Afr J Reprod Health. 2025 Apr 23;29(4):108-119. doi: 10.29063/ajrh2025/v29i4.10.

ABSTRACT

Due to physiological changes during the perimenopausal period, the morphology of cervical cells undergoes certain alterations. Accurate cell image segmentation and lesion identification are of great significance for the early detection of precancerous lesions. Traditional detection methods may have certain limitations, thereby creating an urgent need for the development of more effective models. This study aimed to develop a highly efficient and accurate cervical cell image segmentation and recognition model to enhance the detection of precancerous lesions in perimenopausal women. based on U-shaped Network(U-Net) and Residual Network (ResNet). The model integrates U-Net with Segmentation Network (SegNet) and incorporates the Squeeze-and-Excitation (SE) attention mechanism to create the 2Se/U-Net segmentation model. Additionally, ResNet is optimized with the local discriminant loss function (LD-loss) and deep residual learning (DRL) blocks to develop the LD/ResNet lesion recognition model. The performance of the models is evaluated using data from 103 cytology images of perimenopausal women, focusing on segmentation metrics like mean pixel accuracy (MPA) and mean intersection over union (mIoU), as well as lesion detection metrics such as accuracy (Acc), precision (Pre), recall (Re), and F1-score (F1). Results show that the 2Se/U-Net model achieves an MPA of 92.63% and mIoU of 96.93%, outperforming U-Net by 12.48% and 9.47%, respectively. The LD/ResNet model demonstrates over 97.09% accuracy in recognizing cervical cells and achieves high detection performance for precancerous lesions, with Acc, Pre, and Re at 98.95%, 99.36%, and 98.89%, respectively. The model shows great potential for enhancing cervical cancer screening in clinical settings.

PMID:40314307 | DOI:10.29063/ajrh2025/v29i4.10

Elife. 2025 May 2;13:RP98033. doi: 10.7554/eLife.98033.

ABSTRACT

Protein engineering is a pivotal aspect of synthetic biology, involving the modification of amino acids within existing protein sequences to achieve novel or enhanced functionalities and physical properties. Accurate prediction of protein variant effects requires a thorough understanding of protein sequence, structure, and function. Deep learning methods have demonstrated remarkable performance in guiding protein modification for improved functionality. However, existing approaches predominantly rely on protein sequences, which face challenges in efficiently encoding the geometric aspects of amino acids' local environment and often fall short in capturing crucial details related to protein folding stability, internal molecular interactions, and bio-functions. Furthermore, there lacks a fundamental evaluation for developed methods in predicting protein thermostability, although it is a key physical property that is frequently investigated in practice. To address these challenges, this article introduces a novel pre-training framework that integrates sequential and geometric encoders for protein primary and tertiary structures. This framework guides mutation directions toward desired traits by simulating natural selection on wild-type proteins and evaluates variant effects based on their fitness to perform specific functions. We assess the proposed approach using three benchmarks comprising over 300 deep mutational scanning assays. The prediction results showcase exceptional performance across extensive experiments compared to other zero-shot learning methods, all while maintaining a minimal cost in terms of trainable parameters. This study not only proposes an effective framework for more accurate and comprehensive predictions to facilitate efficient protein engineering, but also enhances the in silico assessment system for future deep learning models to better align with empirical requirements. The PyTorch implementation is available at https://github.com/ai4protein/ProtSSN.

PMID:40314227 | DOI:10.7554/eLife.98033

J Mammary Gland Biol Neoplasia. 2025 May 2;30(1):8. doi: 10.1007/s10911-025-09580-w.

