Orthogonal Projection Research Articles

Norm of the sum of two orthogonal projections

Norm of the sum of two orthogonal projections

Associations of Plasma and Fecal Metabolites with Body Mass Index and Body Fat Distribution in Children.

Childhood obesity continues to be a critical public health concern with far-reaching implications for the well-being. This study aimed to investigate the association between metabolites in plasma and feces and indicators including body mass index (BMI), BMI for age Z score (BMIZ), and body fat distribution among children aged 6-9 years in China. This cross-sectional study enrolled 424 healthy children, including 186 girls and 238 boys. Dual-energy X-ray absorptiometry (DXA) was used to determine the body fat content and regional fat distribution. Plasma and fecal metabolites were analyzed using targeted metabolomic technologies. A total of 200 plasma metabolites and 212 fecal metabolites were accurately quantified via ultra-performance liquid chromatography coupled with tandem mass spectrometry (UPLC-MS/MS). By using Orthogonal Projections to Latent Structures Discriminant Analysis (OPLS-DA) and random forest model, we discovered that 9 plasma metabolites and 11 fecal metabolites were associated with different weight statuses. After adjusting for potential covariates and false discovery rate (FDR) correction, multiple linear regression analyses revealed that plasma metabolites (fumaric acid, glycine, l-glutamine, methylmalonic acid, and succinic acid) and fecal metabolites (protocatechuic acid) were negatively associated (β: -1.373--0.016, pFDR: <0.001-0.031; β: -1.008--0.071, pFDR: 0.005-0.033), while plasma metabolites (isovaleric acid, isovalerylcarnitine, l-glutamic acid, and pyroglutamic acid) and fecal metabolites (3-aminoisobutanoic acid, butyric acid, N-acetylneuraminic acid, octanoylcarnitine, oleoylcarnitine, palmitoylcarnitine, stearoylcarnitine, taurochenodesoxycholic acid, and taurodeoxycholic acid) exhibited positive associations with BMI, BMIZ, and body fat distribution (β: 0.023-2.396, pFDR: <0.001; β: 0.014-1.736, pFDR: <0.001-0.049). Plasma and fecal metabolites such as glutamine may serve as a potential therapeutic target for the development of obesity.

