Computers in biology and medicine最新文献

{"title":"Harnessing network pharmacology and in silico drug discovery to uncover new targets and therapeutics for Alzheimer's disease","authors":"Haitham Al Madhagi ,&nbsp;Husam Nassar","doi":"10.1016/j.compbiomed.2025.109781","DOIUrl":"10.1016/j.compbiomed.2025.109781","url":null,"abstract":"<div><div>Alzheimer's disease (AD) is the leading cause of progressive neurodegenerative dementia, affecting approximately 50 million individuals globally. Recent studies have highlighted the differential expression of circular RNAs (circRNAs) in AD, which may disrupt the circRNA-miRNA-mRNA regulatory networks in neuronal cells. This work aims to integrate network pharmacology with in silico drug design to identify novel druggable targets for AD and propose promising drug candidates. We analyzed two circRNA datasets from the Gene Expression Omnibus, employing enrichment analysis and constructing a circRNA-miRNA-mRNA network. The RNAenrich platform facilitated the identification of hub genes and potential druggable targets. The identified target was subjected to virtual screening against a chemical drug library comprising over 6000 compounds in clinical trials while ensuring compliance with Lipinski's Rule of Five. Our findings reveal that differentially expressed circRNAs are significantly involved in gland development, apoptosis regulation, hypoxic response, and neuronal death. Notably, CDK-6 emerged as the most promising druggable target, exhibiting strong binding affinity with five selected ligands: DB06963, DB06888, DB07020, DB08683, and DB06976. These ligands demonstrated distinct binding modes and stable interactions over 500 ns of molecular dynamics simulations conducted via Desmond. In conclusion, our study identifies CDK-6 as a viable target for therapeutic intervention in Alzheimer's disease. The top five ligands present a compelling case for further investigation as innovative CDK-6 inhibitors and potential drug candidates for AD treatment.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"187 ","pages":"Article 109781"},"PeriodicalIF":7.0,"publicationDate":"2025-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143180256","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Two-stage CNN-based framework for leukocytes classification","authors":"Siraj Khan ,&nbsp;Muhammad Sajjad ,&nbsp;José Escorcia-Gutierrez ,&nbsp;Sami Dhahbi ,&nbsp;Mohammad Hijji ,&nbsp;Khan Muhammad","doi":"10.1016/j.compbiomed.2024.109616","DOIUrl":"10.1016/j.compbiomed.2024.109616","url":null,"abstract":"<div><div>Leukocytes are pivotal markers in health, crucial for diagnosing diseases like malaria and viral infections. Peripheral blood smear tests provide pathologists with vital insights into various medical conditions. Manual leukocyte counting is challenging and error-prone due to their complex structure. Accurate segmentation and classification of leukocytes remain challenging, impacting both accuracy and efficiency in blood microscopic image analysis. To overcome these limitations, we propose a robust two-stage CNN framework that integrates YOLOv8 for precise segmentation and MobileNetV3 for effective classification. Initially, WBCs are segmented using YOLOv8m-seg, extracting ROIs for subsequent analysis. Then, features from segmented ROIs are used to train MobileNetV3, classifying WBCs into lymphocytes, monocytes, basophils, eosinophils, and neutrophils. This framework significantly advances leukocyte categorization, enhancing diagnostic performance and patient outcomes. The proposed technique achieved impressive accuracy rates of 99.56 %, 99.19 % and 98.89 % during segmentation and 99.28 %, 99.63 % and 98.49 % during classification on Raabin-WBC, PBC and LISC datasets, respectively, outperforming state-of-the-art methods.