ABSTRACT

Circular RNAs (circRNAs) have emerged as critical regulators in various biological processes including diseases. In the mammary gland (MG), which undergoes most of its development postnatally, circRNAs play pivotal roles in both physiological and pathological contexts. This review highlights the involvement of circRNAs during key developmental stages of the MG, with particular emphasis on lactation, where circRNA-miRNA networks significantly influence milk secretion and composition. CircRNAs exhibit stage-, breed- and species-specific expression patterns during lactation, which underscores their complexity. This intricate regulation also plays a significant role in pathological conditions of the MG, where dysregulated circRNA expression contributes to disease progression such as mastitis, early breast cancer (BC) stages, and epithelial-to-mesenchymal transition in BC (EMT). In mastitis, altered circRNA expression disrupts immune responses and compromises epithelial integrity. During early BC progression, circRNAs drive cell proliferation, while in EMT, they facilitate metastatic processes. By focusing on the circRNA-miRNA interactions underlying these processes, this review highlights their potential use as biomarkers for MG development, disease progression, and as therapeutic targets.

PMID:40314719 | DOI:10.1007/s10911-025-09580-w

Anal Chem. 2025 May 2. doi: 10.1021/acs.analchem.4c06809. Online ahead of print.

ABSTRACT

N-heterocyclic carbenes (NHCs) are versatile catalysts for organic reactions, characterized by their unique electron-donating properties and high activity. This study introduces NHCs as innovative derivatization reagents for liquid chromatography-mass spectrometry (LC-MS) analysis of aliphatic aldehydes. Five distinct NHC reagents were evaluated, and 2-mesityl-2,5,6,7-tetrahydropyrrolo[2,1-c][1,2,4]triazol-4-ium chloride (MTPTC) was identified as the most promising candidate due to its rapid reaction kinetics, high selectivity, and excellent product stability. The MTPTC-based derivatization reaction effectively addressed the issue of stereoisomeric products, resulting in well-resolved single peaks in the LC separation. Additionally, the derivatized products exhibited high stability, facilitating accurate and reliable quantitative analysis. Using MTPTC as the derivatization reagent, an LC-MS quantitative analysis strategy was developed for the determination of eight aliphatic aldehydes in human sera. The method demonstrated a broad linear range, low limits of detection and quantification, and satisfactory reproducibility and accuracy. The applicability of this method was further validated through the quantification of aliphatic aldehydes in the serum of sepsis patients. This work extends NHCs' utility to analytical chemistry and introduces a novel derivatization reagent for the analysis of carbonyl compounds by LC-MS.

PMID:40314613 | DOI:10.1021/acs.analchem.4c06809

mSystems. 2025 May 2:e0002125. doi: 10.1128/msystems.00021-25. Online ahead of print.

ABSTRACT

Gihawi et al. (mBio 14:e01607-23, 2023, https://doi.org/10.1128/mbio.01607-23) argued that the analysis of tumor-associated microbiome data by Poore et al. (Nature 579:567-574, 2020, https://doi.org/10.1038/s41586-020-2095-1) is invalid because features that were originally very sparse (genera with mostly zero read counts) became associated with the phenotype following batch correction. Here, we examine whether such an observation should necessarily indicate issues with processing or machine learning pipelines. We show counterexamples using the centered log ratio (CLR) transformation, which is often used for analysis of compositional microbiome data. The CLR transformation has similarities to voom-SNM, the batch-correction method brought into question by Gihawi et al., and yet is a sample-wise operation that cannot, in itself, "leak" information or invalidate downstream analyses. We show that because the CLR transformation divides each value by the geometric mean of its sample, common imputation strategies for missing or zero values result in transformed features that are associated with the geometric mean. Through analyses of both synthetic and vaginal microbiome data sets, we demonstrate that when the geometric mean is associated with a phenotype, sparse and CLR-transformed features will also become associated with it. We re-analyze features highlighted by Gihawi et al. and demonstrate that the phenomenon of sparse features becoming phenotype-associated can also be observed after a CLR transformation, which serves as a counterexample to the claim that such an observation necessarily means information leakage. While we do not intend to address other concerns regarding tumor microbiome analyses, validate Poore et al.'s results, or evaluate batch-correction pipelines, we conclude that because phenotype-associated features that were initially sparse can be created by a sample-wise transformation that cannot artifactually inflate machine learning performance, their detection is not independently sufficient to demonstrate information leakage in machine learning pipelines. Microbiome data are multivariate, and as such, a value of 0 carries a different meaning for each sample. Many transformations, including CLR and other batch-correction methods, are likewise multivariate, and, as these issues demonstrate, each individual feature should be interpreted with caution.

