the view that multi-signal monitoring can outperform single-marker approaches when lactate dynamics are influenced by cardiovascular, thermoregulatory, and sweat-gland factors. Practically, these models may facilitate real-time intensity profiling and individualized training prescription without repeated blood sampling. Future studies should validate performance across sex, fitness levels, environments (heat/humidity), and exercise modes, and test device-level implementation with prospective external validation and error thresholds that are meaningful for coaching and clinical decision-making. Declarations Funding This work was supported by Dongwoo Fine-Chem Co., Ltd. (Republic of Korea). Conflicts of Interest This study was funded by Dongwoo Fine-Chem Co., Ltd. S. H. Chon and Y. K. Lee is an employees of Dongwoo Fine-Chem Co., Ltd. The authors declare no other conflicts of interest (The sponsor had no role in the study design, data analysis, or interpretation). Author Contributions Conceptualization: Jaesung Lee, Jihye Moon, Jonghoon Park. Investigation: Jaesung Lee, Jihye Moon, Seunghwan Chon, Youngkeun Lee. Formal analysis: Jaesung Lee, Jihye Moon, Youngim Kim, Hyeonmin Kim, Hyunseob Lee, Sungjin Yoon. Data curation: Jaesung Lee, Jihye Moon. Visualization: Jaesung Lee, Jihye Moon. Writing – original draft: Jaesung Lee, Jihye Moon. Writing – review & editing: Youngim Kim, Hyeonmin Kim, Eunbi Kim, Hyunseob Lee, Sungjin Yoon, Seunghwan Chon, Youngkeun Lee. Supervision: Jonghoon Park. Project administration: Jonghoon Park. Funding acquisition: Jonghoon Park. All authors have read and approved the final version of the manuscript. Data Availability The data supporting the findings of this study, including the raw and processed data underlying the figures and tables, are available from the corresponding author upon reasonable request. Ethics Approval The study protocol was approved by the Institutional Review Board of Korea University (KUIRB‑2024‑0280‑01) Patient consent Written informed consent was obtained from all participants for study participation and the use of their data for research purposes. Permission to reproduce Not applicable. Clinical trial registration Not applicable. Acknowledgements This research was supported by a fund from Dongwoo Fine-Chem Co., Ltd. in 2024–2025. References Beneke, R., Leithäuser, R. M. & Ochentel, O. Blood lactate diagnostics in exercise testing and training. Int. J. Sports Physiol. Perform. 6 , 8–24. https://doi.org/10.1123/ijspp.6.1.8 (2011). Goodwin, M. L., Harris, J. E., Hernández, A. & Gladden, L. B. Blood lactate measurements and analysis during exercise: A guide for clinicians. J. Diabetes Sci. Technol. 1 , 558–569. https://doi.org/10.1177/193229680700100414 (2007). Khabour, O. F., Ali, A., Mahallawi, W. H. & K. H. & Occupational infection and needle stick injury among clinical laboratory workers in Al-Madinah city, Saudi Arabia. J. Occup. Med. Toxicol. 13 , 15. https://doi.org/10.1186/s12995-018-0198-5 (2018). Gao, W. et al. Fully integrated wearable sensor arrays for multiplexed in situ perspiration analysis. Nature 529 , 509–514. https://doi.org/10.1038/nature16521 (2016). Yang, G., Ryu, S., Kim, K. & Kim, J. Wearable device for continuous sweat lactate monitoring in sports. Front. Physiol. 15 , 1376801. https://doi.org/10.3389/fphys.2024.1376801 (2024). Karpova, E. V., Laptev, A. I., Andreev, E. A., Karyakina, E. E. & Karyakin, A. A. Relationship between sweat and blood lactate levels during exhaustive physical exercise. ChemElectroChem 7, 191–194, (2020). https://doi.org/10.1002/celc.201901703 Okawara, H., Sawada, T., Nakashima, D., Fujitsuka, H. & Katsumata, Y. Lactate threshold evaluation in swimming using a sweat lactate sensor: A prospective study. Eur. J. Sport Sci. 24 , 1302–1312. https://doi.org/10.1002/ejsc.12179 (2024). Takei, N. et al. Differential patterns of sweat and blood lactate concentration response during incremental exercise in varied ambient temperatures: A pilot study. Temp. (Austin) . 11 , 247–253. https://doi.org/10.1080/23328940.2024.2375693 (2024). Seki, Y. et al. A novel device for detecting anaerobic threshold using sweat lactate during exercise. Sci. Rep. 11 , 4929. https://doi.org/10.1038/s41598-021-84381-9 (2021). Buono, M. J., Lee, N. V. L. & Miller, P. W. The relationship between exercise intensity and the sweat lactate excretion rate. J. Physiol. Sci. 60 , 103–107. https://doi.org/10.1007/s12576-009-0073-8 (2010). Taylor, N. A. & Machado-Moreira, C. A. Regional variations in transepidermal water loss, eccrine sweat gland density, sweat secretion rates and electrolyte composition in resting and exercising humans. Extrem. Physiol. Med. 2 , 4. https://doi.org/10.1186/2046-7648-2-4 (2013). Bentley, D. J., Newell, J. & Bishop, D. Incremental exercise test design and analysis: Implications for performance diagnostics in endurance athletes. Sports Med. 37 , 575–586. https://doi.org/10.2165/00007256-200737070-00002 (2007). Achten, J. & Jeukendrup, A. E. Heart rate monitoring: Applications and limitations. Sports Med. 33 , 517–538. https://doi.org/10.2165/00007256-200333070-00004 (2003). Febbraio, M. A. Alterations in energy metabolism during exercise and heat stress. Sports Med. 31 , 47–59. https://doi.org/10.2165/00007256-200131010-00004 (2001). van Hall, G. Lactate kinetics in human tissues at rest and during exercise. Acta Physiol. 199 , 499–508. https://doi.org/10.1111/j.1748-1716.2010.02122.x (2010). Armstrong, L. E. et al. American College of Sports Medicine position stand: Exertional heat illness during training and competition. Med. Sci. Sports Exerc. 39 , 556–572. https://doi.org/10.1249/MSS.0b013e31802fa199 (2007). Gerrett, N. et al. Thermal sensitivity to warmth during rest and exercise: A sex comparison. Eur. J. Appl. Physiol. 114 , 1451–1462. https://doi.org/10.1007/s00421-014-2875-0 (2014). Vieira, S. S. et al. Does stroke volume increase during an incremental exercise? A systematic review. Open. Cardiovasc. Med. J. 10 , 57–63. https://doi.org/10.2174/1874192401610010057 (2016). Sawka, M. N. & Wenger, C. B. Physiological responses to acute exercise-heat stress. In Human Performance Physiology and Environmental Medicine at Terrestrial Extremes (eds Pandolf, K. B., Sawka, M. N. & Gonzalez, R. R.) . 97–151Benchmark Press,. (1988). Shen, T. & Wen, X. Heart rate based prediction of velocity at lactate threshold in ordinary adults. J. Exerc. Sci. Fit. 17 , 108–112. https://doi.org/10.1016/j.jesf.2019.07.002 (2019). Periard, J. D., Eijsvogels, T. M. H. & Daanen, H. A. M. Exercise under heat stres : Thermoregulation, hydration, performance implications, and mitigation strategies. Physiol Rev 101, 1873– (1979). https://doi.org/10.1152/physrev.00038.2020 (2021). Lindinger, M. Homeostasis and its disturbance during exercise. In Open Textbook of Exercise Physiology (Open Educational Alberta ,). (2023). https://pressbooks.openeducationalberta.ca/physiologyexercise/ Lim, J., Park, H., Lee, S. & Park, J. Changes in heart rate, muscle temperature, blood lactate concentration, blood pressure, and fatigue perception following jogging and running: An observational study. Exerc. Sci. 31 , 72–79. https://doi.org/10.15857/ksep.2022.00045 (2022). Heikenfeld, J. et al. Accessing analytes in biofluids for peripheral biochemical monitoring. Nat. Biotechnol. 37 , 407–419. https://doi.org/10.1038/s41587-019-0040-3 (2019). Rabost-Garcia, G. et al. Non-Invasive Multiparametric Approach To Determine Sweat-Blood Lactate Bioequivalence. ACS Sens. 8 , 1536–1541. https://doi.org/10.1021/acssensors.2c02614 (2023). Bates, D., Mächler, M., Bolker, B. & Walker, S. Fitting Linear Mixed-Effects Models Using lme4. J. Stat. Softw. 67 , 1–48. https://doi.org/10.18637/jss.v067.i01 (2015). Morris, N. B., Cramer, M. N., Hodder, S. G., Havenith, G. & Jay, O. A comparison between the technical absorbent and ventilated capsule methods for measuring local sweat rate. J. Appl. Physiol. (1985) . 