MRI-Based Spectral Analysis of Fetal Brain Gyrification: Applications to Lissencephaly and Polymicrogyria

MRI-Based Spectral Analysis of Fetal Brain Gyrification: Applications to Lissencephaly and Polymicrogyria | 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 MRI-Based Spectral Analysis of Fetal Brain Gyrification: Applications to Lissencephaly and Polymicrogyria Bossmat Yehuda, Rotem Gal, Yair Wexler, Aviad Rabinowich, Elka Miller, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8011139/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 20 Feb, 2026 Read the published version in Scientific Reports → Version 1 posted 13 You are reading this latest preprint version Abstract Cortical gyrification is a key marker of fetal brain development and is typically assessed qualitatively on ultrasound or MRI. While previous quantitative approaches have characterized gestational trajectories in typically developing (TD) fetuses, only a few studies have investigated cortical malformations such as lissencephaly and polymicrogyria. Spectral analysis, which characterizes signals by their frequency content, has been successfully applied to study gyrification in neonates and adults but has not yet been explored prenatally. In this study, we introduce a spectral framework for quantifying fetal cortical folding from routine fetal MRI. Cerebral contours were extracted from coronal slices, transformed into polar coordinates, and analyzed using Fourier Transform to derive spectral profiles and five gyrification features: non-zero spectral density, entropy, mean frequency, variance, and skewness and the first twelve frequencies. Seventy-three TD fetuses and twenty-four with malformations of cortical development (14 polymicrogyria, 10 lissencephaly) were evaluated across gestation. Differences between TD, lissencephaly, or polymicrogyria fetuses were evaluated using linear mixed models and post-hoc t-tests with Benjamini–Hochberg correction. In TD fetuses, spectral features showed gestational-age–related trajectories, with increasing spectral density and variance and decreasing skewness, corresponding to the sequential folding waves. Fetuses with cortical malformations had lower spectral density and entropy ( p ≤ 0.031), and reduction in most of the twelve frequencies, most prominently in frequencies associated with the Sylvian fissure development ( p < 0.001). Spectral representation may capture both global and local aspects of cortical folding, offering a robust and quantitative biomarker of fetal brain maturation and deviations in cortical development. Biological sciences/Biological techniques Health sciences/Biomarkers Health sciences/Medical research Biological sciences/Neuroscience Cortex Gyrification Fetal MRI Spectral analysis Lissencephaly Polymicrogyria Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Introduction The human’s brain cortical development begins prenatally and involves substantial changes in shape, size, tissue organization, and maturation. Gyrification, the formation of gyri and sulci, emerges during mid-gestation, accelerates throughout the third trimester, and continues after birth.[ 1 ] Gyrification follows an organized spatiotemporal pattern, and progresses in three waves: the primary wave spanning roughly 20–30 weeks (peak ~ 24), the secondary wave across 30–36 weeks (peak ~ 32), and the tertiary wave starts at approximately 38 weeks and extends into early postnatal life. Overall, gyrification shows a strong gestational age (GA)-dependent trajectory and is a key marker of brain development.[ 2 ] Malformations of cortical development (MCDs) may result from a range of etiologies, including genetic mutations, intrauterine infections, toxin exposure or ischemia,[ 3 , 4 ] and typically become detectable after 24 weeks of gestation.[ 5 ] Lissencephaly and polymicrogyria are two types of MCDs associated with a broad spectrum of neurodevelopmental outcomes, including severe epilepsy and motor impairments,[ 6 ] depending on their type, severity, and anatomical extent.[ 2 ] Lissencephaly is characterized by a smooth cortical surface and a thickened, disorganized cortex. In contrast, polymicrogyria is characterized by excessive and irregular folding with thin, fused cortical layers, which may occur focally, diffusely, unilaterally, or bilaterally.[ 7 – 9 ] While typically, the assessment of fetal cortical development is based on qualitative visual inspection of gyrification patterns using ultrasound (US) or magnetic resonance imaging (MRI),[ 10 , 11 ] previous studies used the gyrification index (GI), defined as the ratio between the cortical contour to its convex hull,[ 12 , 13 ] along with additional methods to quantify cortical folding patterns.[ 14 – 16 ] A few studies have developed methods for fetal gyrification analysis based on 3D-reconstructed MRI,[ 17 – 21 ] which is limited by long reconstruction times and specific acquisition requirements. Only one study quantified gyrification based on 2D data using GI, demonstrating GA- dependent changes.[ 22 ] Only two studies to date have quantitatively analyzed gyrification in fetuses with MCDs. One compared three fetuses with polymicrogyria to 14 typically developing (TD) fetuses,[ 21 ] reporting diminished alignment to fetal brain templates among cases with polymicrogyria, but no significant differences using 3D GI. The second study that employed 2D GI quantification, detected differences between TD fetuses and fetuses with lissencephaly and polymicrogyria.[ 22 ] This method was further applied on fetuses with growth restriction, demonstrating reduced gyrification patterns. [ 23 ] While this method was based on a relatively large cohort and successfully identified lissencephaly abnormalities, it underperformed in early gestation (< 28 weeks) and in distinguishing polymicrogyria from TD fetuses, highlighting the need for alternative methodologies. Spectral analysis, a foundational tool in signal processing, has been widely applied across scientific domains to analyze patterns based on frequency content. The Fourier transform (FT) decomposes signals into their frequency components, revealing underlying oscillations and their amplitudes.[ 24 ] Previous spectral methods applications were used to characterize brain gyrification among adults and neonates.[ 25 , 26 ] Collectively, these studies demonstrated that lower frequencies are associated with global brain shape and higher frequencies are associated with local folding patterns. The developing fetal brain is subjected to profound regional and global changes throughout gestation and some MCDs may only be manifested locally, making spectral analysis a potential approach for capturing subtle, spatially confined deviations in cortical folding patterns that might be overlooked by conventional metrics. To our knowledge, spectral analysis has not been previously applied to prenatal gyrification. The aim of this study was to assess fetal gyrification using spectral analysis of fetal MRI, characterize GA-related patterns in TD fetuses, and compare TD fetuses with fetuses with lissencephaly and polymicrogyria. Results A total of 150 participants with singleton pregnancies were included. Eleven cases were excluded due to poor image quality, 29 due to genetic abnormalities or MRI findings other than lissencephaly and polymicrogyria, and 13 with GA below 24 weeks. Supplementary Tables 1 and 2 further details the demographic and MRI indication referrals. The final cohort comprised of 97 fetuses: 73 TD (mean GA at MRI 31.3±2.7 weeks; range, 25-37.6 weeks), 14 with polymicrogyria (mean GA, 31.6±2.5 weeks; range, 27.6–36 weeks), and ten with lissencephaly (mean GA, 29.0±3.0 weeks; range, 25–33, weeks). No significant differences in GA at the time of MRI were observed between TD and MCD fetuses. Spectral feature trajectories of TD controls demonstrated increased gyrification across gestation, except for skewness which decreased with GA. These changes are illustrated in Fig. 1 . Supplementary Fig. 1 highlights the correlation between the dominant frequency and the complexity degree of the contour. The spatial frequencies of TD controls were characterized by two distinct developmental patterns across gestation. Mid-frequency band (frequencies 6,7,9, and 10) showed a progressive increase with gestation, representing global gyrification. In contrast, the lower frequency band (frequencies 1,3,4,5) exhibited accelerated growth between 24 to 32 weeks, followed by a plateau from 32 weeks onwards. The highest frequency band (frequencies 11 and 12) was characterized by accelerated growth beginning around 31 weeks. Notably, frequency 2 decreased along gestation. Changes in spatial frequencies with gestation are further illustrated in Fig. 2 . Symmetry analysis demonstrated homogeneous dispersion among TD, lissencephaly and polymicrogyria cases fetuses and in all features. Representative cases of TD controls, lissencephaly and polymicrogyria are shown in Fig. 3 . Fetuses with MCD demonstrated significant differences in gyrification compared with TD fetuses. Lissencephaly cases exhibited an 11.5% reduction in non-zero spectral density, a 2.7% reduction in entropy, a 4.4% reduction in variance, and a 7% increase in skewness relative to TD fetuses ( p < 0.001). Additionally, apart from frequencies 2 ( p = 0.17) and 9 ( p = 0.06), all other frequencies were significantly lower in lissencephaly compared with TD controls ( p ≤ 0.015). Compared to TD controls, polymicrogyria cases showed a 7.8% reduction in non-zero spectral density ( p < 0.001), and a 1.6% reduction in entropy ( p = 0.03), with no significant differences in spectral variance ( p = 0.17) or skewness ( p = 0.12). Apart from frequencies 2, 9 and 11 ( p = 0.099, 0.547 and 0.19, respectively) all other spatial frequencies were significantly reduced in polymicrogyria compared with TD controls ( p < 0.05). Results of the comparison between TD, lissencephaly and polymicrogyria are summarized in Table 2 and illustrated in Figs. 3 and 4 . Table 2 Gyrification spectral features changes in percentages for lissencephaly and polymicrogyria cases, compared with TD controls. P-values are presented after adjustment with Benjamini-Hochberg. Abbreviations: TD = typically developing. Feature Lissencephaly [%] P-value Polymicrogyria [%] P-value Non-zero Spectral Density -11.5 < 0.001 -7.8 < 0.001 Entropy -2.7 < 0.001 -1.6 0.031 1st Moment - Mean -0.7 0.34 -0.7 0.34 2nd Moment - Variance -4.4 < 0.001 -3.1 0.17 3rd Moment - Skewness 7 < 0.001 4 0.12 Frequency 1 -29.4 < 0.001 -19.1 < 0.001 Frequency 2 -3.3 0.099 -0.6 0.099 Frequency 3 -10.8 0.016 -5.8 0.015 Frequency 4 -2.9 < 0.001 -7.4 0.004 Frequency 5 -19.3 < 0.001 -12.1 0.011 Frequency 6 -16.4 < 0.001 -8.0 0.008 Frequency 7 -14.1 < 0.001 -11.5 0.016 Frequency 8 -18.4 < 0.001 -13.1 0.042 Frequency 9 -18.9 0.062 -6.0 0.547 Frequency 10 -16.6 < 0.001 -12.3 0.003 Frequency 11 -6.8 < 0.001 -8.5 0.19 Frequency 12 -15.1 < 0.001 -13.9 0.03 Discussion This study presents the application of spectral analysis to fetal brain gyrification quantification from 2D-MRI in TD fetuses and those with lissencephally and polymicrogyryia. Five spectral features and spatial profiles of the first twelve frequencies were extracted. TD fetuses exhibited increasing values across all frequencies and features with advancing GA, showing growth characteristics corresponding to the first two gyrification waves. Fetuses with MCDs demonstrated decreased gyrification in most gyrification-based spectral frequencies, compared with TD subjects, with more pronounced differences among cases with lissencephaly. Importantly, the spectral representation also detected distinct patterns in polymicrogyria, despite the heterogenous imaging characteristics of this malformation enabling developmental stage characterization. Cortical folding begins in second trimester and involves cell proliferation, neuronal migration, and cortical organization. [ 1 ] Among fetuses with TD, there was and increase in mean frequency and variance along gestation, reflecting spectral range widening as the cortex develops into a more geometrically complex structure.[ 27 ] Skewness, which quantifies the asymmetry of the spectral distribution relative to its mean frequency,[ 28 ] decreased with gestation. Taken together, these indicate a growing dominance of higher frequencies as pregnancy progresses, highlighting how the spectral features capture global aspects of gyrification during fetal development. Per specific frequencies, distinct patterns of growth along gestation corresponding to the waves of cortical folding were demonstrated. The low frequency band magnitude rapidly increased between the 25 and 32 weeks of gestation, followed by a plateau, suggesting an association with the primary folding wave. These findings generally represent the global brain gyrification and are consistent with previous studies performed in preterm infants.