IMPORTANCE: Gihawi et al. claim that finding that a transformation turned highly sparse (mostly zero) features into features that are associated with a phenotype is sufficient to conclude that there is information leakage and to invalidate an analysis. This claim has critical implications for both the debate regarding The Cancer Genome Atlas (TCGA) cancer microbiome analysis and for interpretation and evaluation of analyses in the microbiome field at large. We show by counterexamples and by reanalysis that such transformations can be valid.

PMID:40314439 | DOI:10.1128/msystems.00021-25

Elife. 2025 May 2;13:RP98033. doi: 10.7554/eLife.98033.

ABSTRACT

Protein engineering is a pivotal aspect of synthetic biology, involving the modification of amino acids within existing protein sequences to achieve novel or enhanced functionalities and physical properties. Accurate prediction of protein variant effects requires a thorough understanding of protein sequence, structure, and function. Deep learning methods have demonstrated remarkable performance in guiding protein modification for improved functionality. However, existing approaches predominantly rely on protein sequences, which face challenges in efficiently encoding the geometric aspects of amino acids' local environment and often fall short in capturing crucial details related to protein folding stability, internal molecular interactions, and bio-functions. Furthermore, there lacks a fundamental evaluation for developed methods in predicting protein thermostability, although it is a key physical property that is frequently investigated in practice. To address these challenges, this article introduces a novel pre-training framework that integrates sequential and geometric encoders for protein primary and tertiary structures. This framework guides mutation directions toward desired traits by simulating natural selection on wild-type proteins and evaluates variant effects based on their fitness to perform specific functions. We assess the proposed approach using three benchmarks comprising over 300 deep mutational scanning assays. The prediction results showcase exceptional performance across extensive experiments compared to other zero-shot learning methods, all while maintaining a minimal cost in terms of trainable parameters. This study not only proposes an effective framework for more accurate and comprehensive predictions to facilitate efficient protein engineering, but also enhances the in silico assessment system for future deep learning models to better align with empirical requirements. The PyTorch implementation is available at https://github.com/ai4protein/ProtSSN.

PMID:40314227 | DOI:10.7554/eLife.98033

Front Bioinform. 2025 Apr 17;5:1522401. doi: 10.3389/fbinf.2025.1522401. eCollection 2025.

ABSTRACT

The incidence of non-alcoholic fatty liver disease (NAFLD), encompassing the more severe non-alcoholic steatohepatitis (NASH), is rising alongside the surges in diabetes and obesity. Increasing evidence indicates that NASH is responsible for a significant share of idiopathic hepatocellular carcinoma (HCC) cases, a fatal cancer with a 5-year survival rate below 22%. Biomarkers can facilitate early screening and monitoring of at-risk NAFLD/NASH patients and assist in identifying potential drug candidates for treatment. This study utilized an ensemble feature selection framework to analyze transcriptomic data, identifying biomarker genes associated with the stage-wise progression of NAFLD-related HCC. Seven machine learning algorithms were assessed for disease stage classification. Twelve feature selection methods including correlation-based techniques, mutual information-based methods, and embedded techniques were utilized to rank the top genes as features, through this approach, multiple feature selection methods were combined to yield more robust features important in this disease progression. Cox regression-based survival analysis was carried out to evaluate the biomarker potentiality of these genes. Furthermore, multiphase drug repurposing strategy and molecular docking were employed to identify potential drug candidates against these biomarkers. Among the seven machine learning models initially evaluated, DISCR resulted as the most accurate disease stage classifier. Ensemble feature selection identified ten top genes, among which eight were recognized as potential biomarkers based on survival analysis. These include genes ABAT, ABCB11, MBTPS1, and ZFP1 mostly involved in alanine and glutamate metabolism, butanoate metabolism, and ER protein processing. Through drug repurposing, 81 candidate drugs were found to be effective against these markers genes, with Diosmin, Esculin, Lapatinib, and Phenelzine as the best candidates screened through molecular docking and MMGBSA. The consensus derived from multiple methods enhances the accuracy of identifying relevant robust biomarkers for NAFLD-associated HCC. The use of these biomarkers in a multiphase drug repurposing strategy highlights potential therapeutic options for early intervention, which is essential to stop disease progression and improve outcomes.