Abstract PO2-28-11: How to improve margins in breast cancer surgery: a

Abstract Objective: Intraoperative specimen mammography (ISM) is a diffuse technique that allows surgeons to check specimens immediately after lumpectomy, instead of intraoperative frozen sections, definitely abandoned. in Italy and Europe, because it is regarded as too expensive and time-consuming examination. Although the specimen is slightly compressed, the radiological image can be distorted by tissue overlap, and this may affect the evaluation of tumor borders, resulting in extension of the lumpectomy. As ISM may be less precise due to inadequate compression, a vacuum effect was applied to the specimen to increase the precision of margin detection. Study design: This study was conducted at St. Anna Hospital Breast Unit, Turin, Italy. Women who underwent lumpectomy for cancer were eligible for inclusion. BCS resection was performed with a free margin, and the surgical specimen was imaged immediately by SSM and evaluated by an expert radiologist and surgeon.Inclusion criteria were: breast cancer detected by a screening test, and treatment with surgical lumpectomy. Both standard ISM (sISM) and vacuum ISM (vISM) were performed. A dedicated breast surgeon at the study institution performed both surgery and radiological scans, and their interpretation. A dedicated radiologist reviewed all of the images. ISM was performed by a digital specimen mammography system (Faxitron, Bioptics Inc., Tucson, AZ, USA) in the operating room. Specimen registration and ISM analysis took 3 min. sISM (two orthogonal projections) was performed. Specimens were sealed in plastic bags in the vacuum apparatus (TissueSAFE Milestone Medical, Sorisole, Italy); this process took a few seconds. 152 specimens obtained after lumpectomy from 1 April 2021 to 31 April 2022 were scanned. sISM (two orthogonal projections) was performed. Next, the specimen was placed in a vacuum bag, and vISM was performed. Additional tissue was removed if the surgeon considered that excision was inadequate. Finally, the specimen was sent for definitive histopathological analysis, which is the gold standard for the assessment of surgical margins. Intraoperative histological margin assessment was not performed. The sISM and vISM images and final histopathology reports were compared. A first experience on a limited number of patients to demonstrate feasibility of the technique, is reported in table 1 and 2. Results: For sISM, specificity was 43,85% [95 % confidence interval (CI) 35.96-51.73], sensitivity was 40.91% (95 % CI 33.09-48.73), positive predictive value (PPV) 10.98% (95 % CI 6.01-15.95) and negative predictive value (NPV) 81.43% (95 % CI 75.25-87.61). For vISM, specificity 92,31% (95 % CI 88,07-96.54), sensitivity 72.73% (95 % CI 65.65-79.81), PPV 61.54% (95 % CI 53.80-69.27) and NPV 95.24% (95 % CI 91.85-98.62). Conclusion: ISM is a valid, safe and quicker alternative to intraoperative histopathological study, but has been reported to be limited due to less compression than standard SSM. Consequently, ISM may result in a high percentage of false-positive results, meaning unnecessary wider resections that lead to increased operative times and worse aesthetic results. A vacuum effect was performed on the specimen to increase the precision of margin detection. vISM images seem to be easier to interpret than sISM images (Figs. 1–3) as they are characterized by a vacuum-created radiolucent rim that better defines the tumor margins. These data suggest that the vacuum technique is feasible, cost-saving and yields results that are similar to those from frozen sections but without the limitations, such as prolonged operating time, high variability in sensitivity due to pathologists abilities, risk of compromising the histology, and unreliability for small lumps and ductal carcinoma in situ. Keywords: Breast cancer; Intraoperative mammography; Lumpectomy; Specimen mammography; Vacuum intraoperative mammography. Table 1. Clinical Data IC, invasive cancer; DCIS ductal carcinoma in situ; ADH, atypical ductal hyperplasia; ILC, invasive lobular cancer; B3, lesions of uncertain malignant potential; IMPC, invasive micropapillary carcinoma; NST, no special type; sISM, standard intraoperative specimen mammography; vISM, vacuum intraoperative specimen mammography. ª Margin: 0, involved; 1, free Table 2. Figure 1 - 2 - 3 Standard intraoperative specimen mammography (left) and vacuum intraoperative specimen mammography (right). Citation Format: Maria Grazia Baù, Vincenzo Marra, Alessandra Surace, Donatella Tota, Maria Piera Mano, Aurelia Mondino, Alessandro Bottero, Elisabetta Robba. How to improve margins in breast cancer surgery: a [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PO2-28-11.

On Orthogonal Projections of Symplectic Balls

On Orthogonal Projections of Symplectic Balls

Validating Thai translations of perceptual soundscape attributes: A non-Procrustean approach with a Procrustes projection

Measurement of a psychological construct across populations without a common linguistic medium often necessitates the development of multiple translations of the psychometric tool across multiple languages, dialects, or other population-specific variations. In this follow-up (Stage 2) study, a listening test using a shared set of 27 stimuli from the Soundscape Attribute Translation Project (SATP) was conducted with Thai-speaking participants using the set of Thai translations of the eight perceptual affective quality (PAQ) descriptors selected in earlier (Stage 1) work through a structured evaluation questionnaire. Principal component analysis was performed on the listening test data to obtain a rank-two reduction of the responses with maximal explained variance. In order to align the principal component space to the two-dimensional circumplex space, this work presents a simple and numerically stable method based on the orthogonal Procrustes projection, to find the optimal two-dimensional orthogonal transform that aligns the first two principal components with the axes corresponding to Pleasantness and Eventfulness as defined in ISO/TS 12913-3:2019. Analysis of the listening test responses indicated good to excellent interrater reliability, reflecting the general comprehensibility of the translations to laypersons. Subsequent analyses yielded a two-dimensional projection with ▪ explained variance and near-perfect alignment of the composite Pleasantness and Eventfulness axes. Angular locations of the individual translated PAQs are located within ▪ of their theoretically ideal locations and preserve angular ordering, albeit with imperfect equiangularity. Cross-analysis against the results from Stage 1 showed that the structured evaluation may be partially useful in anticipating potential imperfections of the PAQ translations and their angular locations in Stage 2.