[ 25 , 26 ] In addition, higher frequencies exhibited an accelerated increase from 32 weeks onwards, consistent with the onset of the secondary folding wave and local gyrification development. As higher frequencies correspond to more complex contour features with shorter distances between oscillations, this increase is likely to reflect the emergence of more intricate sulcal patterns. Conversely, frequency 2 decreased steadily with gestation, likely representing the gradual reduction of the smooth elliptical shape of cerebral contour as it evolves to a more complex structure. Moreover, the Sylvian fissure is the principal contributor to the bi-lobar configuration, suggesting a correlation between the emergence of this primary fissure and lower freqeuncies. Supplementary Fig. 2 illustrates the association between hemispheric contour with and without the Sylvian fissure and frequency 1, which demonstrated the highest significant difference compared to MCDs. Spectral features and the overall spectral profile were symmetrical between the two hemispheres, consistent with prior studies of cortical gyrification.[ 17 , 18 , 20 , 25 , 29 ] Although regional asymmetries have been described in specific anatomical locations,[ 1 , 30 , 31 ] the global nature of this approach precludes detection of such localized variations. Early and accurate diagnosis MCD is crucial, as patients often present within the first year of life with hypotonia, global developmental delay, and seizures, even when no other major brain anomalies are detected. The severity of neurological outcomes is known to correlate with the type and extent of abnormalities observed on MRI.[ 32 ] A small number of studies previously assessed quantitative differences between gyrification in fetuses with MCD compared with TD,[ 17 , 21 – 23 ] demonstrated reduced gyrification among lissencephaly cases. Polymicrogyria cases show some defferences compared with TD, yet its local and heterogenous appearance reduces the ability to reliably separate TD from polymicrogyria.[ 33 ]. These findings are consistent with our results, showing underdeveloped gyrification and reduced spectral dispersion in lissencephaly. Notably, non-zero spectral density values were significantly lower in lissencephaly cases prior to 30 weeks, suggesting that our method is sensitive to early lissencephaly-related gyrification patterns, a stage where the fetal brain surface is typically smooth and may resemble a lissencephalic pattern. Contrary to our hypothesis, the mean frequency (first moment) did not differ between groups, which likely reflects subtle spectral shifts that are not captured by an averaged central measure. Lissencephaly cases did, however, show significantly higher skewness and lower variance values, suggesting that spectral energy was disproportionately concentrated in lower frequencies. The most pronounced decrease was observed in frequency 1 which is most likely linked to the Sylvian fissure and is included in the representing band for the first gyrification wave. This observation quantitatively supports the known malformation of the Sylvian fissure in lissencephaly,[ 34 , 35 ] which disrupts the development of the Sylvian fissure during the first wave of gyrification. The three spectral moments were not different among fetuses with polymicrogyria compared with TD controls. However, non-zero density and entropy were significantly lower among polymicrogyria subjects, suggesting that spectral measures may be sensitive to subtle abnormalities that are characteristic of polymicrogyria. In addition, high-frequency components demonstrated lower values, which may indicate the presence of shallow, irregular sulci which are hallmark feature of polymicrogyria. Since sulcal depth is encoded as the distance from the cortical contour to the brain’s center of mass, shallow sulci correspond to lower intensities at their respective frequencies. Moreover, polymicrogyria cases demonstrated decreased values in the low frequency range, particularly frequency 1, which is directrly linked to the development of the Sylvian fissure. Taken together, these spectral differences between polymicrogyria and TD fetuses indicate measurable alterations in gyrification patterns, suggesting the potential use of this method for assessing the developmental stage characterization and the extent of MCDs. Collectively, our findings support the potential utility of spectral analysis for quantitative assessment of gyrification. Specifically, our results suggest an advantage of this approach over previous methods in detecting abnormalities associated with polymicrogyria. Future studies should aim to localize regions of abnormality and delineate the extent of malformation to enhance model interpretability for clinicians. In addition, the relationship between spectral measures and neurodevelopmental outcomes merits further investigation. Several limitations should be acknowledged. While our cohort is relatively large compared with other studies in the field, the number of subjects with MCD is limited. Our findings should be validated in larger MCD cohorts to assess the correlation gyrification-based frequencies and neurological outcomes, along with genetic findings. Moreover, the cohort was retrospectively collected, and some demographic information was not available. In addition, in this study, manual corrections were applied over automatic deep leaning based segmentation to ensure segmentation accuracy. Future development should include improving segmentation accuracy and explore super resolution reconstruction. Conclusions This study introduces a quantitative spectral framework of fetal brain gyrification from MRI studies. Spectral-based features changed across gestation aligning with developmental trajectories and expected patterns of gyrification waves, and revealed significant differences between fetuses with MCDs and TD controls, including a marked reduction in total gyrification and frequencies linked to the Sylvian fissure development. Our findings suggest that the spectral representation of cortical folding may serve as a biomarker for brain maturation, and distinguish fetuses with lissencephaly and polymicrogyria from TD subjects. This method may also be extended to study the gyrification patterns of other prenatal conditions with delayed cortical maturation such as fetal growth restriction and congenital cytomegalovirus infection. Materials and Methods Population The fetal brain MRI were collected at two clinical sites between 2007 to 2022, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel and the Children's Hospital of Eastern Ontario (CHEO), Ottawa, Canada. Data collection and analysis were performed in accordance with relevant guidelines and regulations. TD controls included singleton pregnancies referred for MRI due to various sonographic suspicious abnormalities, such as ventriculomegaly, microcephaly, and club foot, or a previous pregnancy with abnormal development. Exclusion criteria included MRI detected structural brain anomalies, chromosomal abnormalities, amniocentesis-confirmed cytomegalovirus infection, chronic maternal disease, or poor image quality. Fetuses scanned prior to the 24th gestation week were excluded, as gyrification is not yet reliably visible. MCD cases were defined as fetuses diagnosed with lissencephaly or polymicrogyria, based on expert interpretation of both US and MRI. All scans were evaluated by expert fetal neuroradiologists (LBS or EM), each with more than 20 years of experience. The study was approved by the Tel Aviv Sourasky Medical Center and the Children's Hospital of Eastern Ontario (CHEO Institutional Review Boards. The informed consent was waived by the Institutional Review Boards due to the retrospective nature of the study. MRI data The MRI data included scans acquired using two MRI field strengths,3T and 1.5T, from two vendors (GE Healthcare and Siemens Healthineers). The protocol included brain coronal sequences (Table 1 ) using either Fast Recovery Fast Spin Echo (FRFSE), Fast Imaging Employing Steady-State Acquisition (FIESTA), Half-Fourier Single-Shot Turbo Spin-Echo (HASTE) or True Fast Imaging with Steady-State Free Precession (TruFISP). Table 1 Imaging parameters of MRI sequences. Abbreviations: FRFSE = fast-recovery fast spin-echo; FIESTA- fast imaging employing steady-state acquisition; HASTE- half Fourier single-shot turbo spin-echo; TruFISP- true fast imaging with steady-state free precession. Vendor System (magnetic field) Sequence Echo time [milisec] Repetition time [milisec] In-plane resolution [mm] Slice thickness [mm] GE Healthcare DISCOVERY MR450 (1.5T) FRFSE 119–124 7469–11421 0.7 4.0–5.8 SIGNA (1.5T) FIESTA 1.7–1.8 3.8–4 0.6–0.7 4–6 Signa HDxt (1.5T) FIESTA 1.6–1.7 3.8–3.9 1.4 3.7–5.2 SIEMENS Healthineers Aera (1.5T) HASTE 94 1200 1.0 – 1.2 3–4 Prisma (3T) HASTE 86–109 2000 0.4 2–5 Skyra (3T) HASTE 75–93 1711– 2229 0.4–0.7 2.5–4 TruFISP 2.5 4.9 0.4 3 Image analysis The image processing pipeline (Fig. 1 ) included two main stages: Hemispheric contour extraction (Fig. 5 A) This step follows a previously described methodology,[ 22 ] summarized as follows: (1.1) The right and left cerebral hemispheres were automatically segmented using an in-house developed automatic tool and manually refined as needed (BY, 6 years of experience in image segmentation), using ITK-SNAP (version 3.8)[ 36 ] while blinded to the fetal condition. (1.2) Cerebral contour definition: The contour was defined as the boundary between the cerebral hemispheres and the extra-axial cerebrospinal fluid, based on the segmentation in 1.1. Spectral contour analysis was performed per slice for each hemisphere separately, based on the cerebral contour: (2.1) Polar transformation of the contour (Fig. 5 B): Each contour point was represented by its index along the contour (x-axis) and its distance from the convex center of mass, normalized by the maximal distance (y-axis). (2.2) Discrete FT of the polar representation (Fig. 5 C): The FT was applied, with a sampling size defined as the inverse of the contour length, yielding natural frequency values corresponding to the number of oscillations per full contour traversal. For example, a frequency of 3 corresponds to an oscillation repeating three times along the contour. (2.3) Normalization of the spectral array for each slice (Fig. 5 D): Each spectral vector was scaled by the sum of all frequency amplitudes in the corresponding slice, producing a size-independent spectral probability distribution. A predefined noise threshold was applied to remove artifacts, set at the cumulative sum of spectral energy equal to 95%, yielding a spectral representation of frequencies 1 through 12. (2.4) Spectral profile computation (Fig. 5 E): The final spectral representation was obtained by averaging the normalized spectra of all slices into a single vector containing the amplitudes of the natural frequencies. Gyrification feature extraction The spectral profile extracted from step 2.4, which included the first 12 individual frequencies (p i ), was analyzed across GA and between groups. The twelve frequencies were categorized as low (frequencies 1–5), middle (frequencies 6–10), and high (frequencies 11–12) frequency bands, according to the frequency bands described.[ 25 ] In addition, five features were extracted per fetus for each hemisphere (see Eq. 1–5): Non-zero spectral density: (Eq. 1) Represents total gyrification by capturing the intensity of all non-zero frequencies. Frequency zero corresponds to the mean of the polar-transformed contour, reflecting the average hemisphere radius. Spectral entropy: (Eq. 2) Describes the distribution of the spectral power across frequencies Mean frequency: )Eq. 3) The first moment of the spectrum, reflecting its central tendency. Variance: (Eq. 4) The second moment, relative to the first moment, quantifying spectral spread. Skewness: (Eq. 5) The third moment, relative to the first moment, capturing the asymmetry of the frequency distribution. Eq. 1 \(\:non-zero\:spectral\:density=\:\sum\:_{f=1}^{N}{p}_{f}\) Eq. 2 \(\:Spectral\:Entropy=\:-\sum\:_{f=1}^{N}{p}_{f}\bullet\:{log}_{2}\left({p}_{f}\right)\:\) Eq. 3 \(\:Mean=\:\frac{\sum\:_{f=1}^{N}(f\bullet\:{p}_{f})}{\sum\:_{f=1}^{N}{p}_{f}}\) Eq. 4 \(\:Variance=\:\frac{\sum\:_{f=1}^{N}{\left(\right(f-Mean)}^{2}\bullet\:{p}_{f})}{\sum\:_{f=1}^{N}{p}_{f}}\) Eq. 5 \(\:Skewness=\:\frac{\sum\:_{f=1}^{N}{\left(\right(f-Mean)}^{3}\bullet\:{p}_{f})}{{\left(\sqrt{Variance}\right)}^{3}\sum\:_{f=1}^{N}{p}_{f}}\) Where N is the maximal frequency, and p f is the spectral amplitude of frequency f . Symmetry evaluation was performed for each frequency and for each of the five features using the following equation, with L and R representing the left and right hemispheres, respectively: Eq. 6 \(\:Symmetry=\:\frac{L-R}{L+R\:}\) Statistical analysis Statistical analysis was performed using a designated R script, version 022.07.2. Changes along GA in the TD group were assessed for all features and for all frequencies by using generalized additive models for location, scale and shape (GAMLSS) with box-cox power exponential distribution, as recommended by the World Health Organization.