114 , 816–823. https://doi.org/10.1152/japplphysiol.01088.2012 (2013). Additional Declarations Competing interest reported. This study was funded by Dongwoo Fine-Chem Co., Ltd. Seunghwan Chon and Youngkeun Lee are employees of Dongwoo Fine-Chem Co., Ltd. The sponsor had no role in the study design, data collection, analysis, interpretation, or manuscript preparation. All other authors declare no competing interests. Cite Share Download PDF Status: Published Journal Publication published 07 Apr, 2026 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 06 Mar, 2026 Reviews received at journal 05 Mar, 2026 Reviews received at journal 21 Feb, 2026 Reviewers agreed at journal 13 Feb, 2026 Reviewers agreed at journal 10 Feb, 2026 Reviewers invited by journal 02 Feb, 2026 Editor assigned by journal 30 Jan, 2026 Editor invited by journal 06 Jan, 2026 Submission checks completed at journal 05 Jan, 2026 First submitted to journal 05 Jan, 2026 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-8498078","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":584773699,"identity":"ab40c76f-58ad-4416-9fe7-77bd48245866","order_by":0,"name":"Jaesung Lee","email":"","orcid":"","institution":"Korea University","correspondingAuthor":false,"prefix":"","firstName":"Jaesung","middleName":"","lastName":"Lee","suffix":""},{"id":584773700,"identity":"54f2256d-4ec4-4ad8-8fd9-c98dabece0ef","order_by":1,"name":"Jihye Moon","email":"","orcid":"","institution":"Korea University","correspondingAuthor":false,"prefix":"","firstName":"Jihye","middleName":"","lastName":"Moon","suffix":""},{"id":584773701,"identity":"8774331b-e7e9-49e2-8b97-5d5fa9c1cd32","order_by":2,"name":"Youngim Kim","email":"","orcid":"","institution":"Korea University","correspondingAuthor":false,"prefix":"","firstName":"Youngim","middleName":"","lastName":"Kim","suffix":""},{"id":584773702,"identity":"2bf6ce5c-46a9-42d4-bc51-ac5f32621587","order_by":3,"name":"Hyeonmin Kim","email":"","orcid":"","institution":"Korea University","correspondingAuthor":false,"prefix":"","firstName":"Hyeonmin","middleName":"","lastName":"Kim","suffix":""},{"id":584773703,"identity":"929da7c1-3257-4ae6-9c46-a4dd748975c7","order_by":4,"name":"Eunbi Kim","email":"","orcid":"","institution":"Korea University","correspondingAuthor":false,"prefix":"","firstName":"Eunbi","middleName":"","lastName":"Kim","suffix":""},{"id":584773704,"identity":"b3115ab0-9e6a-41d0-b617-a9356156b65e","order_by":5,"name":"Hyunseob Lee","email":"","orcid":"","institution":"Korea University","correspondingAuthor":false,"prefix":"","firstName":"Hyunseob","middleName":"","lastName":"Lee","suffix":""},{"id":584773705,"identity":"43c9bfe4-656f-4c56-8c36-7a944c83f6c8","order_by":6,"name":"Sungjin Yoon","email":"","orcid":"","institution":"Korea University","correspondingAuthor":false,"prefix":"","firstName":"Sungjin","middleName":"","lastName":"Yoon","suffix":""},{"id":584773706,"identity":"1e6bbaa4-9d0c-4e3d-85a0-4034714edd56","order_by":7,"name":"Seunghwan Chon","email":"","orcid":"","institution":"Dongwoo Fine-Chem Co., Ltd. AR\u0026D center","correspondingAuthor":false,"prefix":"","firstName":"Seunghwan","middleName":"","lastName":"Chon","suffix":""},{"id":584773707,"identity":"851a37db-0f4f-4f07-82f5-80b78f13150a","order_by":8,"name":"Youngkeun Lee","email":"","orcid":"","institution":"Dongwoo Fine-Chem Co., Ltd. AR\u0026D center","correspondingAuthor":false,"prefix":"","firstName":"Youngkeun","middleName":"","lastName":"Lee","suffix":""},{"id":584773708,"identity":"b9b90ea6-8bfc-406e-8f3d-e89dcaae3231","order_by":9,"name":"Jonghoon Park","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA5klEQVRIiWNgGAWjYBACxgYGBmYwi70NLshMpBaeY0RqQaiQSCNSC3N77+HXBRU2eeaSzxIfV1TYMfC3H2A2rsDnsJ5zadYzzqQVW85OO2x45kwyg8SZBObEM/i0zMgxM+ZtO5y44XZ6m2RjG9BNNxiYDzYQpeXmcaCWf/UM8kRoMX4M1nKD7ZhkY8NhBgOglkS8WnrOmDHznElL3NmTlmzYcOw4j+GZxGZDfFoM23uMP/NU2CRuZz9m+LChplpO7vjhw5J4tTQwsEmAGAZQAR5I9OIB8sCo+YCsZRSMglEwCkYBBgAA/r9MhtrfXdUAAAAASUVORK5CYII=","orcid":"","institution":"Korea University","correspondingAuthor":true,"prefix":"","firstName":"Jonghoon","middleName":"","lastName":"Park","suffix":""}],"badges":[],"createdAt":"2026-01-02 04:53:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8498078/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8498078/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-026-47148-8","type":"published","date":"2026-04-07T15:58:16+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":101852856,"identity":"8bd6ec33-d21b-4d91-808d-c5acba3a521b","added_by":"auto","created_at":"2026-02-04 10:18:14","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":119024,"visible":true,"origin":"","legend":"\u003cp\u003eChanges in blood lactate (BLa), heart rate (HR), core body temperature (CBT), sweat lactate concentration ([La⁻]sw), and lactate excretion rate (LER) during incremental exercise.\u003c/p\u003e\n\u003cp\u003eAbbreviations: BLa, Blood Lactate concentration; HR, Heart Rate; CBT, Core Body Temperature; [La⁻]sw, sweat lactate concentration; LER, Lactate excretion rate\u003c/p\u003e","description":"","filename":"image1.png","url":"https://assets-eu.researchsquare.com/files/rs-8498078/v1/77eac8094605e4bcffc10d29.png"},{"id":101852857,"identity":"2c40567e-eb35-42a6-9392-6eb836bf6817","added_by":"auto","created_at":"2026-02-04 10:18:14","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":72354,"visible":true,"origin":"","legend":"\u003cp\u003eCorrelation between blood lactate (BLa) and non-invasive physiological signals during incremental exercise.\u003c/p\u003e\n\u003cp\u003eAbbreviations: BLa, blood lactate concentration; HR, heart rate; CBT, core body temperature; [La⁻]sw, sweat lactate concentration; LER, lactate excretion rate.\u003c/p\u003e","description":"","filename":"image2.png","url":"https://assets-eu.researchsquare.com/files/rs-8498078/v1/6154810e6a1cbc541239ead9.png"},{"id":101852859,"identity":"cc3b4856-ddf7-43e6-959f-d004a781f0ac","added_by":"auto","created_at":"2026-02-04 10:18:15","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":281516,"visible":true,"origin":"","legend":"\u003cp\u003eAgreement between predicted and observed log-transformed blood lactate concentration (log[BLa]) across body regions based on conditional predictions from the linear mixed-effects model (LMM).\u003c/p\u003e\n\u003cp\u003eNote: All input variables (HR, CBT, and [La⁻]sw) were standardized (z-scores) to account for scale differences and multicollinearity. Abbreviations: BLa, blood lactate concentration; HR, heart rate; CBT, core body temperature; [La⁻]sw, sweat lactate concentration; LMM, linear mixed-effects model.\u003c/p\u003e","description":"","filename":"image3.png","url":"https://assets-eu.researchsquare.com/files/rs-8498078/v1/447ac2a9668178339e41bf42.png"},{"id":101852858,"identity":"4dbc441b-6cac-4a01-805a-22122a115dc1","added_by":"auto","created_at":"2026-02-04 10:18:15","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":66106,"visible":true,"origin":"","legend":"\u003cp\u003eOverview of the experimental protocol: screening and main test procedures.\u003c/p\u003e","description":"","filename":"image4.png","url":"https://assets-eu.researchsquare.com/files/rs-8498078/v1/f722309027a73c176f797619.png"},{"id":106808883,"identity":"e9f4c3a4-8e2b-4243-b23d-8d187a0c66db","added_by":"auto","created_at":"2026-04-13 16:04:25","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1237107,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8498078/v1/2512c780-a5f7-40cd-8f96-9b7c181c1dc9.pdf"}],"financialInterests":"Competing interest reported. This study was funded by Dongwoo Fine-Chem Co., Ltd. Seunghwan Chon and Youngkeun Lee are employees of Dongwoo Fine-Chem Co., Ltd. The sponsor had no role in the study design, data collection, analysis, interpretation, or manuscript preparation. All other authors declare no competing interests.","formattedTitle":"Non-Invasive Prediction of Blood Lactate During Incremental Exercise via Heart Rate, Core Body Temperature, and Sweat-Derived Indices","fulltext":[{"header":"Introduction","content":"\u003cp\u003eBlood lactate concentration (BLa) increases with exercise intensity and serves as a key indicator of aerobic capacity and endurance performance; accordingly, it is widely used in sports science and clinical rehabilitation[1,2]. However, conventional BLa assessment during exercise requires invasive blood sampling, which limits continuous and real-time monitoring due to discomfort, procedural complexity, and potential infection risk[3].