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"187 ","pages":"Article 109616"},"PeriodicalIF":7.0,"publicationDate":"2025-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143180255","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Computer-aided design of imatinib derivatives: Overcoming drug-resistance in chronic myeloid leukemia","authors":"S. Gholizadeh ,&nbsp;N. Panahi ,&nbsp;N. Razzaghi-Asl","doi":"10.1016/j.compbiomed.2025.109784","DOIUrl":"10.1016/j.compbiomed.2025.109784","url":null,"abstract":"<div><div>Chronic myeloid leukemia (CML) is a hematologic condition characterized by the overexpression of stem blood cells in the bone marrow. The Philadelphia chromosome encodes the oncogenic tyrosine kinase BCR-ABL, which is a hallmark of CML. Imatinib, a phenylamino pyrimidine derivative, was the first tyrosine kinase inhibitor (TKI) approved in 2001 for CML. Despite launching a new era in cancer targeted therapy, acquired resistance occurred due to the point mutation of a gatekeeper residue (Thr315Ile) at the BCR-ABL catalytic pocket. There are no approved medications for Thr315Ile-BCR-ABL harboring CML patients. Present study was aimed at the in silico identification of synthetically accessible imatinib derivatives that are likely to bind a frequent Thr315Ile-BCR-ABL and overcome drug resistance. 4-((4-benzylpiperazin-1-yl) methyl)-<em>N</em>-(4-methyl-3-(4-(pyridine-3-yl) pyrimidine-2-amino) phenyl) benzamide (SCHEMBL12127861) and 4-(methoxy (methyl) amino)-<em>N</em>-(3-methyl-5-(pyrido[3,4-b] pyrazin-2-yl thio) phenyl) benzamide (<strong>18</strong>) were revealed as top-binders. Molecular dynamics simulations and free energy calculations conferred stable binding features Thr315Ile-BCR-ABL. A new binding model was suggested for <strong>18</strong> that resided outside the kinase domain (∼15 Å from Ile315). Considering stability and binding energy compared to imatinib, the intended binding model may be the subject of further evaluations to overwhelm drug resistance. <span><span>SCHEMBL12127861</span><svg><path></path></svg></span> had appropriate and more buried accommodation than imatinib near the DFG motif and P-loop of the kinase domain. Hydrogen bonds, π-cation interaction, salt bridge, and a vdW cooperative contacts mediated the complex stability. Although validation of proposed models is to be achieved, this study identified synthetically accessible in silico hits with tight binding to the clinically frequent mutant BCR-ABL phenotypes in Thr315Ile-positive CML patients.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"187 ","pages":"Article 109784"},"PeriodicalIF":7.0,"publicationDate":"2025-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143179695","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"In-silico exploring pathway and mechanism-based therapeutics for allergic rhinitis: Network pharmacology, molecular docking, ADMET, quantum chemistry and machine learning based QSAR approaches","authors":"K.M. Tanjida Islam,&nbsp;Shahin Mahmud","doi":"10.1016/j.compbiomed.2025.109754","DOIUrl":"10.1016/j.compbiomed.2025.109754","url":null,"abstract":"<div><div>Allergic rhinitis is a devastating health complication that interrupts the quality of daily life and significantly affects around 40 % of the population worldwide. Despite the availability of various treatment options, many patients continue to struggle with persistent symptoms and side effects, highlighting the need for innovative therapeutic approaches. Therefore, identifying pathway and mechanism-based targeted therapies with more effective and fewer side effects could aid current therapeutics and provide novel therapeutic advantages. This study aimed to identify potential drug candidates for allergic rhinitis treatment by employing <em>in-silico</em> approaches, including network pharmacology, molecular docking, ADMET, similarity, pharmacophore modeling, quantum chemistry, and machine learning-based QSAR modeling. From three traditionally used medicinal plants known as allergic rhinitis curing, <em>Xanthium strumarium</em>, <em>Magnolia liliiflora</em>, and <em>Tylophora indica</em>, 241 compounds were obtained, and their favorable ADMET properties were analyzed. Network pharmacology revealed 203 potential therapeutic targets, with 15 hub targets identified through protein-protein interaction network analysis and most of them play key roles in inflammatory and immune pathways confirmed by KEGG pathway analysis. Molecular docking, similarity testing, and pharmacophore modeling studies identified promising compounds Quercetin, Yinyanghuo E, Uralenin, CID:90643991, CID:42607537, CID:76329670, Heracetin, and Fisetin exhibiting strong binding affinities with key regulatory targets, NFKB1, TRAF6, and key cytokines IL5, and IL6 that directly and indirectly involved in allergic reactions. Quantum chemistry calculations revealed favorable electronic properties and reactivities of these compounds. The machine learning-based QSAR model predicted IC₅₀ &lt; 50 nM for almost all compounds, indicating highly potent inhibitors. Hence, this <em>in-silico</em> study identified some novel promising drug candidates for treating allergic rhinitis by targeting crucial inflammatory and immune pathways, offering improved treatment outcomes and reduced side effects, subject to further experimental validation.