[ 37 ] All features and individual frequencies were normalized due to significant correlation with GA, with absolute Pearson r value ranging from 0.27 to 0.82. Briefly, to remove the influence of GA, each feature X was normalized using a power-law transformation: X’ = X/GA λ , where λ minimized the Pearson correlation X’ and GA. The exponent λ varied across measures, ranging between 0.42 for entropy and 2.40 for p 9 . For negatively correlated measures ( skewness and p 2 ), λ was negative (-1.33 and − 0.95, respectively). After normalization, the absolute correlation was reduced to ≤ 0.07. Importantly, λ was determined only from the TD controls. Log transform was applied to GA-normalized features when necessary to achieve normality and homogeneity of variances. As gyrification was estimated for each subject in the right and left hemispheres separately and no asymmetry was detected, the hemisphere and the subject ID were used as random variables in the linear mixed model. Comparison between TD, lissencephaly and polymicrogyria was done using all 17 normalized spectral frequencies and features. Each frequency or feature of gyrification was compared between MCD group and TD using a linear mixed model, with pathology, GA, and their interaction included as fixed effects. Although normalization eliminated typical GA effects, GA and its interaction with pathology were retained to assess pathology-specific GA effects on gyrification. Post-hoc analysis was performed using t-tests with degrees of freedom estimated by the Kenward-Roger method and only included comparisons to the TD group. All p -values of model effects and post hoc tests for all measures were grouped and corrected for multiple comparisons using the Benjamini-Hochberg procedure,[ 38 ] controlling false discovery rate at the 0.05 level. Throughout the study, p -values < 0.05 were considered statistically significant. Declarations Funding statement: This research was supported by the Israel Innovation Authority; Yoran Institute of Human Genome Research; and March of Dimes. Author Contribution B.Y. conceived and designed the study, developed the image processing pipeline, performed data curation, analysis, and visualization, and wrote the original manuscript draft. R.G. contributed to the methodological design. Y.W. conducted the statistical analyses. A.B. contributed to validation and to manuscript review and editing. D.B. supervised the project and contributed to the study design and critical manuscript revision. E.M. and L.B.S. reviewed the MRI scans and provided clinical interpretation. All authors read and approved the final version of the manuscript. Acknowledgement We would like to thank the participants of this study and the MRI radiographers for scanning the participants. We wish good health to all study participants and their newborns.We would like to sincerely thank Mrs. Cassandra Kapoor for her valuable assistance in managing the data from CHEO.This research was supported by Kamin grants (63418, 72126) from the Israel Innovation Authority, the Yoran Institute of Human Genome Research, and March of Dimes. Data Availability De-identified data supporting the findings reported in this work can be made available by the corresponding author upon reasonable request by researchers who meet the criteria for access to confidential data. References Dubois, J. & Dehaene-Lambertz, G. Fetal and postnatal development of the cortex: MRI and genetics. Brain Mapping: Encyclopedic Ref. 2 , 11–19 (2015). Schmitt, S. et al. European Archives of Psychiatry and Clinical Neuroscience Associations of gestational age with gyrification and neurocognition in healthy adults. Eur Arch. Psychiatry Clin. Neurosci 1 , 3 . Guerrini, R. & Dobyns, W. B. Malformations of cortical development: clinical features and genetic causes. Lancet Neurol. 13 , 710–726 (2014). Klein, B., Levin, B. E., Duchowny, M. S. & Llabre, M. M. Cognitive outcome of children with epilepsy and malformations of cortical development. Neurology 55 , 230–235 (2000). Pogliani, L., Andrea Righini, T., Zirpoli, S. & F. & Early Prenatal MR Imaging Diagnosis of Polymicrogyria. AJNR Am. J. Neuroradiol. 25 , 343–346 (2023). Barkovich, A. J., Guerrini, R., Kuzniecky, R. I., Jackson, G. D. & Dobyns, W. B. A developmental and genetic classification for malformations of cortical development: update 2012. Brain 135 , 1348–1369 (2012). Garel, C. MRI of the Fetal Brain Normal Development and Cerebral Pathologies (Springer-, 2004). Barkovich, A. J. MRI analysis of sulcation morphology in polymicrogyria. Epilesia 17–22. 10.1111/j.1528-1167.2009.02436.x (2010). Di Donato, N. et al. Lissencephaly: Expanded imaging and clinical classification. Am. J. Med. Genet. A . 173 , 1473–1488 (2017). Cassart, M. & Garel, C. European overview of current practice of fetal imaging by pediatric radiologists: a new task force is launched. Pediatr. Radiol. https://doi.org/10.1007/s00247-020-04710-4 (2020). Garel, C. Fetal cerebral biometry: Normal parenchymal findings and ventricular size. European Radiology vol. 15 809–813 Preprint at (2005). https://doi.org/10.1007/s00330-004-2610-z Luders, E. et al. Mapping the Relationship between Cortical Convolution and Intelligence: Effects of Gender. Cereb. Cortex September . 18 , 2019–2026 (2008). Duvernoy, C. S. et al. Gender differences in adverse outcomes after contemporary percutaneous coronary intervention: An analysis from the Blue Cross Blue Shield of Michigan Cardiovascular Consortium (BMC2) percutaneous coronary intervention registry. Am. Heart J. 159 , 677–683e1 (2010). Lohmann, G., von Cramon, D. Y. & Colchester, A. C. F. Deep sulcal landmarks provide an organizing framework for human cortical folding. Cereb. Cortex . 18 , 1415–1420 (2008). Kiselev, V. G., Hahn, K. R. & Auer D. P. Is the brain cortex a fractal? Neuroimage 20, 1765–74 (2003). Chung, M. K., Dalton, K. M., Shen, L., Evans, A. C. & Davidson, R. J. Weighted fourier series representation and its application to quantifying the amount of gray matter. IEEE Trans. Med. Imaging . 26 , 566–581 (2007). Rajagopalan, V. et al. Quantitative in vivo MRI measurement of cortical development in the fetus. Brain Struct. Funct. 217 , 127–139 (2012). Rajagopalan, V. et al. Local tissue growth patterns underlying normal fetal human brain gyrification quantified in utero. J. Neurosci. 31 , 2878–2887 (2011). Wright, R. et al. Automatic quantification of normal cortical folding patterns from fetal brain MRI. Neuroimage 91 , 21–32 (2014). Habas, P. A. et al. Early folding patterns and asymmetries of the normal human brain detected from in utero MRI. Cereb. Cortex . 22 , 13–25 (2012). Im, K. et al. Quantitative folding pattern analysis of early primary sulci in human fetuses with brain abnormalities. Am. J. Neuroradiol. 38 , 1449–1455 (2017). Yehuda, B. et al. Automatic Quantification of Normal Brain Gyrification Patterns and Changes in Fetuses with Polymicrogyria and Lissencephaly Based on MRI. AJNR Am. J. Neuroradiol. 44 , 1432–1439 (2023). Yehuda, B. et al. Reduced gyrification in fetal growth restriction with prenatal magnetic resonance images. Cereb Cortex 34 , (2024). Wong, M. Discrete Fourier Analysis Vol. 5 (Springer Science & Business Media, 2011). Dubois, J. et al. The dynamics of cortical folding waves and prematurity-related deviations revealed by spatial and spectral analysis of gyrification. Neuroimage 185 , 934–946 (2019). Germanaud, D. et al. Larger is twistier: Spectral analysis of gyrification (SPANGY) applied to adult brain size polymorphism. Neuroimage 63 , 1257–1272 (2012). Hinich, M. J. & Clay, C. S. The application of the discrete Fourier transform in the estimation of power spectra, coherence, and bispectra of geophysical data. Reviews of Geophysics vol. 6 347–363 Preprint at (1968). https://doi.org/10.1029/RG006i003p00347 Hamed, H. A., Abdullah, A. K. & Al-Waisawy, S. Frequency Hopping Spread Spectrum Recognition Based on Discrete Fourier Transform and Skewness and Kurtosis . International J. Appl. Eng. Research 13 (2018). http://www.ripublication.com Chen, H. et al. Gyral net: A new representation of cortical folding organization. Med. Image Anal. 42 , 14–25 (2017). Garel, C. et al. Fetal Cerebral Cortex: Normal Gestational Landmarks Identified Using Prenatal MR Imaging. AJNR Am. J. Neuroradiol. 22 , 184–189 (2001). Dubois, J. et al. Primary cortical folding in the human newborn: An early marker of later functional development. Brain 131 , 2028–2041 (2008). Leventer, R. J. et al. Clinical and imaging features of cortical malformations in childhood. Neurology 53 , 715–722 (1999). Leventer, R. J. et al. Clinical and imaging heterogeneity of polymicrogyria: a study of 328 patients. Brain 133 , 1415–1427 (2010). Poon, L. C. et al. Transvaginal three-dimensional ultrasound assessment of Sylvian fissures at 18–30 weeks’ gestation. Ultrasound Obstet. Gynecol. 54 , 190–198 (2019). Guibaud, L. et al. Abnormal Sylvian fissure on prenatal cerebral imaging: significance and correlation with neuropathological and postnatal data. Ultrasound Obstet. Gynecol. 32 , 50–60 (2008). Yushkevich, P. A. et al. User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability. Neuroimage 31 , 1116–1128 (2006). Borghi, E. et al. Construction of the World Health Organization child growth standards: Selection of methods for attained growth curves. Statistics in Medicine vol. 25 247–265 Preprint at (2006). https://doi.org/10.1002/sim.2227 Benjamini, Y. & Hochberg Yosef. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testeing. J. Royal Stat. Soc. Ser. B (Metodological) . 57 , 289–300 (1995). Additional Declarations No competing interests reported. Supplementary Files suppmlementary271025.pdf Cite Share Download PDF Status: Published Journal Publication published 20 Feb, 2026 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 08 Dec, 2025 Reviews received at journal 06 Dec, 2025 Reviews received at journal 02 Dec, 2025 Reviews received at journal 24 Nov, 2025 Reviewers agreed at journal 12 Nov, 2025 Reviewers agreed at journal 12 Nov, 2025 Reviewers agreed at journal 11 Nov, 2025 Reviewers agreed at journal 10 Nov, 2025 Reviewers invited by journal 10 Nov, 2025 Editor assigned by journal 10 Nov, 2025 Editor invited by journal 10 Nov, 2025 Submission checks completed at journal 07 Nov, 2025 First submitted to journal 07 Nov, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8011139","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":544143832,"identity":"90dea0e1-c8e6-4199-a2f9-ef0cf9ea42be","order_by":0,"name":"Bossmat Yehuda","email":"","orcid":"","institution":"Tel-Aviv Sourasky Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Bossmat","middleName":"","lastName":"Yehuda","suffix":""},{"id":544143834,"identity":"36b2b3c1-6034-4f69-999e-10f67ec8731a","order_by":1,"name":"Rotem Gal","email":"","orcid":"","institution":"Tel Aviv University","correspondingAuthor":false,"prefix":"","firstName":"Rotem","middleName":"","lastName":"Gal","suffix":""},{"id":544143839,"identity":"a508b096-38d5-450e-a4e4-1495a2034b65","order_by":2,"name":"Yair Wexler","email":"","orcid":"","institution":"Tel Aviv University","correspondingAuthor":false,"prefix":"","firstName":"Yair","middleName":"","lastName":"Wexler","suffix":""},{"id":544143840,"identity":"fbaafd08-3e08-4ff4-a295-42bd509d6693","order_by":3,"name":"Aviad Rabinowich","email":"","orcid":"","institution":"Tel-Aviv Sourasky Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Aviad","middleName":"","lastName":"Rabinowich","suffix":""},{"id":544143841,"identity":"e7a10481-955d-4bae-855f-d92e000732c6","order_by":4,"name":"Elka Miller","email":"","orcid":"","institution":"University of Toronto","correspondingAuthor":false,"prefix":"","firstName":"Elka","middleName":"","lastName":"Miller","suffix":""},{"id":544143842,"identity":"b877e150-bb16-4d1d-b9af-7d858ee2a648","order_by":5,"name":"Liat Ben Sira","email":"","orcid":"","institution":"Tel Aviv University","correspondingAuthor":false,"prefix":"","firstName":"Liat","middleName":"Ben","lastName":"Sira","suffix":""},{"id":544143844,"identity":"2aa68136-5ecb-4662-8417-62efbefbe522","order_by":6,"name":"Dafna Ben Bashat","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA1UlEQVRIie3PvQrCMBDA8QuF62LbNaHoM1Rci+8SCl1EfAPJlMn6LiKIYyGgS7VrwaXQWeioomAnPxZjN8H8x+N+cAdgMv1iKSmBiBDcx8SyhIZYQUNiik9CviLqhYCGuFuF5XmdT5GOeA2XkAtbQ1gW2f0kO1Ck4yUl85gL3WFBOtlQRzaks1sBmamB9pcgr5Dd5L4NKSL0HZlStJMVwEn1tIQVFfrdLGLSTpaUi7iHOuLmEbLjeuh5lrOo62vY8TyVfiTvcQmA+rW3ri33TSaT6S+6A5A1QbyrKVIMAAAAAElFTkSuQmCC","orcid":"","institution":"Tel-Aviv Sourasky Medical Center","correspondingAuthor":true,"prefix":"","firstName":"Dafna","middleName":"Ben","lastName":"Bashat","suffix":""}],"badges":[],"createdAt":"2025-11-02 13:08:28","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8011139/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8011139/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-026-38229-9","type":"published","date":"2026-02-20T15:57:17+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":96421924,"identity":"0569ae73-3178-4f76-92ea-075072bd5ecf","added_by":"auto","created_at":"2025-11-21 01:07:05","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":485845,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript71125revised.docx","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/c3d630a6412231b0ee64c7d5.docx"},{"id":96453733,"identity":"939cc855-0d91-4364-a159-37a773b9b6eb","added_by":"auto","created_at":"2025-11-21 10:01:25","extension":"jpg","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1304684,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/02b10c309bf7020baa62a1d0.jpg"},{"id":96421926,"identity":"3183e932-cc58-4941-bf15-719895fa8d19","added_by":"auto","created_at":"2025-11-21 01:07:05","extension":"jpg","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1819738,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/c216611cfcf31cf133f91f0d.jpg"},{"id":96454682,"identity":"878e7cb0-ede0-4dab-9cf2-b966a46f693a","added_by":"auto","created_at":"2025-11-21 10:03:02","extension":"jpg","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":337033,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/3f3f4aa8dc58a4c48884d21a.jpg"},{"id":96421932,"identity":"73a97497-dd4e-468b-8111-93887127333c","added_by":"auto","created_at":"2025-11-21 01:07:05","extension":"jpg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":727171,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/fd7df5aa0bb1fc9f598565c9.jpg"},{"id":96421936,"identity":"fd4b2d09-6464-4199-afa9-f81cb3b37dc7","added_by":"auto","created_at":"2025-11-21 01:07:05","extension":"jpg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1295045,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/06656a267290ad6854262ea5.jpg"},{"id":96454546,"identity":"0e33a3e5-be51-4632-8c29-30528b61afd9","added_by":"auto","created_at":"2025-11-21 10:02:52","extension":"json","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":9266,"visible":true,"origin":"","legend":"","description":"","filename":"1b4484c79e5340919d8ffe3bb6a0d03b.json","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/c62ad95e8b6be065a46c6893.json"},{"id":96455108,"identity":"c411b040-93f0-4148-b55d-fb158c0f9156","added_by":"auto","created_at":"2025-11-21 10:03:34","extension":"pdf","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":313519,"visible":true,"origin":"","legend":"","description":"","filename":"suppmlementary271025.