PMID:40313868 | PMC:PMC12043677 | DOI:10.3389/fbinf.2025.1522401

Int J Neuropsychopharmacol. 2025 May 2:pyaf029. doi: 10.1093/ijnp/pyaf029. Online ahead of print.

ABSTRACT

BACKGROUND: Vortioxetine is a serotonin reuptake inhibitor and serotonin receptor modulator used for the treatment of major depressive disorder, but recent studies have also reported anticancer effects in models of glioblastoma. Given the well-established benefits of drug repositioning, we examined the pharmacological mechanism for these anticancer actions using bioinformatics and molecular docking.

METHODS: Putative molecular targets for vortioxetine were identified by searching DrugBank, GeneCards, SwissTargetPrediction, CTD, and SuperPred databases, while glioblastoma-related proteins were identified using GeneCards, OMIM, and TTD. A protein-protein interaction (PPI) network was constructed from vortioxetine targets also involved in glioblastoma to identify core (hub) targets, which were then characterized by GO and KEGG pathway enrichment analyses using DAVID. Cytoscape was utilized to generate a drug-pathway-target-disease network, and molecular docking simulations were performed to evaluate direct interactions between vortioxetine and core target proteins.

RESULTS: A total of 234 unique vortioxetine protein targets were identified. Among 234 vortioxetine targets identified, 48 were also related to glioblastoma. Topological analysis of the PPI network revealed five core targets: the serine/threonine kinase AKT1, transcription factor hypoxia-inducible factor (HIF)-1, cell adhesion molecule cadherin-E, NF-κB subunit p105, and prostaglandin-endoperoxide synthase 2. According to GO and KEGG pathway analyses, the anticancer efficacy of vortioxetine may be mediated by effects on glucose metabolism, cell migration, phosphorylation, inflammatory responses, apoptosis, and signaling via Rap1, chemical carcinogenesis-reactive oxygen species, and HIF-1. Molecular docking revealed moderately strong affinities between vortioxetine and four core targets.

CONCLUSIONS: This study suggests that vortioxetine may inhibit glioblastoma development through direct effects on multiple targets, and further emphasizes the value of bioinformatics analyses for drug repositioning.

PMID:40312983 | DOI:10.1093/ijnp/pyaf029

Front Genet. 2025 Apr 17;16:1509042. doi: 10.3389/fgene.2025.1509042. eCollection 2025.

ABSTRACT

OBJECTIVES: This study aims to identify specific single nucleotide polymorphism (SNP) correlated to perioperative analgesia in patients undergoing laparoscopic gynecological surgery.

METHODS: A total of 200 females meeting specific criteria underwent gynecological laparoscopic procedures under general anesthesia. Preoperative pain sensitivity was evaluated using Pain Sensitivity Questionnaire and Pain Catastrophizing Scale (PCS). Venous blood samples were collected for SNP analysis of nine genes. The study analyzed the correlation between SNPs and pre-operative pain assessment, analgesics usage, and the occurrence of related adverse effects.