\u003c/p\u003e\n\u003cp\u003eRecent advances in wearable devices for sweat lactate measurement have enabled non-invasive approaches to estimating BLa[4-7]. Importantly, sweat lactate concentration ([La⁻]ₛw) does not reflect passive diffusion from blood. It can also be locally produced through glycolytic activity within sweat gland cells, and is influenced by multiple physiological and environmental factors, including sweat rate, skin blood flow, and thermal stress[8,9]. As sweat rate increases, dilution effects may attenuate the association between [La⁻]ₛw and BLa[9]. To partially account for dilution, lactate excretion rate (LER; [La⁻]ₛw × sweat rate) has been proposed as a more informative index[10]. Nevertheless, regional differences in sweat gland density and sweating responses, as well as variation in sensor attachment and skin–sensor contact, may limit the representativeness of a single local measurement[11]. This suggests that additional physiological variables may be necessary to improve the robustness of non-invasive BLa estimation.\u003c/p\u003e\n\u003cp\u003eHeart rate (HR) and core body temperature (CBT) are potential physiological signals that may reflect metabolic strain beyond cardiovascular or sweat-derived indices[12,13]. HR typically increases with exercise intensity and shows moderate-to-strong associations with BLa, supporting its use for lactate estimation[12]. However, inter-individual variability in cardiovascular responses, training status, and fatigue resistance can reduce predictive reliability[13]. CBT also rises progressively during exercise due to metabolic heat accumulation, and elevations in body temperature have been linked to increased glycolytic activation and lactate accumulation[14,15].\u0026nbsp;Despite this physiological rationale, CBT has rarely been incorporated as a predictor of lactate dynamics, as prior work has primarily focused on thermoregulatory function or heat-related illness risk[16].\u003c/p\u003e\n\u003cp\u003eAccordingly, the aim of this study was to develop and validate multivariable models to predict BLa during incremental exercise using only non-invasive physiological signals. We constructed regression models incorporating HR, CBT, and sweat-derived lactate indices. Based on the optimal combination of explanatory variables identified using a linear mixed-effects model, we additionally applied a random forest algorithm using the same inputs to evaluate predictive performance.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eStage-dependent Responses of Physiological Variables\u003c/strong\u003e\u003cstrong\u003e.\u0026nbsp;\u003c/strong\u003eAs shown in Figure 1, blood lactate concentration (BLa) increased progressively with increasing exercise intensity, showing marked elevation from stage 4 and peaking at stage 7 (p \u0026lt; 0.001; Figure 1A). Heart rate (HR) and core body temperature (CBT) significantly increased with workload (p \u0026lt; 0.001; Figure 1B), with HR increasing steadily from ~100 bpm to ~180 bpm and plateauing near stage 7, and CBT rising sharply between stages 6 and 8 to ~39 \u0026deg;C. Lactate excretion rate (LER) at the forehead, chest, and back significantly increased with exercise intensity (p \u0026lt; 0.001; Figure 1D), with the forehead showing the largest rise before stabilizing at higher stages. In contrast, sweat lactate concentration ([La⁻]ₛw) significantly decreased across stages (p \u0026lt; 0.001; Figure 1C), peaking at stage 0 and a progressive decline that stabilized from stage 5 onward.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCorrelations Between BLa and Non-Invasive Physiological Variables.\u0026nbsp;\u003c/strong\u003eFigure 2 shows the correlations between BLa and non-invasive physiological variables. HR strongly correlated with BLa (r = 0.814, p \u0026lt; 0.001; Figure 2A), and CBT showed a moderate positive correlation (r = 0.563, p \u0026lt; 0.001; Figure 2B). [La⁻]ₛw exhibited weak negative correlations across measurement sites\u0026mdash;forehead (r = \u0026ndash;0.242, p = 0.002; Figure 2C) and back (r = \u0026ndash;0.250, p \u0026lt; 0.001; Figure 2E). LER showed significant positive correlations with BLa\u0026mdash;forehead (r = 0.391, p \u0026lt; 0.001; Figure 2F), chest (r = 0.254, p \u0026lt; 0.001; Figure 2G), and back (r = 0.224, p \u0026lt; 0.001; Figure 2H).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRegression Models for Predicting Blood Lactate Concentration.\u0026nbsp;\u003c/strong\u003eTable 1 summarizes the regression performance for predicting BLa using various physiological signal combinations. Conditional and marginal R\u0026sup2; values and corresponding RMSE were used to evaluate each model\u0026rsquo;s explanatory power and prediction accuracy. Models that included HR demonstrated the highest predictive capability, with conditional R\u0026sup2; values consistently \u0026gt; 0.90. Among these, combining HR, CBT, and forehead LER yielded the best performance (conditional R\u0026sup2; = 0.939, marginal R\u0026sup2; = 0.788, conditional RMSE = 0.2288). Models using CBT alone showed moderate predictive accuracy (conditional R\u0026sup2; = 0.799, marginal R\u0026sup2; = 0.481). Predictive performance improved when CBT was combined with sweat-derived parameters such as LER or [La⁻]ₛw. The CBT + forehead LER model achieved a conditional R\u0026sup2; of 0.901 and a conditional RMSE of 0.3662. Overall, regression models integrating cardiovascular (HR) and thermoregulatory (CBT) variables with sweat-derived parameters (LER or [La⁻]ₛw) exhibited conditional R\u0026sup2; values ranging from 0.921 to 0.939.\u003c/p\u003e\n\u003cp\u003eTable 1. Regression Performance (R\u0026sup2;, RMSE) by physiological signal Combination\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"668\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eFactor\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;Conditional R\u0026sup2;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eMarginal R\u0026sup2;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eConditional RMSE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eMarginal RMSE\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eHR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.908***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.795***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.2973\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.4678\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eHR+CBT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.917***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.807***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.2818\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.4554\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eHR+Forehead[La-]sw\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.926***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.807***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.2648\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.4618\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eHR+Chest[La-]sw\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.918***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.808***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.2795\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.4510\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eHR+Back[La-]sw\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.916***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.791***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.2839\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.4681\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eHR+Forehead LER\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.921***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.785***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.2596\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.4561\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eHR+Chest LER\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.915***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.797***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.2779\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.4556\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eHR+Back