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/30f0d53c28a513a4bd1b3c03.pdf"},{"id":96421931,"identity":"04fc5ee6-3c41-445a-9a76-c7dfd0e3dea5","added_by":"auto","created_at":"2025-11-21 01:07:05","extension":"xml","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":105685,"visible":true,"origin":"","legend":"","description":"","filename":"1b4484c79e5340919d8ffe3bb6a0d03b1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/e5e72d4793d2dec27ee6f7c6.xml"},{"id":96454567,"identity":"05a634ef-4be6-41a4-b39e-f520199478fe","added_by":"auto","created_at":"2025-11-21 10:02:54","extension":"jpg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1304684,"visible":true,"origin":"","legend":"","description":"","filename":"Figure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/618e2763d2cab4faaf8e7371.jpg"},{"id":96421944,"identity":"581c6a81-25ed-40e6-9482-e7f37930e334","added_by":"auto","created_at":"2025-11-21 01:07:06","extension":"jpg","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1819738,"visible":true,"origin":"","legend":"","description":"","filename":"Figure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/194158ec4aa98a92a6565fea.jpg"},{"id":96421937,"identity":"a1f64cd8-9ba1-4219-904b-c772e35ab1ee","added_by":"auto","created_at":"2025-11-21 01:07:05","extension":"jpg","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":337033,"visible":true,"origin":"","legend":"","description":"","filename":"Figure3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/7d163ff12c541ee3ac90984b.jpg"},{"id":96421933,"identity":"40ea5f85-0504-4402-bbdd-ebb09bd118b3","added_by":"auto","created_at":"2025-11-21 01:07:05","extension":"jpg","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":727171,"visible":true,"origin":"","legend":"","description":"","filename":"Figure4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/0052d530d771ca22416af827.jpg"},{"id":96421945,"identity":"0b747d3b-231e-42d9-8422-953b6b6c7906","added_by":"auto","created_at":"2025-11-21 01:07:06","extension":"jpg","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1295045,"visible":true,"origin":"","legend":"","description":"","filename":"Figure5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/52035f1243a3dfd80863ed44.jpg"},{"id":96421941,"identity":"f8da6fa1-5de7-459d-922b-3419cedebf12","added_by":"auto","created_at":"2025-11-21 01:07:06","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":329712,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure1.png","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/2765ff69455e7b0f2b96c32a.png"},{"id":96421943,"identity":"052692ca-3ac8-42c3-8bf3-3dc91f0fdb24","added_by":"auto","created_at":"2025-11-21 01:07:06","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":527202,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure2.png","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/b14d5594243953f3096fdb80.png"},{"id":96454772,"identity":"bd5e9e0f-45fe-4b7d-92eb-3ea40a445e27","added_by":"auto","created_at":"2025-11-21 10:03:07","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1310023,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure3.png","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/c7b60f37dfc6d63a71322dba.png"},{"id":96453960,"identity":"3bf5e149-f6f7-4cf6-82cd-b457662d08b4","added_by":"auto","created_at":"2025-11-21 10:02:09","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":256997,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure4.png","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/36217c9cf1b95b1248077e1c.png"},{"id":96421938,"identity":"b1ea15f4-3a00-4dd5-8bc1-5b8483973b5e","added_by":"auto","created_at":"2025-11-21 01:07:05","extension":"png","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":452789,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineFigure5.png","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/cd3fe415e8e309be0c7cf29f.png"},{"id":96421947,"identity":"13d660d1-fc54-4f2a-b052-383cf908ce40","added_by":"auto","created_at":"2025-11-21 01:07:06","extension":"xml","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":103838,"visible":true,"origin":"","legend":"","description":"","filename":"1b4484c79e5340919d8ffe3bb6a0d03b1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/ec2d8a06b4869fce868bf7c3.xml"},{"id":96421946,"identity":"55a215c7-ff8c-4c1d-8a4d-e0cfe6469eb3","added_by":"auto","created_at":"2025-11-21 01:07:06","extension":"html","order_by":20,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":116890,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/235b61fa9b4b38cd49c8b739.html"},{"id":96421930,"identity":"3b984e1b-1d27-45f0-8f32-f785c3085c1e","added_by":"auto","created_at":"2025-11-21 01:07:05","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1304684,"visible":true,"origin":"","legend":"\u003cp\u003eGyrification spectral features along gestation for TD controls using GAMLSS model. Cases with LIS and PMG are also indicated. Centile lines: 3rd, 15th, 50th, 85th, 97th. Abbreviations: GA = gestational age, GAMLSS = generalized additive models for location, scale and shape, TD = typically developing, LIS = lissencephaly, PMG = polymicrogyria.\u003c/p\u003e","description":"","filename":"Figure1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/201eefc75a59cf05f385bf31.jpg"},{"id":96421925,"identity":"031f55d7-f164-4f31-8625-762e5b2cb0f9","added_by":"auto","created_at":"2025-11-21 01:07:05","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1819738,"visible":true,"origin":"","legend":"\u003cp\u003eRight hemisphere’s growth curves of the different frequencies across gestation for TD controls using GAMLSS model. Cases with LIS and PMG are also indicated. Centile lines: 3rd, 15th, 50th, 85th, 97th. Abbreviations: GA = gestational age, GAMLSS = generalized additive models for location, scale and shape, LIS = lissencephaly, PMG = polymicrogyria, TD = typically developing.\u003c/p\u003e","description":"","filename":"Figure2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/0eaec65dd339f5bbbc7c329f.jpg"},{"id":96421929,"identity":"caee986e-6ad0-4315-af59-840fd61a56e4","added_by":"auto","created_at":"2025-11-21 01:07:05","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":337033,"visible":true,"origin":"","legend":"\u003cp\u003eCoronal view of three cases. (\u003cstrong\u003ea\u003c/strong\u003e) A TD fetus at 33 weeks of gestation; (\u003cstrong\u003eb\u003c/strong\u003e) a fetus diagnosed with lissencephaly at 32 weeks of gestation; and (\u003cstrong\u003ec\u003c/strong\u003e) a fetus diagnosed with polymicrogyria at 33 weeks of gestation. Arrows pointing on abnormal excessive gyri. Abbreviations: GA = gestational age, TD = typically developing.\u003c/p\u003e","description":"","filename":"Figure3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/db90e273cc87529b0660379e.jpg"},{"id":96453932,"identity":"113b1fcf-e02c-4c53-a3aa-d934144a1789","added_by":"auto","created_at":"2025-11-21 10:02:06","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":727171,"visible":true,"origin":"","legend":"\u003cp\u003eSpectral gyrification parameters, after normalization with respect to GA. Significance level are as follows: * \u0026lt;0.05; ***\u0026lt;0.001. Abbreviations: TD = typically developing, LIS = lissencephaly; PMG = polymicrogyria.\u003c/p\u003e","description":"","filename":"Figure4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/08fd18ae019dab782dfa8dc0.jpg"},{"id":96454729,"identity":"f44a4bbe-0999-469c-b7df-90ffc6b158d6","added_by":"auto","created_at":"2025-11-21 10:03:04","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1295045,"visible":true,"origin":"","legend":"\u003cp\u003eImage processing pipeline: (\u003cstrong\u003ea\u003c/strong\u003e) Stack of fetal MRI scan of a TD fetus at the 34\u003csup\u003eth\u003c/sup\u003e gestation week following brain detection. Example shown is slice 11 with the extracted hemisphere contour (green) and the center of mass (red dot). (\u003cstrong\u003eb\u003c/strong\u003e) Polar transformation of the contour. (\u003cstrong\u003ec\u003c/strong\u003e) Discrete Fourier Transform of the polar contour, with the color bar indicating normalized frequency amplitudes. (\u003cstrong\u003ed\u003c/strong\u003e) Spectral matrix for a single fetus: each row corresponds to a slice, and each column to a frequency component. (\u003cstrong\u003ee\u003c/strong\u003e) Final spectral representation obtained by averaging frequency amplitudes across all slices. Abbreviations: MRI = magnetic resonance imaging, TD = typically developing.\u003c/p\u003e","description":"","filename":"Figure5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/8e83b05de332a794e80b4fb1.jpg"},{"id":103251244,"identity":"9fd506ad-31b0-44b2-aa07-4291f1212b87","added_by":"auto","created_at":"2026-02-23 16:07:01","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":6230965,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/3b4d5245-da3b-473d-9122-cf90e9752d8d.pdf"},{"id":96421922,"identity":"79a8cd80-5ba3-4df2-8805-681d0ae6f6c8","added_by":"auto","created_at":"2025-11-21 01:07:05","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":313519,"visible":true,"origin":"","legend":"","description":"","filename":"suppmlementary271025.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8011139/v1/8b1a1308a8c3f8ce422ec910.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"MRI-Based Spectral Analysis of Fetal Brain Gyrification: Applications to Lissencephaly and Polymicrogyria","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThe human\u0026rsquo;s brain cortical development begins prenatally and involves substantial changes in shape, size, tissue organization, and maturation. Gyrification, the formation of gyri and sulci, emerges during mid-gestation, accelerates throughout the third trimester, and continues after birth.[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e] Gyrification follows an organized spatiotemporal pattern, and progresses in three waves: the primary wave spanning roughly 20\u0026ndash;30 weeks (peak\u0026thinsp;~\u0026thinsp;24), the secondary wave across 30\u0026ndash;36 weeks (peak\u0026thinsp;~\u0026thinsp;32), and the tertiary wave starts at approximately 38 weeks and extends into early postnatal life. Overall, gyrification shows a strong gestational age (GA)-dependent trajectory and is a key marker of brain development.[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]\u003c/p\u003e\u003cp\u003eMalformations of cortical development (MCDs) may result from a range of etiologies, including genetic mutations, intrauterine infections, toxin exposure or ischemia,[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e] and typically become detectable after 24 weeks of gestation.[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e] Lissencephaly and polymicrogyria are two types of MCDs associated with a broad spectrum of neurodevelopmental outcomes, including severe epilepsy and motor impairments,[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e] depending on their type, severity, and anatomical extent.[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e] Lissencephaly is characterized by a smooth cortical surface and a thickened, disorganized cortex. In contrast, polymicrogyria is characterized by excessive and irregular folding with thin, fused cortical layers, which may occur focally, diffusely, unilaterally, or bilaterally.[\u003cspan additionalcitationids=\"CR8\" citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]\u003c/p\u003e\u003cp\u003eWhile typically, the assessment of fetal cortical development is based on qualitative visual inspection of gyrification patterns using ultrasound (US) or magnetic resonance imaging (MRI),[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e] previous studies used the gyrification index (GI), defined as the ratio between the cortical contour to its convex hull,[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e] along with additional methods to quantify cortical folding patterns.