RESULTS: Six out of nine identified loci showed polymorphisms. The PCS scores were higher in the mutation group (GG + GC) for ADRA2A rs1800544 compared to the CC group (P < 0.05). No differences were observed in visual analog scale or Ramsay sedation scores between the mutation and wild-type groups for any of the SNPs (P < 0.05). Patients in the mutant group (AG + GG) for OPRM1 rs1799971 had higher analgesic usage within 24 h compared to the wild-type group (P < 0.05). The consumption of intraoperative remifentanil was higher in the mutation group (GG + GC) of ADRA2A rs1800544 than in the CC group. The Multifactorial Dimensionality Reduction analysis suggests that the optimal interaction model includes OPRM1 rs179971 and CYP450 3A4 * 1G rs2242480 together.

CONCLUSION: Patients with GG and AG genotypes of OPRM1 rs1799971 gene required more 24-h postoperative analgesics after gynecological surgery compared to those with AA genotype. A SNP-SNP interaction was observed between OPRM1 rs179971 and CYP450 3A4 * 1G rs2242480. Clinical Trial Registration: (www.chictr.org.cn, registration number: ChiCTR2200062425).

PMID:40313598 | PMC:PMC12043472 | DOI:10.3389/fgene.2025.1509042

Int J Mol Med. 2025 Jun;55(6):99. doi: 10.3892/ijmm.2025.5540. Epub 2025 May 2.

ABSTRACT

Cation channels have been extensively studied in the context of digestive disorders, but comparatively little attention has been given to anions and their associated channels. Chloride ions, the most abundant anions in the human body, act as signaling molecules, modulating cellular behavior and playing a key role in regulating multiorgan physiological and pathophysiological mechanisms. The intra‑ and extracellular distributions of chloride ions are primarily controlled by various chloride channels and transporters. Currently, these chloride channels are classified into several groups: The chloride channels family, cystic fibrosis transmembrane conductance regulator, calcium‑activated chloride channels, volume‑regulated anion channels, proton‑activated chloride channels and ligand‑gated anion channels. This review aims to summarize the roles of chloride ion channels and transporter proteins in digestive system diseases, providing a theoretical basis for future research and offering potential new strategies for disease treatment.

PMID:40314091 | DOI:10.3892/ijmm.2025.5540

Ulster Med J. 2025 Apr;94(1):28-30. Epub 2025 Apr 30.

NO ABSTRACT

PMID:40313987 | PMC:PMC12042854

Case Rep Med. 2025 Apr 23;2025:9935363. doi: 10.1155/carm/9935363. eCollection 2025.

ABSTRACT

This case report describes a 39-year-old phenotypically normal male patient of a married couple with primary infertility presenting as candidates for assisted reproductive techniques. The medical history of the couple is unremarkable, with both partners phenotypically normal. Semen analysis revealed oligoasthenzoospermia (OAT), 15% sperm DNA fragmentation and 4% aneuploidies in the sperm nuclei. Genetic analysis showed no Y chromosome of cystic fibrosis transmembrane conductance regulator gene mutations. Karyotype analysis in the male partner revealed a small supernumerary marker chromosome (sSMC) derived from chromosome 15, specifically inverted and duplicated (inv dup(15)) corresponding to the 15q11.2 region but lacking the Prader-Willi/Angelman syndrome critical region (PWACR). Further investigations revealed that 35% of the patient's spermatozoa carried the sSMC(15). This case study highlights the potential association between the presence of an inv dup(15) sSMC, without the involvement of the PWACR, and male infertility. sSMC(15) may disrupt spermatogenesis and contribute to oligoasthenozoospermia in males with primary infertility. Further research into the association of mechanism mechanisms of male infertility related to the 15q11.2 region is warranted.

PMID:40313645 | PMC:PMC12043387 | DOI:10.1155/carm/9935363

FEMS Microbiol Rev. 2025 May 1:fuaf018. doi: 10.1093/femsre/fuaf018. Online ahead of print.