Chemical impurity profiling: Linkage of starting materials and an intermediate synthesis product of a carbamate chemical warfare agent

Establishing links between starting materials and products is highly valuable in the investigation of the use of toxic chemicals for illicit purposes. In this study, impurity profiling was performed on starting materials and their synthesis product, 2-([dimethylamino]methyl)pyridin-3-yl dimethylcarbamate, an intermediate compound in the production route for the carbamate class of Chemical Warfare Agents. The aim was to link the five commercial starting materials to the correct synthesis products. Initially, the intermediate compound was synthesized using different batches of the two starting materials (2 plus 3 batches), producing six unique combinations. All synthesis batches and the different starting materials were analysed by gas chromatography-high resolution mass spectrometry (GC-HRMS). Chemometrics analyses were conducted with principal component analysis and orthogonal projections to latent structures discriminant analysis to extract chemical impurity profiles and to build supervised classification models.Additionally, 12 test set samples, produced using the same starting materials by two different chemists in another laboratory, were analysed by GC-HRMS. A classification model able to distinguish all supplier combinations was successfully created and used to link the test set samples to their corresponding starting material.Furthermore, a new set of synthesis samples was extracted with a work-up procedure before analysis to investigate the effect of higher sample purity on the classification model. The results show that linking the synthesis products to their starting materials was successful for one of the starting materials, despite the purification procedure.

IModel: an interactive 3D crystal structure visualization program

This work presents present iModel, an interactive 3D crystal structure visualization program capable of reading crystal structure datafiles in various formats and rendering corresponding 3D models. Users can interact with the models by translating, rotating and zooming in on them. The software visualizes crystal structures, including atoms, bonds, unit cells and polyhedral shapes using spheres, lines and polyhedra, with customizable colors and sizes for individual atoms and bonds. The software features perspective and orthogonal projections, multiple chemical bond search modes, and a user-friendly toolbar for view control, and exports models as images or saves progress in project files. Developed on the LabVIEW platform with Python extensions, the iModel program leverages a queue message handler and producer–consumer pattern for a modular, reusable and powerful crystal structure visualization tool.

Intermediate dimensions under self-affine codings

Intermediate dimensions were recently introduced by Falconer et al. (Math Z 296:813–830, 2020) to interpolate between the Hausdorff and box-counting dimensions. In this paper, we show that for every subset E of the symbolic space, the intermediate dimensions of the projections of E under typical self-affine coding maps are constant and given by formulas in terms of capacities. Moreover, we extend the results to the generalized intermediate dimensions introduced by Banaji (Monatsh Math 202: 465–506, 2023) in several settings, including the orthogonal projections in Euclidean spaces and the images of fractional Brownian motions.

Reconstruction method suitable for fast CT imaging

Reconstructing computed tomography (CT) images from an extremely limited set of projections is crucial in practical applications. As the available projections significantly decrease, traditional reconstruction and model-based iterative reconstruction methods become constrained. This work aims to seek a reconstruction method applicable to fast CT imaging when available projections are highly sparse. To minimize the time and cost associated with projections acquisition, we propose a deep learning model, X-CTReNet, which parameterizes a nonlinear mapping function from orthogonal projections to CT volumes for 3D reconstruction. The proposed model demonstrates effective capability in inferring CT volumes from two-view projections compared to baseline methods, highlighting the significant potential for drastically reducing projection acquisition in fast CT imaging.

On the maximal hyperplane in [formula omitted]

In this paper we introduce the family of projections Pf,α:ℓpn→ker⁡fPf,α=Id−1∑i=1n|fi|αf⊗〈f〉α−1 and study their maximal p-norms. Note that α=2 and α=q corresponds to orthogonal and polar projections respectively. For fixed α>1, we are able to compute maxf⁡||Pf,α||p and show that, for α≠q, any maximizer f has at most two non-zero coordinates different. For α=q there are more maximizers but we can always choose one that has at most two non-zero coordinates different. Based on that we compute orthogonal λorthH(ℓpn) and polar λpolH(ℓpn) hyperplane constants of ℓpn. Furthermore, the general formula for maxf⁡||Pf,α||p allows us to show that for n=3 and any hyperplane H=ker⁡fλ(H=ker⁡f,ℓpn)≤λ(ker⁡1,ℓpn), i.e., ker⁡1 is a maximal hyperplane in ℓp3, confirming the n=3 case of the long standing conjecture.

Eigenvalue estimates for Fourier concentration operators on two domains.