LER\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.912***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.789***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.2829\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.4614\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eHR+CBT+Forehead[La-]sw\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.933***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.818***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.2429\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.4248\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eHR+CBT+Chest[La-]sw\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.921***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.828***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.2698\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.4231\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eHR+CBT+Back[La-]sw\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.926***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.799***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.2675\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.4634\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eHR+CBT+Forehead LER\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.939***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.788***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.2288\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.4572\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eHR+CBT+Chest LER\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.921***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.806***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.2684\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.4502\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eHR+CBT+Back LER\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.933***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.792***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.2668\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.4671\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eCBT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.799***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.481***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.5549\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.9517\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eCBT+Forehead[La-]sw\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.871***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.460***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.4283\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.9407\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eCBT+Chest[La-]sw\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.854***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.503***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.4691\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.9873\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eCBT+Back[La-]sw\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.838***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.522***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.4816\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.9244\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eCBT+Forehead LER\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.901***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.512***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.3662\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.8807\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eCBT+Chest LER\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.854***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.503***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.4679\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.9804\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eCBT+Back LER\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.877***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.505***\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.4507\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.0101\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003csup\u003e***\u003c/sup\u003e\u003cem\u003ep\u003c/em\u003e\u0026lt;0.001, Note: RMSE values are reported on the log-transformed BLa scale (log units).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAgreement Between Predicted and Observed BLa.\u0026nbsp;\u003c/strong\u003eFigure 3 presents the agreement between predicted and observed log-transformed blood lactate concentration (Log[BLa]) values across different body regions, based on conditional predictions from the linear mixed-effects model (LMM). Log-scale values can be back-transformed to the original scale using the exponential function (BLa = exp(Log[BLa])). Therefore, differences on the log scale correspond to multiplicative (ratio) differences in BLa. Predictive models incorporating HR, CBT, and sweat-derived variables showed strong consistency between predicted and measured values. Among [La⁻]ₛw-based models, combining forehead [La⁻]ₛw, HR, and CBT yielded the highest predictive accuracy (R\u0026sup2; = 0.933, RMSE = 0.4204; Figure 3A), followed by chest [La⁻]ₛw (R\u0026sup2; = 0.921, RMSE = 0.4109; Figure 3B) and back [La⁻]ₛw (R\u0026sup2; = 0.926, RMSE = 0.4289; Figure 3C). When LER was used instead of concentration, similar predictive performance was observed. The forehead LER model showed the strongest agreement (R\u0026sup2; = 0.938, RMSE = 0.4637; Figure 3D). The chest and back LER models also exhibited high predictive accuracy (R\u0026sup2; = 0.921, RMSE = 0.4432; Figure 3E; R\u0026sup2; = 0.923, RMSE = 0.4544; Figure 3F). The LMM models achieved high predictive accuracy for Log(BLa) across all body regions, indicating consistently strong relative agreement in (multiplicative) terms after back-transformation to the original BLa scale.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study demonstrated that blood lactate concentration during graded exercise can be reliably estimated by integrating non-invasive physiological signals. Specifically, HR and CBT exhibited strong and moderate positive correlations with BLa, whereas the lactate excretion rate (LER) showed a significant yet modest correlation. Regression analyses revealed that combined models incorporating HR, CBT, and sweat-derived parameters\u0026mdash;particularly forehead LER\u0026mdash;achieved the highest predictive accuracy, with conditional R\u0026sup2; values \u0026gt; 0.93. Linear mixed-effects modeling enhanced interpretability and stability of predictions across participants. Collectively, these findings highlight the feasibility of integrated cardiovascular, thermoregulatory, and sweat-based indices for non-invasive monitoring of metabolic stress, with applications in exercise testing, performance optimization, and clinical rehabilitation.\u003c/p\u003e\n\u003cp\u003eIn this study, all measured physiological variables exhibited characteristic response patterns across progressive exercise stages, reflecting coordinated cardiovascular, thermoregulatory, and metabolic system activation. BLa increased progressively with exercise intensity, consistent with established lactate accumulation patterns during incremental workloads[1]. HR increased in a near-linear pattern before plateauing at maximal exertion (117.3 \u0026plusmn; 14.1 \u0026rarr; 182.1 \u0026plusmn; 16.0 bpm; Figure 1), aligning with previous reports showing that HR increases steadily with exercise intensity and levels off as sympathetic activation reaches maximum capacity[13,18]. CBT rose continuously throughout all stages, indicating cumulative heat storage as workload increased\u0026mdash;a physiological response attributed to metabolic heat production-dissipation imbalance)[9,19]. In contrast, [La⁻]ₛw significantly decreased with increasing workload, whereas LER increased. This inverse relationship reflects dilutional effects associated with rising sweat rates at higher exercise intensities[8, 9]. Collectively, these results represented typical physiological adaptations to incremental exercise and confirmed that the variables measured appropriately characterize metabolic, cardiovascular, and thermoregulatory stress.