[\u003cspan additionalcitationids=\"CR15\" citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e] A few studies have developed methods for fetal gyrification analysis based on 3D-reconstructed MRI,[\u003cspan additionalcitationids=\"CR18 CR19 CR20\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e] which is limited by long reconstruction times and specific acquisition requirements. Only one study quantified gyrification based on 2D data using GI, demonstrating GA- dependent changes.[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]\u003c/p\u003e\u003cp\u003eOnly two studies to date have quantitatively analyzed gyrification in fetuses with MCDs. One compared three fetuses with polymicrogyria to \u003cb\u003e14\u003c/b\u003e typically developing (TD) fetuses,[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e] reporting diminished alignment to fetal brain templates among cases with polymicrogyria, but no significant differences using 3D GI. The second study that employed 2D GI quantification, detected differences between TD fetuses and fetuses with lissencephaly and polymicrogyria.[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e] This method was further applied on fetuses with growth restriction, demonstrating reduced gyrification patterns. [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] While this method was based on a relatively large cohort and successfully identified lissencephaly abnormalities, it underperformed in early gestation (\u0026lt;\u0026thinsp;28 weeks) and in distinguishing polymicrogyria from TD fetuses, highlighting the need for alternative methodologies.\u003c/p\u003e\u003cp\u003eSpectral analysis, a foundational tool in signal processing, has been widely applied across scientific domains to analyze patterns based on frequency content. The Fourier transform (FT) decomposes signals into their frequency components, revealing underlying oscillations and their amplitudes.[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e] Previous spectral methods applications were used to characterize brain gyrification among adults and neonates.[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e] Collectively, these studies demonstrated that lower frequencies are associated with global brain shape and higher frequencies are associated with local folding patterns. The developing fetal brain is subjected to profound regional and global changes throughout gestation and some MCDs may only be manifested locally, making spectral analysis a potential approach for capturing subtle, spatially confined deviations in cortical folding patterns that might be overlooked by conventional metrics. To our knowledge, spectral analysis has not been previously applied to prenatal gyrification.\u003c/p\u003e\u003cp\u003eThe aim of this study was to assess fetal gyrification using spectral analysis of fetal MRI, characterize GA-related patterns in TD fetuses, and compare TD fetuses with fetuses with lissencephaly and polymicrogyria.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eA total of 150 participants with singleton pregnancies were included. Eleven cases were excluded due to poor image quality, 29 due to genetic abnormalities or MRI findings other than lissencephaly and polymicrogyria, and 13 with GA below 24 weeks. Supplementary Tables\u0026nbsp;1 and 2 further details the demographic and MRI indication referrals. The final cohort comprised of 97 fetuses: 73 TD (mean GA at MRI 31.3\u0026plusmn;2.7 weeks; range, 25-37.6 weeks), 14 with polymicrogyria (mean GA, 31.6\u0026plusmn;2.5 weeks; range, 27.6\u0026ndash;36 weeks), and ten with lissencephaly (mean GA, 29.0\u0026plusmn;3.0 weeks; range, 25\u0026ndash;33, weeks). No significant differences in GA at the time of MRI were observed between TD and MCD fetuses.\u003c/p\u003e\u003cp\u003eSpectral feature trajectories of TD controls demonstrated increased gyrification across gestation, except for skewness which decreased with GA. These changes are illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Supplementary Fig.\u0026nbsp;1 highlights the correlation between the dominant frequency and the complexity degree of the contour. The spatial frequencies of TD controls were characterized by two distinct developmental patterns across gestation. Mid-frequency band (frequencies 6,7,9, and 10) showed a progressive increase with gestation, representing global gyrification. In contrast, the lower frequency band (frequencies 1,3,4,5) exhibited accelerated growth between 24 to 32 weeks, followed by a plateau from 32 weeks onwards. The highest frequency band (frequencies 11 and 12) was characterized by accelerated growth beginning around 31 weeks. Notably, frequency 2 decreased along gestation. Changes in spatial frequencies with gestation are further illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. Symmetry analysis demonstrated homogeneous dispersion among TD, lissencephaly and polymicrogyria cases fetuses and in all features.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eRepresentative cases of TD controls, lissencephaly and polymicrogyria are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. Fetuses with MCD demonstrated significant differences in gyrification compared with TD fetuses. Lissencephaly cases exhibited an 11.5% reduction in non-zero spectral density, a 2.7% reduction in entropy, a 4.4% reduction in variance, and a 7% increase in skewness relative to TD fetuses (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Additionally, apart from frequencies 2 (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.17) and 9 (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.06), all other frequencies were significantly lower in lissencephaly compared with TD controls (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026le;\u0026thinsp;0.015).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eCompared to TD controls, polymicrogyria cases showed a 7.8% reduction in non-zero spectral density (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001), and a 1.6% reduction in entropy (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.03), with no significant differences in spectral variance (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.17) or skewness (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.12). Apart from frequencies 2, 9 and 11 (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.099, 0.547 and 0.19, respectively) all other spatial frequencies were significantly reduced in polymicrogyria compared with TD controls (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.05).\u003c/p\u003e\u003cp\u003eResults of the comparison between TD, lissencephaly and polymicrogyria are summarized in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e2\u003c/span\u003e and illustrated in Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e and \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eGyrification spectral features changes in percentages for lissencephaly and polymicrogyria cases, compared with TD controls. P-values are presented after adjustment with Benjamini-Hochberg. Abbreviations: TD\u0026thinsp;=\u0026thinsp;typically developing.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"5\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFeature\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eLissencephaly [%]\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eP-value\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003ePolymicrogyria [%]\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eP-value\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNon-zero Spectral Density\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-11.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-7.8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEntropy\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-2.7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-1.6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.031\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e1st Moment - Mean\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-0.7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.34\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-0.7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.34\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2nd Moment - Variance\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-4.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-3.1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.17\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e3rd Moment - Skewness\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.12\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFrequency 1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-29.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-19.1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFrequency 2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-3.3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.099\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-0.6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.099\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFrequency 3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-10.8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.016\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-5.8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.015\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFrequency 4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-2.9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-7.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.004\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFrequency 5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-19.3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-12.1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.011\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFrequency 6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-16.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-8.0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.008\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFrequency 7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-14.1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-11.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.016\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFrequency 8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-18.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-13.1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.042\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFrequency 9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-18.9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.062\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-6.0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.547\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFrequency 10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-16.6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-12.3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.003\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFrequency 11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-6.8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-8.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.19\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eFrequency 12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-15.1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-13.9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.03\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study presents the application of spectral analysis to fetal brain gyrification quantification from 2D-MRI in TD fetuses and those with lissencephally and polymicrogyryia. Five spectral features and spatial profiles of the first twelve frequencies were extracted. TD fetuses exhibited increasing values across all frequencies and features with advancing GA, showing growth characteristics corresponding to the first two gyrification waves. Fetuses with MCDs demonstrated decreased gyrification in most gyrification-based spectral frequencies, compared with TD subjects, with more pronounced differences among cases with lissencephaly. Importantly, the spectral representation also detected distinct patterns in polymicrogyria, despite the heterogenous imaging characteristics of this malformation enabling developmental stage characterization.\u003c/p\u003e\u003cp\u003eCortical folding begins in second trimester and involves cell proliferation, neuronal migration, and cortical organization. [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e] Among fetuses with TD, there was and increase in mean frequency and variance along gestation, reflecting spectral range widening as the cortex develops into a more geometrically complex structure.[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e] Skewness, which quantifies the asymmetry of the spectral distribution relative to its mean frequency,[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e] decreased with gestation. Taken together, these indicate a growing dominance of higher frequencies as pregnancy progresses, highlighting how the spectral features capture global aspects of gyrification during fetal development.