We study concentration operators associated with either the discrete or the continuous Fourier transform, that is, operators that incorporate a spatial cut-off and a subsequent frequency cut-off to the Fourier inversion formula. The spectral profiles of these operators describe the number of prominent degrees of freedom in problems where functions are assumed to be supported on a certain domain and their Fourier transforms are known or measured on a second domain. We derive eigenvalue estimates that quantify the extent to which Fourier concentration operators deviate from orthogonal projectors, by bounding the number of eigenvalues that are away from 0 and 1 in terms of the geometry of the spatial and frequency domains, and a factor that grows at most poly-logarithmically on the inverse of the spectral margin. The estimates are non-asymptotic in the sense that they are applicable to concrete domains and spectral thresholds, and almost match asymptotic benchmarks. Our work covers, for the first time, non-convex and non-symmetric spatial and frequency concentration domains, as demanded by numerous applications that exploit the expected approximate low dimensionality of the modeled phenomena. The proofs build on Israel's work on one dimensional intervals arXiv:1502.04404v1. The new ingredients are the use of redundant wave-packet expansions and a dyadic decomposition argument to obtain Schatten norm estimates for Hankel operators.

Revealing the 1H-NMR Profiling of Six Edible Mushrooms Consumed in the Northeastern Highlands of Puebla, Mexico.

The 1H-NMR metabolomics profiling of six edible mushrooms consumed in the northeastern highlands of Puebla, Mexico is presented. These fungi were morpho- and molecularly identified as Infundibulicybe squamulosa, Amanita jacksonii, Lepista nuda, Russula delica, Russula brevipes, and Lactarius indigo. The chemical profiling confirmed the presence of eight essential amino acids and their derivatives, six organic acids, six nucleosides, low amounts of reducing sugars, and valuable nutraceuticals such as betaine, carnitine, glycero-3-phosphocholine and O-acetylcarnitine which were differentially determined and quantified in the six mushrooms by qNMR. Principal component analysis (PCA) and orthogonal projections to latent structures discriminant analysis (OPLS-DA) generated four different groups. Two of these groups were constituted by fungal species with phylogenic relationships whereas non-phylogenetic related species were separated from each other. The potential use of 1H-NMR metabolomics and chemometrics to group macromycetes and determine the nutritional and nutraceutical potential of these local foods is demonstrated.

Urinary metabolomic profiles uncover metabolic pathways in children with asthma

Objective The prevalence of asthma has gradually increased worldwide in recent years, which has made asthma a global public health problem. However, due to its complexity and heterogeneity, there are a few academic debates on the pathogenic mechanism of asthma. The study of the pathogenesis of asthma through metabolomics has become a new research direction. We aim to uncover the metabolic pathway of children with asthma. Methods Liquid chromatography (LC)–mass spectrometry (MS)–based metabolomic analysis was conducted to compare urine metabolic profiles between asthmatic children (n = 30) and healthy controls (n = 10). Results Orthogonal projections to latent structures-discrimination analysis (OPLS-DA) showed that there were significant differences in metabolism between the asthma group and the control group with three different metabolites screened out, including traumatic acid, dodecanedioic acid, and glucobrassicin, and the levels of traumatic acid and dodecanedioic acid in the urine samples of asthmatic children were lower than those of healthy controls therein. Pathway enrichment analysis of differentially abundant metabolites suggested that α-linolenic acid metabolism was an asthma-related pathway. Conclusions This study suggests that there are significant metabolic differences in the urine of asthmatic children and healthy controls, and α-linolenic acid metabolic pathways may be involved in the pathogenesis of asthma.

XTransCT: ultra-fast volumetric CT reconstruction using two orthogonal x-ray projections for image-guided radiation therapy via a transformer network

Objective. The aim of this study was to reconstruct volumetric computed tomography (CT) images in real-time from ultra-sparse two-dimensional x-ray projections, facilitating easier navigation and positioning during image-guided radiation therapy. Approach. Our approach leverages a voxel-sapce-searching Transformer model to overcome the limitations of conventional CT reconstruction techniques, which require extensive x-ray projections and lead to high radiation doses and equipment constraints. Main results. The proposed XTransCT algorithm demonstrated superior performance in terms of image quality, structural accuracy, and generalizability across different datasets, including a hospital set of 50 patients, the large-scale public LIDC-IDRI dataset, and the LNDb dataset for cross-validation. Notably, the algorithm achieved an approximately 300% improvement in reconstruction speed, with a rate of 44 ms per 3D image reconstruction compared to former 3D convolution-based methods. Significance. The XTransCT architecture has the potential to impact clinical practice by providing high-quality CT images faster and with substantially reduced radiation exposure for patients. The model's generalizability suggests it has the potential applicable in various healthcare settings.

Correction: Relating moments of self-adjoint polynomials in two orthogonal projections

Correction: Relating moments of self-adjoint polynomials in two orthogonal projections

Distinct Changes in Metabolic Profile and Sensory Quality with Different Varieties of Chrysanthemum (Juhua) Tea Measured by LC-MS-Based Untargeted Metabolomics and Electronic Tongue.