\u003c/p\u003e\n\u003cp\u003eCorrelation analyses were conducted to evaluate the physiological relationships among these variables during exercise. HR showed a strong positive correlation with BLa (r = 0.814, P \u0026lt;0.001), consistent with prior evidence linking HR to metabolic intensity and lactate kinetics during graded exercise[12,20]. CBT demonstrated a moderate positive association with BLa (r = 0.563, P \u0026lt;0.001), aligning with previous findings that elevated internal temperature parallels increased lactate accumulation owing to enhanced glycolytic flux under thermal load[10.14]. The weaker correlation of CBT relative to HR likely reflects the multifactorial nature of thermoregulation, which is influenced by metabolic heat production, heat dissipation mechanisms such as sweating, skin blood flow, and environmental conditions[21,22]. Prior studies have similarly reported progressive increases in CBT with workload; however, it shows greater variability due to environmental and individual thermoregulatory differences, weakening its direct association with lactate kinetics[19,23].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e[La⁻]ₛw exhibited the weakest negative correlations with BLa (\u0026ndash;0.135\u0026le; r \u0026le;\u0026ndash;0.250) among the measured variables in this study. This weak association reflects the strong influence of local factors on [La⁻]ₛw rather than systemic metabolic status. Specifically, sweat composition depends on sweat gland activity, skin blood flow, evaporation rate, and local dilution effects, which vary across body sites and individuals. With increasing exercise intensity, total lactate excretion increases; however, sweat concentration decreases owing to higher sweat volume and dilution, obscuring its direct relationship with blood lactate[8,9].Regional variability (forehead vs. chest vs. back) and temporal lag between sweat and blood lactate changes further reduce the predictive value of sweat lactate concentation[24].\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eRegression analyses further demonstrated that models combining HR, CBT, and sweat-derived parameters yielded the most accurate predictions of BLa (Table 1). Conditional and marginal R\u0026sup2; values and RMSE showed that integrated models consistently outperformed single-variable approaches, consistent with recent findings emphasizing multi-signal integration for physiological modeling[5]. The model combining HR, CBT, and forehead LER achieved the best performance (conditional R\u0026sup2; = 0.939; marginal R\u0026sup2; = 0.788; RMSE = 0.2288). This suggested that forehead-derived signals may offer practical advantages owing easier sensor attachment and enhanced signal stability in real-world monitoring. Similarly, models pairing CBT with sweat-derived variables improved prediction accuracy compared with CBT alone, consistent with prior studies reporting the additive predictive value of thermal and metabolic markers[10,25]. Integrating cardiovascular and thermoregulatory responses (HR and CBT) with peripheral sweat-based metrics enhances overall predictive capability.\u003c/p\u003e\n\u003cp\u003eLog transformation of lactate data and application of the linear mixed-effects model (LMM) improved model fit and interpretability, as recommended for physiological datasets exhibiting non-linear accumulation patterns[26]. The LMM incorporated both fixed (HR, CBT, and sweat-derived indices) and random effects (participant variability), enabling accurate estimation of within-subject trends. Although results are presented on the log scale, model effects can be interpreted on the original BLa scale via back-transformation; specifically, exp(\u0026beta;) denotes a multiplicative change (ratio) in BLa for a one-unit change in the predictor, and log-scale predicted values can be converted to BLa using BLa = exp(Log[BLa]). Conditional R\u0026sup2; values across all body regions ranged from 0.921 to 0.939, indicating consistent predictive accuracy between predicted and observed log-transformed BLa values. This important methodological approach improves model stability and interpretability, particularly during high-intensity exercise stages.\u003c/p\u003e\n\u003cp\u003eDespite these promising findings, this study had several limitations. First, this study was conducted in a controlled laboratory setting with a relatively homogeneous group of healthy young adults. Future research should validate this model in more diverse populations, including athletes, clinical patients, and different environmental conditions. Second, manual and intermittent sweat sampling was performed may have limited temporal resolution. The use of wearable, real-time sweat biosensors could improve capture of dynamic lactate fluctuations with higher fidelity. Third, although HR, CBT, and sweat-derived parameters captured key physiological domains, integrating additional signals\u0026mdash;such as skin temperature response\u0026mdash;may further enhance prediction accuracy.\u003c/p\u003e\n\u003cp\u003eCollectively, these findings demonstrated that integrating non-invasive physiological signals\u0026mdash;specifically HR, CBT, and sweat-derived lactate measures\u0026mdash;provides a robust and physiologically interpretable framework for estimating BLa during graded exercise. The strong agreement between predicted and observed outcomes confirms the method\u0026rsquo;s applicability for real-time, non-invasive metabolic monitoring, with potential implications for individualized exercise testing, performance optimization, and clinical rehabilitation.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eParticipants.\u0026nbsp;\u003c/strong\u003eThirty-one healthy adult males aged 20\u0026ndash;39 years were recruited. All participants engaged in \u0026ge;150 min of exercise per week for \u0026ge;3 months and maintained stable body weight (\u0026plusmn;10%) during the previous 6 months. Table 2 presents the physical characteristics of participants (n = 31). None used medication and all complied with a 12‑hour fast before testing. Exclusion criteria included metabolic, cardiovascular, or neurological disorders; Type 1 or Type 2 diabetes; orthopedic limitations; history of surgery; or non-obesity related cancer. This study was approved by the Institutional Review Board of Korea University (KUIRB‑2024‑0280‑01), and written informed consent was obtained from all participants. All methods were carried out in accordance with relevant guidelines and regulations and the Declaration of Helsinki.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTable 2. Descriptive characteristics of participants (n=31)\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"599\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd style=\"width: 136px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMean\u0026plusmn;SD\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 115px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMinimum\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMaximum\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 234px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge(yr)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 136px;\"\u003e\n \u003cp\u003e26.2\u0026plusmn;5.82\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 115px;\"\u003e\n \u003cp\u003e18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e37\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 234px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHeight(cm)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 136px;\"\u003e\n \u003cp\u003e176\u0026plusmn;4.96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 115px;\"\u003e\n \u003cp\u003e167\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e188\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 234px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eWeight(kg)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 136px;\"\u003e\n \u003cp\u003e73.4\u0026plusmn;9.69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 115px;\"\u003e\n \u003cp\u003e58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e99.9\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 234px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eBody Fat(%)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 136px;\"\u003e\n \u003cp\u003e14.7\u0026plusmn;4.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 115px;\"\u003e\n \u003cp\u003e6.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 113px;\"\u003e\n \u003cp\u003e23.3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eExperimental p\u003c/strong\u003e\u003cstrong\u003erocedure\u003c/strong\u003e\u003cstrong\u003e.