\u003c/p\u003e\u003cp\u003ePer specific frequencies, distinct patterns of growth along gestation corresponding to the waves of cortical folding were demonstrated. The low frequency band magnitude rapidly increased between the 25 and 32 weeks of gestation, followed by a plateau, suggesting an association with the primary folding wave. These findings generally represent the global brain gyrification and are consistent with previous studies performed in preterm infants.[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e] In addition, higher frequencies exhibited an accelerated increase from 32 weeks onwards, consistent with the onset of the secondary folding wave and local gyrification development. As higher frequencies correspond to more complex contour features with shorter distances between oscillations, this increase is likely to reflect the emergence of more intricate sulcal patterns. Conversely, frequency 2 decreased steadily with gestation, likely representing the gradual reduction of the smooth elliptical shape of cerebral contour as it evolves to a more complex structure. Moreover, the Sylvian fissure is the principal contributor to the bi-lobar configuration, suggesting a correlation between the emergence of this primary fissure and lower freqeuncies. Supplementary Fig.\u0026nbsp;2 illustrates the association between hemispheric contour with and without the Sylvian fissure and frequency 1, which demonstrated the highest significant difference compared to MCDs.\u003c/p\u003e\u003cp\u003eSpectral features and the overall spectral profile were symmetrical between the two hemispheres, consistent with prior studies of cortical gyrification.[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e] Although regional asymmetries have been described in specific anatomical locations,[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e] the global nature of this approach precludes detection of such localized variations.\u003c/p\u003e\u003cp\u003eEarly and accurate diagnosis MCD is crucial, as patients often present within the first year of life with hypotonia, global developmental delay, and seizures, even when no other major brain anomalies are detected. The severity of neurological outcomes is known to correlate with the type and extent of abnormalities observed on MRI.[\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e] A small number of studies previously assessed quantitative differences between gyrification in fetuses with MCD compared with TD,[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan additionalcitationids=\"CR22\" citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] demonstrated reduced gyrification among lissencephaly cases. Polymicrogyria cases show some defferences compared with TD, yet its local and heterogenous appearance reduces the ability to reliably separate TD from polymicrogyria.[\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. These findings are consistent with our results, showing underdeveloped gyrification and reduced spectral dispersion in lissencephaly. Notably, non-zero spectral density values were significantly lower in lissencephaly cases prior to 30 weeks, suggesting that our method is sensitive to early lissencephaly-related gyrification patterns, a stage where the fetal brain surface is typically smooth and may resemble a lissencephalic pattern. Contrary to our hypothesis, the mean frequency (first moment) did not differ between groups, which likely reflects subtle spectral shifts that are not captured by an averaged central measure. Lissencephaly cases did, however, show significantly higher skewness and lower variance values, suggesting that spectral energy was disproportionately concentrated in lower frequencies. The most pronounced decrease was observed in frequency 1 which is most likely linked to the Sylvian fissure and is included in the representing band for the first gyrification wave. This observation quantitatively supports the known malformation of the Sylvian fissure in lissencephaly,[\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e] which disrupts the development of the Sylvian fissure during the first wave of gyrification.\u003c/p\u003e\u003cp\u003eThe three spectral moments were not different among fetuses with polymicrogyria compared with TD controls. However, non-zero density and entropy were significantly lower among polymicrogyria subjects, suggesting that spectral measures may be sensitive to subtle abnormalities that are characteristic of polymicrogyria. In addition, high-frequency components demonstrated lower values, which may indicate the presence of shallow, irregular sulci which are hallmark feature of polymicrogyria. Since sulcal depth is encoded as the distance from the cortical contour to the brain\u0026rsquo;s center of mass, shallow sulci correspond to lower intensities at their respective frequencies. Moreover, polymicrogyria cases demonstrated decreased values in the low frequency range, particularly frequency 1, which is directrly linked to the development of the Sylvian fissure. Taken together, these spectral differences between polymicrogyria and TD fetuses indicate measurable alterations in gyrification patterns, suggesting the potential use of this method for assessing the developmental stage characterization and the extent of MCDs.\u003c/p\u003e\u003cp\u003eCollectively, our findings support the potential utility of spectral analysis for quantitative assessment of gyrification. Specifically, our results suggest an advantage of this approach over previous methods in detecting abnormalities associated with polymicrogyria. Future studies should aim to localize regions of abnormality and delineate the extent of malformation to enhance model interpretability for clinicians. In addition, the relationship between spectral measures and neurodevelopmental outcomes merits further investigation.\u003c/p\u003e\u003cp\u003eSeveral limitations should be acknowledged. While our cohort is relatively large compared with other studies in the field, the number of subjects with MCD is limited. Our findings should be validated in larger MCD cohorts to assess the correlation gyrification-based frequencies and neurological outcomes, along with genetic findings. Moreover, the cohort was retrospectively collected, and some demographic information was not available. In addition, in this study, manual corrections were applied over automatic deep leaning based segmentation to ensure segmentation accuracy. Future development should include improving segmentation accuracy and explore super resolution reconstruction.\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eThis study introduces a quantitative spectral framework of fetal brain gyrification from MRI studies. Spectral-based features changed across gestation aligning with developmental trajectories and expected patterns of gyrification waves, and revealed significant differences between fetuses with MCDs and TD controls, including a marked reduction in total gyrification and frequencies linked to the Sylvian fissure development.\u003c/p\u003e\u003cp\u003eOur findings suggest that the spectral representation of cortical folding may serve as a biomarker for brain maturation, and distinguish fetuses with lissencephaly and polymicrogyria from TD subjects. This method may also be extended to study the gyrification patterns of other prenatal conditions with delayed cortical maturation such as fetal growth restriction and congenital cytomegalovirus infection.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003ePopulation\u003c/h2\u003e\u003cp\u003eThe fetal brain MRI were collected at two clinical sites between 2007 to 2022, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel and the Children's Hospital of Eastern Ontario (CHEO), Ottawa, Canada. Data collection and analysis were performed in accordance with relevant guidelines and regulations.\u003c/p\u003e\u003cp\u003eTD controls included singleton pregnancies referred for MRI due to various sonographic suspicious abnormalities, such as ventriculomegaly, microcephaly, and club foot, or a previous pregnancy with abnormal development. Exclusion criteria included MRI detected structural brain anomalies, chromosomal abnormalities, amniocentesis-confirmed cytomegalovirus infection, chronic maternal disease, or poor image quality. Fetuses scanned prior to the 24th gestation week were excluded, as gyrification is not yet reliably visible. MCD cases were defined as fetuses diagnosed with lissencephaly or polymicrogyria, based on expert interpretation of both US and MRI.\u003c/p\u003e\u003cp\u003eAll scans were evaluated by expert fetal neuroradiologists (LBS or EM), each with more than 20 years of experience.\u003c/p\u003e\u003cp\u003e The study was approved by the Tel Aviv Sourasky Medical Center and the Children's Hospital of Eastern Ontario (CHEO Institutional Review Boards. The informed consent was waived by the Institutional Review Boards due to the retrospective nature of the study.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eMRI data\u003c/h3\u003e\n\u003cp\u003eThe MRI data included scans acquired using two MRI field strengths,3T and 1.5T, from two vendors (GE Healthcare and Siemens Healthineers). The protocol included brain coronal sequences (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e1\u003c/span\u003e) using either Fast Recovery Fast Spin Echo (FRFSE), Fast Imaging Employing Steady-State Acquisition (FIESTA), Half-Fourier Single-Shot Turbo Spin-Echo (HASTE) or True Fast Imaging with Steady-State Free Precession (TruFISP).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eImaging parameters of MRI sequences. Abbreviations: FRFSE\u0026thinsp;=\u0026thinsp;fast-recovery fast spin-echo; FIESTA- fast imaging employing steady-state acquisition; HASTE- half Fourier single-shot turbo spin-echo; TruFISP- true fast imaging with steady-state free precession.\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVendor\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSystem (magnetic field)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSequence\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eEcho time [milisec]\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eRepetition time [milisec]\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eIn-plane resolution [mm]\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eSlice thickness [mm]\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003eGE Healthcare\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eDISCOVERY MR450 (1.5T)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eFRFSE\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e119\u0026ndash;124\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e7469\u0026ndash;11421\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e4.0\u0026ndash;5.8\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSIGNA (1.5T)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eFIESTA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.7\u0026ndash;1.8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e3.8\u0026ndash;4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.6\u0026ndash;0.7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e4\u0026ndash;6\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSigna HDxt (1.5T)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eFIESTA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.6\u0026ndash;1.7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e3.8\u0026ndash;3.9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e3.7\u0026ndash;5.2\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"3\" rowspan=\"4\"\u003e\u003cp\u003eSIEMENS Healthineers\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAera (1.5T)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eHASTE\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e94\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1200\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1.0 \u0026ndash; 1.2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e3\u0026ndash;4\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003ePrisma (3T)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eHASTE\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e86\u0026ndash;109\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e2000\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e2\u0026ndash;5\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eSkyra (3T)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eHASTE\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e75\u0026ndash;93\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e1711\u0026ndash; 2229\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.4\u0026ndash;0.7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e2.5\u0026ndash;4\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eTruFISP\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e2.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e4.9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eImage analysis\u003c/h2\u003e\u003cp\u003eThe image processing pipeline (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) included two main stages:\u003c/p\u003e\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003eHemispheric contour extraction (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA)\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"BlockQuote\"\u003e\u003cp\u003eThis step follows a previously described methodology,[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e] summarized as follows:\u003c/p\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e\u003col\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e(1.1) The right and left cerebral hemispheres were automatically segmented using an in-house developed automatic tool and manually refined as needed (BY, 6 years of experience in image segmentation), using ITK-SNAP (version 3.8)[\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e] while blinded to the fetal condition.