ABSTRACT

The exquisite ability of bacteria to adapt to their environment is essential for their capacity to colonise hostile niches. In the cystic fibrosis (CF) lung, hypoxia is among several environmental stresses that opportunistic pathogens must overcome to persist and chronically colonise. Although the role of hypoxia in the host has been widely reviewed, the impact of hypoxia on bacterial pathogens has not yet been studied extensively. This review considers the bacterial oxygen-sensing mechanisms in three species that effectively colonise the lungs of people with CF, namely Pseudomonas aeruginosa, Burkholderia cepacia complex and Mycobacterium abscessus and draws parallels between their three proposed oxygen-sensing two-component systems: BfiSR, FixLJ, and DosRS, respectively. Moreover, each species expresses regulons that respond to hypoxia: Anr, Lxa, and DosR, and encode multiple proteins that share similar homologies and function. Many adaptations that these pathogens undergo during chronic infection, including antibiotic resistance, protease expression, or changes in motility, have parallels in the responses of the respective species to hypoxia. It is likely that exposure to hypoxia in their environmental habitats predispose these pathogens to colonisation of hypoxic niches, arming them with mechanisms than enable their evasion of the immune system and establish chronic infections. Overcoming hypoxia presents a new target for therapeutic options against chronic lung infections.

PMID:40312783 | DOI:10.1093/femsre/fuaf018

J Neural Eng. 2025 May 2;22(3). doi: 10.1088/1741-2552/adcfc8.

ABSTRACT

Objective.Major depressive disorder (MDD) is a widespread mental disorder that affects health. Many methods combining electroencephalography (EEG) with machine learning or deep learning have been proposed to objectively distinguish between MDD and healthy individuals. However, most current methods detect depression based on multichannel EEG signals, which constrains its application in daily life. The context in which EEG is obtained can vary in terms of study designs and EEG equipment settings, and the available depression EEG data is limited, which could also potentially lessen the efficacy of the model in differentiating between MDD and healthy subjects. To solve the above challenges, a depression detection model leveraging transfer learning with the single-channel EEG is advanced.Approach.We utilized a pretrained ResNet152V2 network to which a flattening layer and dense layer were appended. The method of feature extraction was applied, meaning that all layers within ResNet152V2 were frozen and only the parameters of the newly added layers were adjustable during training. Given the superiority of deep neural networks in image processing, the temporal sequences of EEG signals are first converted into images, transforming the problem of EEG signal categorization into an image classification task. Subsequently, a cross-subject experimental strategy was adopted for model training and performance evaluation.Main results.The model was capable of precisely (approaching 100% accuracy) identifying depression in other individuals by employing single-channel EEG samples obtained from a limited number of subjects. Furthermore, the model exhibited superior performance across four publicly available depression EEG datasets, thereby demonstrating good adaptability in response to variations in EEG caused by the context.Significance.This research not only highlights the impressive potential of deep transfer learning techniques in EEG signal analysis but also paves the way for innovative technical approaches to facilitate early diagnosis of associated mental disorders in the future.

PMID:40314182 | DOI:10.1088/1741-2552/adcfc8

Brief Bioinform. 2025 May 1;26(3):bbaf218. doi: 10.1093/bib/bbaf218.

NO ABSTRACT

PMID:40314061 | DOI:10.1093/bib/bbaf218

Front Neuroinform. 2025 Apr 17;19:1550432. doi: 10.3389/fninf.2025.1550432. eCollection 2025.

ABSTRACT

INTRODUCTION: Brain tumors are a leading cause of mortality worldwide, with early and accurate diagnosis being essential for effective treatment. Although Deep Learning (DL) models offer strong performance in tumor detection and segmentation using MRI, their black-box nature hinders clinical adoption due to a lack of interpretability.