\u0026nbsp;\u003c/strong\u003eParticipants visited the laboratory on two occasions. The indoor environment was maintained at 29\u0026ndash;31 \u0026deg;C and 50\u0026ndash;70% relative humidity. At visit 1, written informed consent was obtained, followed by a preliminary assessment to confirm adequate sweating capacity. Body composition was assessed using a stadiometer and bioelectrical impedance analysis (InBody 270, Biospace, Seoul, Korea). While wearing a heart-rate monitor (Polar 810i, Polar Electro Oy, Finland), participants performed 30 min of treadmill exercise at 60% intensity based on previous protocols[7,10]. Using a gravimetric method[27], participants with a local sweat rate (LSR) \u0026le; 0.4 mg/cm2/min were excluded. For visit 2, participants were instructed to refrain from alcohol consumption, strenuous exercise, and smoking for at least 24 h before testing. They also fasted for \u0026ge; 12 h, avoided caffeine-containing beverages, and maintained adequate hydration. These procedures were implemented to standardize physiological conditions and minimize potential confounding factors. Participants consumed the telemetric core-body temperature capsule with water approximately 2 h before testing. A licensed nurse collected venous blood samples immediately after each exercise stage. Thereafter, sweat samples were collected using double-film absorbent patches placed on the forehead, chest, and upper back.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eExercise Protocol.\u0026nbsp;\u003c/strong\u003eThe exercise test was performed on a motorized treadmill (Drex NR20 NR20X, Drex, Korea) according to a modified Bruce protocol (Figure 4). The test began with a 10-minute warm-up phase at 4.6 km/h and 0% incline, followed by a 2-minute rest period. The modified Bruce protocol was adopted to secure sufficient sweat volume over progressive workloads, enabling reliable sampling across both pre- and post-lactate threshold stages. This gradual intensity progression ensured stable thermoregulatory activation, facilitating continuous sweat excretion while delaying abrupt lactate accumulation. The protocol was suitable for this study because sustained sweat availability was essential for analyzing sweat-derived lactate biomarkers throughout dynamic metabolic transitions. The incremental treadmill subsequently commenced, consisting of seven mandatory 5-minute stages, with a 2-minute rests between stages. Speed increased by 0.5 km/h and incline by 1% per stage, as detailed in Figure 4. The test continued until volitional exhaustion, defined inability to maintain the required pace despite verbal encouragement.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBlood Sampling and Analysis.\u0026nbsp;\u003c/strong\u003eA 22 G catheter (Kovax-Cath, Korea) was inserted into the antecubital vein of the non-dominant arm before exercise. Venous blood samples (~3 mL) were collected immediately after each exercise stage. Samples were stored at 4 \u0026deg;C and analyzed colorimetrically for blood lactate concentration at a certified clinical laboratory (Seegene Medical Foundation, Seoul, Korea).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMeasurements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eHeart Rate (HR).\u003c/strong\u003e HR was continuously monitored using a chest-worn heart rate sensor. Resting HR was recorded during a seated rest period before exercise. HR values were recorded 10 s before each stage ended.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSweat Sampling and Analysis.\u003c/strong\u003e\u003cem\u003e\u0026nbsp;\u003c/em\u003eSweat collection patches were applied after the warm-up phase, after skin cleansing with 70% isopropyl alcohol and deionized water to remove debris and oils. Sweat samples were collected continuously using double-film patches placed on three the forehead (midline, 1 cm above the eyebrows), chest (lateral side, near the pectoral muscle), and upper back (most medial side of the scapula), based on an anatomical mapping method\u003csup\u003e15\u003c/sup\u003e. Each patch consisted of an absorbent cotton layer, a parafilm barrier ((Disposable consumable laboratory film, CNWTC, Chongqing, China), and a transparent waterproof dressing (Tegaderm Film, 3M, St. Paul, MN, USA). Cotton sizes were 16.8, 33.6, and 33.6 cm\u0026sup2; for the forehead, chest, and back, respectively, with corresponding parafilm sizes of 19.2, 38.2, and 38.4 cm. Patches were secured during each 5-minute exercise stage and replaced afterward. Only sweat patches producing \u0026ge;2 \u0026micro;L of sweat were analyzed to ensure adequate sample size. Collected sweat samples were immediately weighed using an analytical balance (Hansung HS-103, Hansung Instrument, South Korea; precision: 0.001 g) to determine sweat volume, and subsequently centrifuged at 2,000 rpm for 5 min at 4 \u0026deg;C (LC-8S, JoanLab, China) to separate the supernatant for further analysis. Samples were stored at 4 \u0026deg;C prior to analysis. [La⁻]sw was quantified amperometrically at 0.1 V with a calibrated lactate sensor (Dongwoo Fine-chem Inc.) connected to a CHI-630 electrochemical workstation (CH Instruments, Inc.). The sensor was calibrated daily with standard lactate solutions (0\u0026ndash;20 mM) to ensure measurement accuracy within \u0026plusmn;5%.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCore Body Temperature (CBT).\u0026nbsp;\u003c/strong\u003eCBT was continuously monitored using an ingestible telemetric capsule transmitting data wirelessly to an external receiver (e-Viewer\u0026reg; Performance). Data were recorded throughout the test and recovery period to assess thermoregulatory and metabolic responses.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical Analyses.\u0026nbsp;\u003c/strong\u003eStatistical analyses were performed using Jamovi version 2.3 for Mac (The Jamovi Project, Sydney, Australia). Statistical significance was set at \u0026alpha; = 0.05, data are presented as the mean \u0026plusmn; standard deviation (SD) unless otherwise stated. Linear mixed-effects models (LMMs) were assessed associations between blood lactate concentration (BLa) and non-invasive physiological variables (HR, CBT, [La⁻]ₛw, and LER). BLa data were log-transformed to improve normality, and participant-level random effects accounted for repeated measures across exercise stages. Pearson\u0026rsquo;s correlation analyses additionally assessed linear relationships between BLa and each non-invasive variable. To confirm the robustness of the LMM findings, a supplementary random forest regression was applied using the same input variables (HR, CBT, and sweat-derived lactate variables) to explore the predictive relationships under a non-parametric framework. This analysis served as a secondary reference, suggesting that the variable combination retained predictive relevance beyond linear assumptions.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePerspective\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIncreasing interest in wearable biosensors has renewed the need for valid, non-invasive indicators of metabolic strain during exercise. Our findings show that integrating HR and CBT with sweat-derived lactate indices, particularly forehead LER, enables accurate estimation of log-transformed blood lactate during incremental running. This supports the view that multi-signal monitoring can outperform single-marker approaches when lactate dynamics are influenced by cardiovascular, thermoregulatory, and sweat-gland factors. Practically, these models may facilitate real-time intensity profiling and individualized training prescription without repeated blood sampling. Future studies should validate performance across sex, fitness levels, environments (heat/humidity), and exercise modes, and test device-level implementation with prospective external validation and error thresholds that are meaningful for coaching and clinical decision-making.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by Dongwoo Fine-Chem Co., Ltd. (Republic of Korea).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflicts of Interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was funded by Dongwoo Fine-Chem Co., Ltd. S. H. Chon and Y. K. Lee is an employees of Dongwoo Fine-Chem Co., Ltd. The authors declare no other conflicts of interest (The sponsor had no role in the study design, data analysis, or interpretation).