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e(1.2) Cerebral contour definition: The contour was defined as the boundary between the cerebral hemispheres and the extra-axial cerebrospinal fluid, based on the segmentation in 1.1.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003eSpectral contour analysis was performed per slice for each hemisphere separately, based on the cerebral contour:\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e(2.1) Polar transformation of the contour (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB): Each contour point was represented by its index along the contour (x-axis) and its distance from the convex center of mass, normalized by the maximal distance (y-axis).\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e(2.2) Discrete FT of the polar representation (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC): The FT was applied, with a sampling size defined as the inverse of the contour length, yielding natural frequency values corresponding to the number of oscillations per full contour traversal. For example, a frequency of 3 corresponds to an oscillation repeating three times along the contour.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e(2.3) Normalization of the spectral array for each slice (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD): Each spectral vector was scaled by the sum of all frequency amplitudes in the corresponding slice, producing a size-independent spectral probability distribution. A predefined noise threshold was applied to remove artifacts, set at the cumulative sum of spectral energy equal to 95%, yielding a spectral representation of frequencies 1 through 12.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003cspan\u003e\u003cli\u003e\u003cp\u003e(2.4) Spectral profile computation (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eE): The final spectral representation was obtained by averaging the normalized spectra of all slices into a single vector containing the amplitudes of the natural frequencies.\u003c/p\u003e\u003c/li\u003e\u003c/span\u003e\u003c/ol\u003e\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eGyrification feature extraction\u003c/h3\u003e\n\u003cp\u003eThe spectral profile extracted from step 2.4, which included the first 12 individual frequencies (p\u003csub\u003ei\u003c/sub\u003e), was analyzed across GA and between groups. The twelve frequencies were categorized as low (frequencies 1\u0026ndash;5), middle (frequencies 6\u0026ndash;10), and high (frequencies 11\u0026ndash;12) frequency bands, according to the frequency bands described.[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]\u003c/p\u003e\u003cp\u003eIn addition, five features were extracted per fetus for each hemisphere (see Eq.\u0026nbsp;1\u0026ndash;5):\u003c/p\u003e\u003cp\u003e\u003cul\u003e\u003cli\u003e\u003cp\u003eNon-zero spectral density: (Eq.\u0026nbsp;1) Represents total gyrification by capturing the intensity of all non-zero frequencies. Frequency zero corresponds to the mean of the polar-transformed contour, reflecting the average hemisphere radius.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eSpectral entropy: (Eq.\u0026nbsp;2) Describes the distribution of the spectral power across frequencies\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eMean frequency: )Eq.\u0026nbsp;3) The first moment of the spectrum, reflecting its central tendency.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eVariance: (Eq.\u0026nbsp;4) The second moment, relative to the first moment, quantifying spectral spread.\u003c/p\u003e\u003c/li\u003e\u003cli\u003e\u003cp\u003eSkewness: (Eq.\u0026nbsp;5) The third moment, relative to the first moment, capturing the asymmetry of the frequency distribution.\u003c/p\u003e\u003c/li\u003e\u003c/ul\u003e\u003c/p\u003e\u003cp\u003eEq.\u0026nbsp;1 \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:non-zero\\:spectral\\:density=\\:\\sum\\:_{f=1}^{N}{p}_{f}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\u003cp\u003eEq.\u0026nbsp;2 \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:Spectral\\:Entropy=\\:-\\sum\\:_{f=1}^{N}{p}_{f}\\bullet\\:{log}_{2}\\left({p}_{f}\\right)\\:\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\u003cp\u003eEq.\u0026nbsp;3 \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:Mean=\\:\\frac{\\sum\\:_{f=1}^{N}(f\\bullet\\:{p}_{f})}{\\sum\\:_{f=1}^{N}{p}_{f}}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\u003cp\u003eEq.\u0026nbsp;4 \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:Variance=\\:\\frac{\\sum\\:_{f=1}^{N}{\\left(\\right(f-Mean)}^{2}\\bullet\\:{p}_{f})}{\\sum\\:_{f=1}^{N}{p}_{f}}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\u003cp\u003eEq.\u0026nbsp;5 \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:Skewness=\\:\\frac{\\sum\\:_{f=1}^{N}{\\left(\\right(f-Mean)}^{3}\\bullet\\:{p}_{f})}{{\\left(\\sqrt{Variance}\\right)}^{3}\\sum\\:_{f=1}^{N}{p}_{f}}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\u003cp\u003eWhere \u003cem\u003eN\u003c/em\u003e is the maximal frequency, and \u003cem\u003ep\u003c/em\u003e\u003csub\u003e\u003cem\u003ef\u003c/em\u003e\u003c/sub\u003e is the spectral amplitude of frequency \u003cem\u003ef\u003c/em\u003e.\u003c/p\u003e\u003cp\u003eSymmetry evaluation was performed for each frequency and for each of the five features using the following equation, with L and R representing the left and right hemispheres, respectively:\u003c/p\u003e\u003cp\u003eEq.\u0026nbsp;6 \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:Symmetry=\\:\\frac{L-R}{L+R\\:}\\)\u003c/span\u003e\u003c/span\u003e\u003c/p\u003e\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e\u003ch2\u003eStatistical analysis\u003c/h2\u003e\u003cp\u003eStatistical analysis was performed using a designated R script, version 022.07.2.\u003c/p\u003e\u003cp\u003eChanges along GA in the TD group were assessed for all features and for all frequencies by using generalized additive models for location, scale and shape (GAMLSS) with box-cox power exponential distribution, as recommended by the World Health Organization.[\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]\u003c/p\u003e\u003cp\u003eAll features and individual frequencies were normalized due to significant correlation with GA, with absolute Pearson r value ranging from 0.27 to 0.82. Briefly, to remove the influence of GA, each feature X was normalized using a power-law transformation: X\u0026rsquo; = X/GA\u003csup\u003e\u003cem\u003eλ\u003c/em\u003e\u003c/sup\u003e, where \u003cem\u003eλ\u003c/em\u003e minimized the Pearson correlation X\u0026rsquo; and GA. The exponent \u003cem\u003eλ\u003c/em\u003e varied across measures, ranging between 0.42 for entropy and 2.40 for \u003cem\u003ep\u003c/em\u003e\u003csub\u003e\u003cem\u003e9\u003c/em\u003e\u003c/sub\u003e. For negatively correlated measures (\u003cem\u003eskewness\u003c/em\u003e and \u003cem\u003ep\u003c/em\u003e\u003csub\u003e\u003cem\u003e2\u003c/em\u003e\u003c/sub\u003e), \u003cem\u003eλ\u003c/em\u003e was negative (-1.33 and \u0026minus;\u0026thinsp;0.95, respectively). After normalization, the absolute correlation was reduced to \u0026le;\u0026thinsp;0.07. Importantly, \u003cem\u003eλ\u003c/em\u003e was determined only from the TD controls. Log transform was applied to GA-normalized features when necessary to achieve normality and homogeneity of variances. As gyrification was estimated for each subject in the right and left hemispheres separately and no asymmetry was detected, the hemisphere and the subject ID were used as random variables in the linear mixed model.\u003c/p\u003e\u003cp\u003eComparison between TD, lissencephaly and polymicrogyria was done using all 17 normalized spectral frequencies and features. Each frequency or feature of gyrification was compared between MCD group and TD using a linear mixed model, with pathology, GA, and their interaction included as fixed effects. Although normalization eliminated typical GA effects, GA and its interaction with pathology were retained to assess pathology-specific GA effects on gyrification. Post-hoc analysis was performed using t-tests with degrees of freedom estimated by the Kenward-Roger method and only included comparisons to the TD group. All \u003cem\u003ep\u003c/em\u003e-values of model effects and post hoc tests for all measures were grouped and corrected for multiple comparisons using the Benjamini-Hochberg procedure,[\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e] controlling false discovery rate at the 0.05 level. Throughout the study, \u003cem\u003ep\u003c/em\u003e-values\u0026thinsp;\u0026lt;\u0026thinsp;0.05 were considered statistically significant.\u003c/p\u003e\u003c/div\u003e"},{"header":"Declarations","content":"\u003ch2\u003eFunding statement:\u003c/h2\u003e\u003cp\u003eThis research was supported by the Israel Innovation Authority; Yoran Institute of Human Genome Research; and March of Dimes.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eB.Y. conceived and designed the study, developed the image processing pipeline, performed data curation, analysis, and visualization, and wrote the original manuscript draft. R.G. contributed to the methodological design. Y.W. conducted the statistical analyses. A.B. contributed to validation and to manuscript review and editing. D.B. supervised the project and contributed to the study design and critical manuscript revision. E.M. and L.B.S. reviewed the MRI scans and provided clinical interpretation. All authors read and approved the final version of the manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eWe would like to thank the participants of this study and the MRI radiographers for scanning the participants. We wish good health to all study participants and their newborns.We would like to sincerely thank Mrs. Cassandra Kapoor for her valuable assistance in managing the data from CHEO.This research was supported by Kamin grants (63418, 72126) from the Israel Innovation Authority, the Yoran Institute of Human Genome Research, and March of Dimes.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eDe-identified data supporting the findings reported in this work can be made available by the corresponding author upon reasonable request\u0026nbsp;by researchers who meet the criteria for access to confidential data.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eDubois, J. \u0026amp; Dehaene-Lambertz, G. Fetal and postnatal development of the cortex: MRI and genetics. \u003cem\u003eBrain Mapping: Encyclopedic Ref.\u003c/em\u003e \u003cb\u003e2\u003c/b\u003e, 11\u0026ndash;19 (2015).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSchmitt, S. et al. European Archives of Psychiatry and Clinical Neuroscience Associations of gestational age with gyrification and neurocognition in healthy adults. \u003cem\u003eEur Arch. Psychiatry Clin. Neurosci\u003c/em\u003e \u003cb\u003e1\u003c/b\u003e, 3 .\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGuerrini, R. \u0026amp; Dobyns, W. B. Malformations of cortical development: clinical features and genetic causes. \u003cem\u003eLancet Neurol.\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e, 710\u0026ndash;726 (2014).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKlein, B., Levin, B. E., Duchowny, M. S. \u0026amp; Llabre, M. M. Cognitive outcome of children with epilepsy and malformations of cortical development. \u003cem\u003eNeurology\u003c/em\u003e \u003cb\u003e55\u003c/b\u003e, 230\u0026ndash;235 (2000).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePogliani, L., Andrea Righini, T., Zirpoli, S. \u0026amp; F. \u0026amp; Early Prenatal MR Imaging Diagnosis of Polymicrogyria. \u003cem\u003eAJNR Am. J. Neuroradiol.\u003c/em\u003e \u003cb\u003e25\u003c/b\u003e, 343\u0026ndash;346 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBarkovich, A. J., Guerrini, R., Kuzniecky, R. I., Jackson, G. D. \u0026amp; Dobyns, W. B. A developmental and genetic classification for malformations of cortical development: update 2012. \u003cem\u003eBrain\u003c/em\u003e \u003cb\u003e135\u003c/b\u003e, 1348\u0026ndash;1369 (2012).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGarel, C. \u003cem\u003eMRI of the Fetal Brain Normal Development and Cerebral Pathologies\u003c/em\u003e (Springer-, 2004).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBarkovich, A. J. MRI analysis of sulcation morphology in polymicrogyria. \u003cem\u003eEpilesia\u003c/em\u003e 17\u0026ndash;22. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1111/j.1528-1167.2009.02436.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1528-1167.2009.02436.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2010).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDi Donato, N. et al. Lissencephaly: Expanded imaging and clinical classification. \u003cem\u003eAm. J. Med. Genet. A\u003c/em\u003e. \u003cb\u003e173\u003c/b\u003e, 1473\u0026ndash;1488 (2017).