METHODS: We present a hybrid AI framework that integrates a 3D U-Net Convolutional Neural Network for MRI-based tumor segmentation with radiomic feature extraction. Dimensionality reduction is performed using machine learning, and an Adaptive Neuro-Fuzzy Inference System (ANFIS) is employed to produce interpretable decision rules. Each experiment is constrained to a small set of high-impact radiomic features to enhance clarity and reduce complexity.

RESULTS: The framework was validated on the BraTS2020 dataset, achieving an average DICE Score of 82.94% for tumor core segmentation and 76.06% for edema segmentation. Classification tasks yielded accuracies of 95.43% for binary (healthy vs. tumor) and 92.14% for multi-class (healthy vs. tumor core vs. edema) problems. A concise set of 18 fuzzy rules was generated to provide clinically interpretable outputs.

DISCUSSION: Our approach balances high diagnostic accuracy with enhanced interpretability, addressing a critical barrier in applying DL models in clinical settings. Integrating of ANFIS and radiomics supports transparent decision-making, facilitating greater trust and applicability in real-world medical diagnostics assistance.

PMID:40313917 | PMC:PMC12043696 | DOI:10.3389/fninf.2025.1550432

Front Bioinform. 2025 Apr 17;5:1554010. doi: 10.3389/fbinf.2025.1554010. eCollection 2025.

ABSTRACT

Single-cell and spatial technologies have transformed our understanding of brain immunology, providing unprecedented insights into immune cell heterogeneity and spatial organisation within the central nervous system. These methods have uncovered complex cellular interactions, rare cell populations, and the dynamic immune landscape in neurological disorders. This review highlights recent advances in single-cell "omics" data analysis and discusses their applicability for brain immunology. Traditional statistical techniques, adapted for single-cell omics, have been crucial in categorizing cell types and identifying gene signatures, overcoming challenges posed by increasingly complex datasets. We explore how machine learning, particularly deep learning methods like autoencoders and graph neural networks, is addressing these challenges by enhancing dimensionality reduction, data integration, and feature extraction. Newly developed foundation models present exciting opportunities for uncovering gene expression programs and predicting genetic perturbations. Focusing on brain development, we demonstrate how single-cell analyses have resolved immune cell heterogeneity, identified temporal maturation trajectories, and uncovered potential therapeutic links to various pathologies, including brain malignancies and neurodegeneration. The integration of single-cell and spatial omics has elucidated the intricate cellular interplay within the developing brain. This mini-review is intended for wet lab biologists at all career stages, offering a concise overview of the evolving landscape of single-cell omics in the age of widely available artificial intelligence.

PMID:40313869 | PMC:PMC12043695 | DOI:10.3389/fbinf.2025.1554010

Front Comput Neurosci. 2025 Apr 16;19:1569828. doi: 10.3389/fncom.2025.1569828. eCollection 2025.

ABSTRACT

INTRODUCTION: Major Depressive Disorder (MDD) remains a critical mental health concern, necessitating accurate detection. Traditional approaches to diagnosing MDD often rely on manual Electroencephalography (EEG) analysis to identify potential disorders. However, the inherent complexity of EEG signals along with the human error in interpreting these readings requires the need for more reliable, automated methods of detection.

METHODS: This study utilizes EEG signals to classify MDD and healthy individuals through a combination of machine learning, deep learning, and split learning approaches. State of the art machine learning models i.e., Random Forest, Support Vector Machine, and Gradient Boosting are utilized, while deep learning models such as Transformers and Autoencoders are selected for their robust feature-extraction capabilities. Traditional methods for training machine learning and deep learning models raises data privacy concerns and require significant computational resources. To address these issues, the study applies a split learning framework. In this framework, an ensemble learning technique has been utilized that combines the best performing machine and deep learning models.

RESULTS: Results demonstrate a commendable classification performance with certain ensemble methods, and a Transformer-Random Forest combination achieved 99% accuracy. In addition, to address data-sharing constraints, a split learning framework is implemented across three clients, yielding high accuracy (over 95%) while preserving privacy. The best client recorded 96.23% accuracy, underscoring the robustness of combining Transformers with Random Forest under resource-constrained conditions.