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceptualization: Jaesung Lee, Jihye Moon, Jonghoon Park. Investigation: Jaesung Lee, Jihye Moon, Seunghwan Chon, Youngkeun Lee. Formal analysis: Jaesung Lee, Jihye Moon, Youngim Kim, Hyeonmin Kim, Hyunseob Lee, Sungjin Yoon. Data curation: Jaesung Lee, Jihye Moon. Visualization: Jaesung Lee, Jihye Moon. Writing – original draft: Jaesung Lee, Jihye Moon. Writing – review \u0026amp; editing: Youngim Kim, Hyeonmin Kim, Eunbi Kim, Hyunseob Lee, Sungjin Yoon, Seunghwan Chon, Youngkeun Lee. Supervision: Jonghoon Park. Project administration: Jonghoon Park. Funding acquisition: Jonghoon Park. All authors have read and approved the final version of the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData Availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data supporting the findings of this study, including the raw and processed data underlying the figures and tables, are available from the corresponding author upon reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics Approval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study protocol was approved by the Institutional Review Board of Korea University (KUIRB‑2024‑0280‑01)\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePatient consent\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWritten informed consent was obtained from all participants for study participation and the use of their data for research purposes.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePermission to reproduce\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eClinical trial registration\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research was supported by a fund from Dongwoo Fine-Chem Co., Ltd. in 2024–2025.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eBeneke, R., Leith\u0026auml;user, R. M. \u0026amp; Ochentel, O. Blood lactate diagnostics in exercise testing and training. \u003cem\u003eInt. J. Sports Physiol. Perform.\u003c/em\u003e \u003cb\u003e6\u003c/b\u003e, 8\u0026ndash;24. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1123/ijspp.6.1.8\u003c/span\u003e\u003cspan address=\"10.1123/ijspp.6.1.8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoodwin, M. L., Harris, J. E., Hern\u0026aacute;ndez, A. \u0026amp; Gladden, L. B. Blood lactate measurements and analysis during exercise: A guide for clinicians. \u003cem\u003eJ. Diabetes Sci. Technol.\u003c/em\u003e \u003cb\u003e1\u003c/b\u003e, 558\u0026ndash;569. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/193229680700100414\u003c/span\u003e\u003cspan address=\"10.1177/193229680700100414\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKhabour, O. F., Ali, A., Mahallawi, W. H. \u0026amp; K. H. \u0026amp; Occupational infection and needle stick injury among clinical laboratory workers in Al-Madinah city, Saudi Arabia. \u003cem\u003eJ. Occup. Med. Toxicol.\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e, 15. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12995-018-0198-5\u003c/span\u003e\u003cspan address=\"10.1186/s12995-018-0198-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGao, W. et al. Fully integrated wearable sensor arrays for multiplexed in situ perspiration analysis. \u003cem\u003eNature\u003c/em\u003e \u003cb\u003e529\u003c/b\u003e, 509\u0026ndash;514. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/nature16521\u003c/span\u003e\u003cspan address=\"10.1038/nature16521\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang, G., Ryu, S., Kim, K. \u0026amp; Kim, J. Wearable device for continuous sweat lactate monitoring in sports. \u003cem\u003eFront. Physiol.\u003c/em\u003e \u003cb\u003e15\u003c/b\u003e, 1376801. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fphys.2024.1376801\u003c/span\u003e\u003cspan address=\"10.3389/fphys.2024.1376801\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKarpova, E. V., Laptev, A. I., Andreev, E. A., Karyakina, E. E. \u0026amp; Karyakin, A. A. Relationship between sweat and blood lactate levels during exhaustive physical exercise. \u003cem\u003eChemElectroChem\u003c/em\u003e 7, 191\u0026ndash;194, (2020). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/celc.201901703\u003c/span\u003e\u003cspan address=\"10.1002/celc.201901703\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOkawara, H., Sawada, T., Nakashima, D., Fujitsuka, H. \u0026amp; Katsumata, Y. Lactate threshold evaluation in swimming using a sweat lactate sensor: A prospective study. \u003cem\u003eEur. J. Sport Sci.\u003c/em\u003e \u003cb\u003e24\u003c/b\u003e, 1302\u0026ndash;1312. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/ejsc.12179\u003c/span\u003e\u003cspan address=\"10.1002/ejsc.12179\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTakei, N. et al. Differential patterns of sweat and blood lactate concentration response during incremental exercise in varied ambient temperatures: A pilot study. \u003cem\u003eTemp. (Austin)\u003c/em\u003e. \u003cb\u003e11\u003c/b\u003e, 247\u0026ndash;253. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/23328940.2024.2375693\u003c/span\u003e\u003cspan address=\"10.1080/23328940.2024.2375693\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSeki, Y. et al. A novel device for detecting anaerobic threshold using sweat lactate during exercise. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e, 4929. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-021-84381-9\u003c/span\u003e\u003cspan address=\"10.1038/s41598-021-84381-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBuono, M. J., Lee, N. V. L. \u0026amp; Miller, P. W. The relationship between exercise intensity and the sweat lactate excretion rate. \u003cem\u003eJ. Physiol. Sci.\u003c/em\u003e \u003cb\u003e60\u003c/b\u003e, 103\u0026ndash;107. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s12576-009-0073-8\u003c/span\u003e\u003cspan address=\"10.1007/s12576-009-0073-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTaylor, N. A. \u0026amp; Machado-Moreira, C. A. Regional variations in transepidermal water loss, eccrine sweat gland density, sweat secretion rates and electrolyte composition in resting and exercising humans. \u003cem\u003eExtrem. Physiol. Med.\u003c/em\u003e \u003cb\u003e2\u003c/b\u003e, 4. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/2046-7648-2-4\u003c/span\u003e\u003cspan address=\"10.1186/2046-7648-2-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBentley, D. J., Newell, J. \u0026amp; Bishop, D. Incremental exercise test design and analysis: Implications for performance diagnostics in endurance athletes. \u003cem\u003eSports Med.\u003c/em\u003e \u003cb\u003e37\u003c/b\u003e, 575\u0026ndash;586. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2165/00007256-200737070-00002\u003c/span\u003e\u003cspan address=\"10.2165/00007256-200737070-00002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAchten, J. \u0026amp; Jeukendrup, A. E. Heart rate monitoring: Applications and limitations. \u003cem\u003eSports Med.\u003c/em\u003e \u003cb\u003e33\u003c/b\u003e, 517\u0026ndash;538. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2165/00007256-200333070-00004\u003c/span\u003e\u003cspan address=\"10.2165/00007256-200333070-00004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2003).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFebbraio, M. A. Alterations in energy metabolism during exercise and heat stress. \u003cem\u003eSports Med.\u003c/em\u003e \u003cb\u003e31\u003c/b\u003e, 47\u0026ndash;59. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2165/00007256-200131010-00004\u003c/span\u003e\u003cspan address=\"10.2165/00007256-200131010-00004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2001).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan Hall, G. Lactate kinetics in human tissues at rest and during exercise. \u003cem\u003eActa Physiol.\u003c/em\u003e \u003cb\u003e199\u003c/b\u003e, 499\u0026ndash;508. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1748-1716.2010.02122.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1748-1716.2010.02122.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArmstrong, L. E. et al. American College of Sports Medicine position stand: Exertional heat illness during training and competition. \u003cem\u003eMed. Sci. Sports Exerc.\u003c/em\u003e \u003cb\u003e39\u003c/b\u003e, 556\u0026ndash;572. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1249/MSS.0b013e31802fa199\u003c/span\u003e\u003cspan address=\"10.1249/MSS.0b013e31802fa199\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGerrett, N. et al. Thermal sensitivity to warmth during rest and exercise: A sex comparison. \u003cem\u003eEur. J. Appl. Physiol.