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCassart, M. \u0026amp; Garel, C. European overview of current practice of fetal imaging by pediatric radiologists: a new task force is launched. \u003cem\u003ePediatr. Radiol.\u003c/em\u003e \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00247-020-04710-4\u003c/span\u003e\u003cspan address=\"10.1007/s00247-020-04710-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGarel, C. Fetal cerebral biometry: Normal parenchymal findings and ventricular size. \u003cem\u003eEuropean Radiology\u003c/em\u003e vol. 15 809\u0026ndash;813 Preprint at (2005). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00330-004-2610-z\u003c/span\u003e\u003cspan address=\"10.1007/s00330-004-2610-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLuders, E. et al. Mapping the Relationship between Cortical Convolution and Intelligence: Effects of Gender. \u003cem\u003eCereb. Cortex September\u003c/em\u003e. \u003cb\u003e18\u003c/b\u003e, 2019\u0026ndash;2026 (2008).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDuvernoy, C. S. et al. Gender differences in adverse outcomes after contemporary percutaneous coronary intervention: An analysis from the Blue Cross Blue Shield of Michigan Cardiovascular Consortium (BMC2) percutaneous coronary intervention registry. \u003cem\u003eAm. Heart J.\u003c/em\u003e \u003cb\u003e159\u003c/b\u003e, 677\u0026ndash;683e1 (2010).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLohmann, G., von Cramon, D. Y. \u0026amp; Colchester, A. C. F. Deep sulcal landmarks provide an organizing framework for human cortical folding. \u003cem\u003eCereb. Cortex\u003c/em\u003e. \u003cb\u003e18\u003c/b\u003e, 1415\u0026ndash;1420 (2008).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKiselev, V. G., Hahn, K. R. \u0026amp; Auer D. P. Is the brain cortex a fractal? \u003cem\u003eNeuroimage\u003c/em\u003e 20, 1765\u0026ndash;74 (2003).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChung, M. K., Dalton, K. M., Shen, L., Evans, A. C. \u0026amp; Davidson, R. J. Weighted fourier series representation and its application to quantifying the amount of gray matter. \u003cem\u003eIEEE Trans. Med. Imaging\u003c/em\u003e. \u003cb\u003e26\u003c/b\u003e, 566\u0026ndash;581 (2007).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRajagopalan, V. et al. Quantitative in vivo MRI measurement of cortical development in the fetus. \u003cem\u003eBrain Struct. Funct.\u003c/em\u003e \u003cb\u003e217\u003c/b\u003e, 127\u0026ndash;139 (2012).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRajagopalan, V. et al. Local tissue growth patterns underlying normal fetal human brain gyrification quantified in utero. \u003cem\u003eJ. Neurosci.\u003c/em\u003e \u003cb\u003e31\u003c/b\u003e, 2878\u0026ndash;2887 (2011).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWright, R. et al. Automatic quantification of normal cortical folding patterns from fetal brain MRI. \u003cem\u003eNeuroimage\u003c/em\u003e \u003cb\u003e91\u003c/b\u003e, 21\u0026ndash;32 (2014).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHabas, P. A. et al. Early folding patterns and asymmetries of the normal human brain detected from in utero MRI. \u003cem\u003eCereb. Cortex\u003c/em\u003e. \u003cb\u003e22\u003c/b\u003e, 13\u0026ndash;25 (2012).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eIm, K. et al. Quantitative folding pattern analysis of early primary sulci in human fetuses with brain abnormalities. \u003cem\u003eAm. J. Neuroradiol.\u003c/em\u003e \u003cb\u003e38\u003c/b\u003e, 1449\u0026ndash;1455 (2017).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYehuda, B. et al. Automatic Quantification of Normal Brain Gyrification Patterns and Changes in Fetuses with Polymicrogyria and Lissencephaly Based on MRI. \u003cem\u003eAJNR Am. J. Neuroradiol.\u003c/em\u003e \u003cb\u003e44\u003c/b\u003e, 1432\u0026ndash;1439 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYehuda, B. et al. Reduced gyrification in fetal growth restriction with prenatal magnetic resonance images. \u003cem\u003eCereb Cortex\u003c/em\u003e \u003cb\u003e34\u003c/b\u003e, (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWong, M. \u003cem\u003eDiscrete Fourier Analysis\u003c/em\u003e Vol. 5 (Springer Science \u0026amp; Business Media, 2011).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDubois, J. et al. The dynamics of cortical folding waves and prematurity-related deviations revealed by spatial and spectral analysis of gyrification. \u003cem\u003eNeuroimage\u003c/em\u003e \u003cb\u003e185\u003c/b\u003e, 934\u0026ndash;946 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGermanaud, D. et al. Larger is twistier: Spectral analysis of gyrification (SPANGY) applied to adult brain size polymorphism. \u003cem\u003eNeuroimage\u003c/em\u003e \u003cb\u003e63\u003c/b\u003e, 1257\u0026ndash;1272 (2012).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHinich, M. J. \u0026amp; Clay, C. S. The application of the discrete Fourier transform in the estimation of power spectra, coherence, and bispectra of geophysical data. \u003cem\u003eReviews of Geophysics\u003c/em\u003e vol. 6 347\u0026ndash;363 Preprint at (1968). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/RG006i003p00347\u003c/span\u003e\u003cspan address=\"10.1029/RG006i003p00347\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHamed, H. A., Abdullah, A. K. \u0026amp; Al-Waisawy, S. \u003cem\u003eFrequency Hopping Spread Spectrum Recognition Based on Discrete Fourier Transform and Skewness and Kurtosis\u003c/em\u003e. \u003cem\u003eInternational J. Appl. Eng. Research\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e (2018). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.ripublication.com\u003c/span\u003e\u003cspan address=\"http://www.ripublication.com\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChen, H. et al. Gyral net: A new representation of cortical folding organization. \u003cem\u003eMed. Image Anal.\u003c/em\u003e \u003cb\u003e42\u003c/b\u003e, 14\u0026ndash;25 (2017).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGarel, C. et al. Fetal Cerebral Cortex: Normal Gestational Landmarks Identified Using Prenatal MR Imaging. \u003cem\u003eAJNR Am. J. Neuroradiol.\u003c/em\u003e \u003cb\u003e22\u003c/b\u003e, 184\u0026ndash;189 (2001).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDubois, J. et al. Primary cortical folding in the human newborn: An early marker of later functional development. \u003cem\u003eBrain\u003c/em\u003e \u003cb\u003e131\u003c/b\u003e, 2028\u0026ndash;2041 (2008).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLeventer, R. J. et al. Clinical and imaging features of cortical malformations in childhood. \u003cem\u003eNeurology\u003c/em\u003e \u003cb\u003e53\u003c/b\u003e, 715\u0026ndash;722 (1999).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLeventer, R. J. et al. Clinical and imaging heterogeneity of polymicrogyria: a study of 328 patients. \u003cem\u003eBrain\u003c/em\u003e \u003cb\u003e133\u003c/b\u003e, 1415\u0026ndash;1427 (2010).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePoon, L. C. et al. Transvaginal three-dimensional ultrasound assessment of Sylvian fissures at 18\u0026ndash;30 weeks\u0026rsquo; gestation. \u003cem\u003eUltrasound Obstet. Gynecol.\u003c/em\u003e \u003cb\u003e54\u003c/b\u003e, 190\u0026ndash;198 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGuibaud, L. et al. Abnormal Sylvian fissure on prenatal cerebral imaging: significance and correlation with neuropathological and postnatal data. \u003cem\u003eUltrasound Obstet. Gynecol.\u003c/em\u003e \u003cb\u003e32\u003c/b\u003e, 50\u0026ndash;60 (2008).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYushkevich, P. A. et al. User-guided 3D active contour segmentation of anatomical structures: significantly improved efficiency and reliability. \u003cem\u003eNeuroimage\u003c/em\u003e \u003cb\u003e31\u003c/b\u003e, 1116\u0026ndash;1128 (2006).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBorghi, E. et al. Construction of the World Health Organization child growth standards: Selection of methods for attained growth curves. \u003cem\u003eStatistics in Medicine\u003c/em\u003e vol. 25 247\u0026ndash;265 Preprint at (2006). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/sim.2227\u003c/span\u003e\u003cspan address=\"10.1002/sim.2227\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBenjamini, Y. \u0026amp; Hochberg Yosef. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testeing. \u003cem\u003eJ. Royal Stat. Soc. Ser. B (Metodological)\u003c/em\u003e. \u003cb\u003e57\u003c/b\u003e, 289\u0026ndash;300 (1995).\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":"Cortex, Gyrification, Fetal MRI, Spectral analysis, Lissencephaly, Polymicrogyria","lastPublishedDoi":"10.21203/rs.3.rs-8011139/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8011139/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eCortical gyrification is a key marker of fetal brain development and is typically assessed qualitatively on ultrasound or MRI. While previous quantitative approaches have characterized gestational trajectories in typically developing (TD) fetuses, only a few studies have investigated cortical malformations such as lissencephaly and polymicrogyria. Spectral analysis, which characterizes signals by their frequency content, has been successfully applied to study gyrification in neonates and adults but has not yet been explored prenatally. In this study, we introduce a spectral framework for quantifying fetal cortical folding from routine fetal MRI. Cerebral contours were extracted from coronal slices, transformed into polar coordinates, and analyzed using Fourier Transform to derive spectral profiles and five gyrification features: non-zero spectral density, entropy, mean frequency, variance, and skewness and the first twelve frequencies. Seventy-three TD fetuses and twenty-four with malformations of cortical development (14 polymicrogyria, 10 lissencephaly) were evaluated across gestation. Differences between TD, lissencephaly, or polymicrogyria fetuses were evaluated using linear mixed models and post-hoc t-tests with Benjamini\u0026ndash;Hochberg correction. In TD fetuses, spectral features showed gestational-age\u0026ndash;related trajectories, with increasing spectral density and variance and decreasing skewness, corresponding to the sequential folding waves. Fetuses with cortical malformations had lower spectral density and entropy (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026le;\u0026thinsp;0.031), and reduction in most of the twelve frequencies, most prominently in frequencies associated with the Sylvian fissure development (\u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;0.001). Spectral representation may capture both global and local aspects of cortical folding, offering a robust and quantitative biomarker of fetal brain maturation and deviations in cortical development.\u003c/p\u003e","manuscriptTitle":"MRI-Based Spectral Analysis of Fetal Brain Gyrification: Applications to Lissencephaly and Polymicrogyria","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-21 01:07:00","doi":"10.21203/rs.3.rs-8011139/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-12-08T09:58:19+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-06T17:23:16+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-02T12:36:35+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-24T10:00:29+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"69534803281636641111556112551659848573","date":"2025-11-12T19:07:53+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"82361473803136622555495805714633956216","date":"2025-11-12T05:11:10+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"160880715615080128360225055915430560196","date":"2025-11-11T08:14:45+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"216953271512594816132787208456102148613","date":"2025-11-10T18:48:48+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-11-10T17:06:02+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-11-10T14:12:32+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-11-10T10:57:55+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-11-07T14:30:14+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-11-07T14:26:45+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":"3629b071-acb8-4099-986e-7bc73bc9525f","owner":[],"postedDate":"November 21st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":57894810,"name":"Biological sciences/Biological techniques"},{"id":57894811,"name":"Health sciences/Biomarkers"},{"id":57894812,"name":"Health sciences/Medical research"},{"id":57894813,"name":"Biological sciences/Neuroscience"}],"tags":[],"updatedAt":"2026-02-23T16:03:35+00:00","versionOfRecord":{"articleIdentity":"rs-8011139","link":"https://doi.org/10.1038/s41598-026-38229-9","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2026-02-20 15:57:17","publishedOnDateReadable":"February 20th, 2026"},"versionCreatedAt":"2025-11-21 01:07:00","video":"","vorDoi":"10.1038/s41598-026-38229-9","vorDoiUrl":"https://doi.org/10.1038/s41598-026-38229-9","workflowStages":[]},"version":"v1","identity":"rs-8011139","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8011139","identity":"rs-8011139","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}