Chrysanthemum tea, a typical health tea with the same origin as medicine and food, is famous for its unique health benefits and flavor. The taste and sensory quality of chrysanthemum (Juhua) tea are mainly determined by secondary metabolites. Therefore, the present research adopted untargeted metabolomics combined with an electronic tongue system to analyze the correlation between the metabolite profiles and taste characteristics of different varieties of chrysanthemum tea. The results of sensory evaluation showed that there were significant differences in the sensory qualities of five different varieties of chrysanthemum tea, especially bitterness and astringency. The results of principal component analysis (PCA) indicated that there were significant metabolic differences among the five chrysanthemum teas. A total of 1775 metabolites were identified by using untargeted metabolomics based on UPLC-Q-TOF/MS analysis. According to the variable importance in projection (VIP) values of the orthogonal projections to latent structures discriminant analysis (OPLS-DA), 143 VIP metabolites were found to be responsible for metabolic changes between Huangju and Jinsi Huangju tea; among them, 13 metabolites were identified as the key metabolites of the differences in sensory quality between them. Kaempferol, luteolin, genistein, and some quinic acid derivatives were correlated with the "astringency" attributes. In contrast, l-(-)-3 phenyllactic acid and L-malic acid were found to be responsible for the "bitterness" and "umami" attributes in chrysanthemum tea. Kyoto Encyclopedia of Genes and Genomes pathway enrichment analysis showed that the flavonoid and flavonol biosynthesis pathways had important effects on the sensory quality of chrysanthemum tea. These findings provide the theoretical basis for understanding the characteristic metabolites that contribute to the distinctive sensory qualities of chrysanthemum tea.

GCMSFormer: A Fully Automatic Method for the Resolution of Overlapping Peaks in Gas Chromatography-Mass Spectrometry.

Gas chromatography-mass spectrometry (GC-MS) is one of the most important instruments for analyzing volatile organic compounds. However, the complexity of real samples and the limitations of chromatographic separation capabilities lead to coeluting compounds without ideal separation. In this study, a Transformer-based automatic resolution method (GCMSFormer) is proposed to resolve mass spectra from GC-MS peaks in an end-to-end manner, predicting the mass spectra of components directly from the raw overlapping peaks data. Furthermore, orthogonal projection resolution (OPR) was integrated into GCMSFormer to resolve minor components. The GCMSFormer model was trained, validated, and tested using 100,000 augmented data. It achieves 99.88% of the bilingual evaluation understudy (BLEU) value on the test set, significantly higher than the 97.68% BLEU value of the baseline sequence-to-sequence model long short-term memory (LSTM). GCMSFormer was also compared with two nondeep learning resolution tools (MZmine and AMDIS) and two deep learning resolution tools (PARAFAC2 with DL and MSHub/GNPS) on a real plant essential oil GC-MS data set. Their resolution results were compared on evaluation metrics, including the number of compounds resolved, mass spectral match score, correlation coefficient, explained variance, and resolution speed. The results demonstrate that GCMSFormer has better resolution performance, higher automation, and faster resolution speed. In summary, GCMSFormer is an end-to-end, fast, fully automatic, and accurate method for analyzing GC-MS data of complex samples.

Characteristics of kopyor coconut (Cocos nucifera L.) using sensory analysis and metabolomics-based approach

Kopyor is a coconut with unique characteristics from Indonesia, one of the largest coconut producers in the world. Kopyor is an edible mature coconut with soft endosperm. Although this fruit is one of the most popular coconuts in the world, there are limited studies on its properties, including its sensory attributes and metabolite profiles. This study investigates the characteristics of kopyor using sensory evaluation, a widely targeted metabolomics approach, and multivariate analysis. The liquid (water) and solid (flesh) endosperms were collected as the samples. The results showed that kopyor has characteristics that distinguish it from normal mature and young coconuts. Kopyor water has a milky, creamy, nutty, bitter, and astringent taste with an oily aftertaste and mouthfeel. Kopyor flesh is soft and moist and gives a sandy mouth feel. This study analyzed the sensory attributes of the kopyor endosperm for the first time and compared it with those of normal mature and young coconuts. A gas chromatography mass spectrometry analysis showed that kopyor contained wider variety of metabolites than normal coconuts of the same age. Based on the differential analysis and orthogonal projections to latent structures-regression, kopyor water was characterized by the accumulation of flavor-related metabolites, such as amino acids and organic acids, which contributed to its sensory complexity. This study solidified the effects of maturation and endosperm type on metabolite accumulation in kopyor endosperm. This pioneering information will lead to the future use of kopyor and other unique coconuts worldwide for food, contributing to the sustainability of the coconut industry.