DISCUSSION: These findings demonstrate that distributed deep learning pipelines can deliver precise MDD detection from EEG data without compromising data security. Proposed framework keeps data on local nodes and only exchanges intermediate representations. This approach meets institutional privacy requirements while providing robust classification outcomes.

PMID:40313734 | PMC:PMC12044669 | DOI:10.3389/fncom.2025.1569828

Front Pediatr. 2025 Apr 17;13:1514447. doi: 10.3389/fped.2025.1514447. eCollection 2025.

ABSTRACT

Artificial Intelligence is revolutionizing prenatal diagnostics by enhancing the accuracy and efficiency of procedures. This review explores AI and machine learning (ML) in the early detection, prediction, and assessment of neural tube defects (NTDs) through prenatal ultrasound imaging. Recent studies highlight the effectiveness of AI techniques, such as convolutional neural networks (CNNs) and support vector machines (SVMs), achieving detection accuracy rates of up to 95% across various datasets, including fetal ultrasound images, genetic data, and maternal health records. SVM models have demonstrated 71.50% accuracy on training datasets and 68.57% on testing datasets for NTD classification, while advanced deep learning (DL) methods report patient-level prediction accuracy of 94.5% and an area under the receiver operating characteristic curve (AUROC) of 99.3%. AI integration with genomic analysis has identified key biomarkers associated with NTDs, such as Growth Associated Protein 43 (GAP43) and Glial Fibrillary Acidic Protein (GFAP), with logistic regression models achieving 86.67% accuracy. Current AI-assisted ultrasound technologies have improved diagnostic accuracy, yielding sensitivity and specificity rates of 88.9% and 98.0%, respectively, compared to traditional methods with 81.5% sensitivity and 92.2% specificity. AI systems have also streamlined workflows, reducing median scan times from 19.7 min to 11.4 min, allowing sonographers to prioritize critical patient care. Advancements in DL algorithms, including Oct-U-Net and PAICS, have achieved recall and precision rates of 0.93 and 0.96, respectively, in identifying fetal abnormalities. Moreover, AI's evolving role in genetic research supports personalized NTD prevention strategies and enhances public awareness through AI-generated health messages. In conclusion, the integration of AI in prenatal diagnostics significantly improves the detection and assessment of NTDs, leading to greater accuracy and efficiency in ultrasound imaging. As AI continues to advance, it has the potential to further enhance personalized healthcare strategies and raise public awareness about NTDs, ultimately contributing to better maternal and fetal outcomes.

PMID:40313675 | PMC:PMC12043698 | DOI:10.3389/fped.2025.1514447

Proc IEEE Int Symp Biomed Imaging. 2024 May;2024. doi: 10.1109/ISBI56570.2024.10635349. Epub 2024 Aug 22.

ABSTRACT

Data limitation is a significant challenge in applying deep learning to medical images. Recently, the diffusion probabilistic model (DPM) has shown the potential to generate high-quality images by converting Gaussian random noise into realistic images. In this paper, we apply the DPM to augment the deep ultraviolet fluorescence (DUV) image dataset with an aim to improve breast cancer classification for intra-operative margin assessment. For classification, we divide the whole surface DUV image into small patches and extract convolutional features for each patch by utilizing the pre-trained ResNet. Then, we feed them into an XGBoost classifier for patch-level decisions and then fuse them with a regional importance map computed by Grad-CAM++ for whole surface-level prediction. Our experimental results show that augmenting the training dataset with the DPM significantly improves breast cancer detection performance in DUV images, increasing accuracy from 93% to 97%, compared to using Affine transformations and ProGAN.

PMID:40313564 | PMC:PMC12045284 | DOI:10.1109/ISBI56570.2024.10635349