\u003c/em\u003e \u003cb\u003e114\u003c/b\u003e, 1451\u0026ndash;1462. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00421-014-2875-0\u003c/span\u003e\u003cspan address=\"10.1007/s00421-014-2875-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVieira, S. S. et al. Does stroke volume increase during an incremental exercise? A systematic review. \u003cem\u003eOpen. Cardiovasc. Med. J.\u003c/em\u003e \u003cb\u003e10\u003c/b\u003e, 57\u0026ndash;63. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2174/1874192401610010057\u003c/span\u003e\u003cspan address=\"10.2174/1874192401610010057\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2016).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSawka, M. N. \u0026amp; Wenger, C. B. \u003cem\u003ePhysiological responses to acute exercise-heat stress. In Human Performance Physiology and Environmental Medicine at Terrestrial Extremes (eds Pandolf, K. B., Sawka, M. N. \u0026amp; Gonzalez, R. R.)\u003c/em\u003e. 97\u0026ndash;151Benchmark Press,. (1988).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShen, T. \u0026amp; Wen, X. Heart rate based prediction of velocity at lactate threshold in ordinary adults. \u003cem\u003eJ. Exerc. Sci. Fit.\u003c/em\u003e \u003cb\u003e17\u003c/b\u003e, 108\u0026ndash;112. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jesf.2019.07.002\u003c/span\u003e\u003cspan address=\"10.1016/j.jesf.2019.07.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePeriard, J. D., Eijsvogels, T. M. H. \u0026amp; Daanen, H. A. M. Exercise under heat stres : Thermoregulation, hydration, performance implications, and mitigation strategies. \u003cem\u003ePhysiol Rev\u003c/em\u003e 101, 1873\u0026ndash; (1979). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1152/physrev.00038.2020\u003c/span\u003e\u003cspan address=\"10.1152/physrev.00038.2020\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLindinger, M. \u003cem\u003eHomeostasis and its disturbance during exercise. In Open Textbook of Exercise Physiology (Open Educational Alberta\u003c/em\u003e,). (2023). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://pressbooks.openeducationalberta.ca/physiologyexercise/\u003c/span\u003e\u003cspan address=\"https://pressbooks.openeducationalberta.ca/physiologyexercise/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLim, J., Park, H., Lee, S. \u0026amp; Park, J. Changes in heart rate, muscle temperature, blood lactate concentration, blood pressure, and fatigue perception following jogging and running: An observational study. \u003cem\u003eExerc. Sci.\u003c/em\u003e \u003cb\u003e31\u003c/b\u003e, 72\u0026ndash;79. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.15857/ksep.2022.00045\u003c/span\u003e\u003cspan address=\"10.15857/ksep.2022.00045\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHeikenfeld, J. et al. Accessing analytes in biofluids for peripheral biochemical monitoring. \u003cem\u003eNat. Biotechnol.\u003c/em\u003e \u003cb\u003e37\u003c/b\u003e, 407\u0026ndash;419. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41587-019-0040-3\u003c/span\u003e\u003cspan address=\"10.1038/s41587-019-0040-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRabost-Garcia, G. et al. Non-Invasive Multiparametric Approach To Determine Sweat-Blood Lactate Bioequivalence. \u003cem\u003eACS Sens.\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e, 1536\u0026ndash;1541. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1021/acssensors.2c02614\u003c/span\u003e\u003cspan address=\"10.1021/acssensors.2c02614\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBates, D., M\u0026auml;chler, M., Bolker, B. \u0026amp; Walker, S. Fitting Linear Mixed-Effects Models Using lme4. \u003cem\u003eJ. Stat. Softw.\u003c/em\u003e \u003cb\u003e67\u003c/b\u003e, 1\u0026ndash;48. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.18637/jss.v067.i01\u003c/span\u003e\u003cspan address=\"10.18637/jss.v067.i01\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorris, N. B., Cramer, M. N., Hodder, S. G., Havenith, G. \u0026amp; Jay, O. A comparison between the technical absorbent and ventilated capsule methods for measuring local sweat rate. \u003cem\u003eJ. Appl. Physiol. (1985)\u003c/em\u003e. \u003cb\u003e114\u003c/b\u003e, 816\u0026ndash;823. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1152/japplphysiol.01088.2012\u003c/span\u003e\u003cspan address=\"10.1152/japplphysiol.01088.2012\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2013).\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":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Blood lactate, heart rate, core body temperature, sweat lactate, non-invasive monitoring, wearable biosensor","lastPublishedDoi":"10.21203/rs.3.rs-8498078/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8498078/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eBlood lactate concentration (BLa) is a key marker of metabolic stress, but invasive sampling limits real-time monitoring. We developed a non-invasive model to estimate BLa during incremental exercise using heart rate (HR), core body temperature (CBT), and sweat-derived indices. Thirty-one healthy adult males performed a graded treadmill test. HR and CBT were monitored continuously. Sweat was sampled from the forehead, chest, and back to quantify sweat lactate concentration ([La\u0026minus;]sw) and lactate excretion rate (LER = [La\u0026minus;]sw \u0026times; sweat rate). Linear mixed-effects models (LMMs) were fitted with log-transformed BLa (Log[BLa]) and participant-level random effects. BLa increased with exercise intensity (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), accompanied by increases in HR, CBT and LER (both p\u0026thinsp;\u0026lt;\u0026thinsp;0.001). LMMs combining HR, CBT, and sweat indices showed strong performance for Log[BLa]. The best model (HR\u0026thinsp;+\u0026thinsp;CBT\u0026thinsp;+\u0026thinsp;forehead LER) achieved conditional R\u0026sup2;=0.939 and RMSE\u0026thinsp;=\u0026thinsp;0.229 (log units), and forehead-based models outperformed chest and back. Combined cardiovascular, thermoregulatory, and sweat-derived measures enable accurate, non-invasive estimation of BLa during graded exercise, supporting wearable-based metabolic monitoring and individualized exercise prescription.\u003c/p\u003e","manuscriptTitle":"Non-Invasive Prediction of Blood Lactate During Incremental Exercise via Heart Rate, Core Body Temperature, and Sweat-Derived Indices","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-04 10:18:10","doi":"10.21203/rs.3.rs-8498078/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-03-06T05:48:53+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-05T18:58:28+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-22T00:07:01+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"302238007465999588706164622081363667985","date":"2026-02-13T17:32:25+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"328043411912555525976344750425641071517","date":"2026-02-10T08:44:09+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-02-03T02:00:01+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-30T17:11:47+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-01-06T11:45:13+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-05T09:56:44+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-01-05T09:39:31+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"771207a0-2490-4c31-8806-a50ee352120d","owner":[],"postedDate":"February 4th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":62209028,"name":"Health sciences/Biomarkers"},{"id":62209029,"name":"Health sciences/Cardiology"},{"id":62209030,"name":"Health sciences/Health care"},{"id":62209031,"name":"Health sciences/Medical research"},{"id":62209032,"name":"Biological sciences/Physiology"}],"tags":[],"updatedAt":"2026-04-13T16:01:06+00:00","versionOfRecord":{"articleIdentity":"rs-8498078","link":"https://doi.org/10.1038/s41598-026-47148-8","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2026-04-07 15:58:16","publishedOnDateReadable":"April 7th, 2026"},"versionCreatedAt":"2026-02-04 10:18:10","video":"","vorDoi":"10.1038/s41598-026-47148-8","vorDoiUrl":"https://doi.org/10.1038/s41598-026-47148-8","workflowStages":[]},"version":"v1","identity":"rs-8498078","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8498078","identity":"rs-8498078","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}