Using clustering as pre-processing in the framework of signal unmixing for exhaustive exploration of archaeological artefacts in Raman imaging

Analytical chemistry on archaeological material is an essential part of modern archaeological investigations and from year to year, instrumental improvement has made it possible to generate data at a high spatial and temporal frequency. In particular, Raman spectral imaging can be successfully applied in archaeological research by its simplicity of implementation to study past human societies through the analysis of their material remains. This technique makes it possible to simultaneously obtain spatial and spectral information by preserving sample integrity. However, because of the inherent complexity of the samples in Archaeology (e.g. seniority, fragility, lack or full absence of any information about its composition), chemical interpretation can be difficult at first glance. Indeed, specific problems of spectral selectivity related to unexpected chemical compounds could appear due to their state of conservation. Furthermore, detecting minor compounds becomes challenging as major components impose their contributions in the acquired spectra. Therefore, a relevant chemometric approach has been introduced in this context to characterize distinct spectral sources in a Raman imaging dataset of an archaeological specimen – a mosaic fragment. The fragment was unearthed during the Ruscino archaeological dig on the outskirts of Perpignan, France. It dates back to the oppidum period. The aim is to extract selective spectral information from pixel clustering analysis in order to enhance the initial optimisation step within the Multivariate Curve Resolution and Alternating Least-Squares (MCR-ALS) algorithm, a well-known signal unmixing technique. The underlying principle of the MCR-ALS is that the acquired spectra can be expressed as linear combinations of pure spectra of all individual components present in the chemical system under study. Sometimes it can be difficult to obtain the desired results through the algorithm, particularly if initial estimates of spectral or concentration profiles are inaccurate due to complex signals, noise or lack of selectivity, resulting in rank deficiency (i.e. a poor estimation of the total number of pure signals). For this reason, an innovative threshold-based clustering algorithm, combined with multiple Orthogonal Projection Approaches (OPA), has been developed to improve matrix rank investigation and thus the initialisation step of the MCR-ALS approach before optimisation. The effective analysis of Raman imaging data for an archaeological mosaic played a crucial role in uncovering significant chemical information about a particular biogenic material. This insight sheds light on the origins of mortar manufacture during the oppidum period.

Attenuated Total Reflectance–Fourier Transform Infrared (ATR-FTIR) Spectroscopy Combined With Deep Learning for Classification of Idiopathic Recurrent Spontaneous Miscarriage (IRSM)

Idiopathic recurrent spontaneous miscarriage (IRSM) is characterized by the repeated losses of two or more clinically detected pregnancies before 24 week of gestation with the cause remaining unknown. We aim to ascertain whether attenuated total reflectance-Fourier transform infrared (ATR-FTIR) spectroscopy with deep learning (DL) can accurately distinguish IRSM from controls. Endometrial tissue (internal lining of the uterine cavity) was collected from 20 women with IRSM and 20 controls during window of implantation. ATR-FTIR spectroscopy was used to acquire spectra (400–4000 cm−1) from three random locations across each specimen and averaged. Multivariate analysis was used to assess group differentiation. Furthermore, DL classifiers were implemented to predict IRSM. Hierarchical cluster analysis (HCA) demonstrated distinct clustering of the two groups. The principal component analysis (PCA) loading plot indicated contributions from amide I, II and III bands, symmetric methyl (CH3) bending of proteins, lipids, nucleic acids, asymmetric stretching of CH2 due to lipids, collagen, and glucose in discriminating between IRSM and controls. Supervised orthogonal projections to latent structures discriminant analysis (OPLS-DA) exhibited optimized discrimination between the groups. Convolutional neural network (CNN) and variational autoencoder-artificial neural network (VAE-ANN) classifiers exhibited good performance and showed classification accuracy of 94% and 94% and F1 scores (mean of recall and precision) of 95% and 93%, respectively. The optimal area under the receiver operating characteristic curve (AUC) value of both CNN and VAE-ANN was 0.94. ATR-FTIR spectroscopy combined with DL approach is promising for understanding complex disease pathophysiology and enabling early prediction.