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"187 ","pages":"Article 109754"},"PeriodicalIF":7.0,"publicationDate":"2025-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143180254","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Finite element analysis of anatomically-modelled prosthetic incus for ossicular chain reconstruction","authors":"Masoud Mohseni-Dargah ,&nbsp;Christopher Pastras ,&nbsp;Payal Mukherjee ,&nbsp;Kai Cheng ,&nbsp;Khosro Khajeh ,&nbsp;Mohsen Asadnia","doi":"10.1016/j.compbiomed.2025.109770","DOIUrl":"10.1016/j.compbiomed.2025.109770","url":null,"abstract":"<div><div>Ossicular chain reconstruction (OCR) is the gold standard for repairing conductive hearing loss (CHL) using prosthetic devices. However, few techniques can validate prosthesis performance before translational implementation. This study not only validates a Finite Element (FE) model for examining (reconstructed) middle ear biomechanics, but also evaluates the potential of personalised prosthetic devices for OCR. Additionally, this research examines tricalcium phosphate (TCP), as a potential biomaterial, besides titanium and hydroxyapatite (HA), as commonly used biomaterials for use in (personalised) OCR. Here, the designed FE model tested the hypothesis that 3D anatomically modelled prosthetic devices (prosthetic incus) can reliably restore sound transmission using appropriate biomaterials, including titanium, HA, or TCP. Our FE modelling examined middle ear biomechanics before and after prosthesis replacement for each biomaterial assignment. Results demonstrated the FE model is in agreement with experimental vibrometry recordings of ossicular motion and earlier numerical simulations. Additionally, the anatomically modelled prosthetic incus closely mimicked normal middle ear biomechanics, revealing its potential for OCR. FE analysis revealed no significant differences between titanium, HA, and TCP prostheses functions, serving as first-order evidence for their support in OCR. This research establishes a FE-based framework for personalised OCR following imaging, which is valuable for future personalised treatments for patients with CHL due to ossicular dysfunction. FE simulations can evaluate the biomechanics and function of prostheses, helping the surgeon make well-informed decisions regarding OCR for translational outcomes.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"187 ","pages":"Article 109770"},"PeriodicalIF":7.0,"publicationDate":"2025-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143181296","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MRI morphological features combined with apparent diffusion coefficient can predict brain invasion in meningioma","authors":"Xiaoyu Huang ,&nbsp;Yuntai Cao ,&nbsp;Guojin Zhang ,&nbsp;FuQiang Tang ,&nbsp;Dandan Sun ,&nbsp;Jialiang Ren ,&nbsp;Wenyi Li ,&nbsp;Junlin Zhou ,&nbsp;Jing Zhang","doi":"10.1016/j.compbiomed.2025.109763","DOIUrl":"10.1016/j.compbiomed.2025.109763","url":null,"abstract":"<div><h3>Objectives</h3><div>Accurately predicting meningioma brain invasion preoperatively helps to select the appropriate surgical approach and predict prognosis, but there are few imaging features that are sufficient for discriminating it alone. We investigate the joint MR imaging features and apparent diffusion coefficient (ADC) to predict the risk of brain invasion of meningiomas preoperatively.</div></div><div><h3>Methods</h3><div>In this retrospective study, 143 patients (invasion group:51, non-invasion group: 92) diagnosed with meningioma by histopathology were included. The maximum (ADC_max), minimum (ADC_min) and mean (ADC_mean) values of ADC and the mean ADC values of a comparative ROI in the normal appearing white matter (ADC_NAWM) were calculated. Differences between clinical features, MRI morphological features, and all ADC values were assessed by Pearson's chi-square test and Kruskal-Wallis rank-sum test. Stepwise logistic regression analysis was used to select the optimal features and construct a prediction model. Furthermore, A nomogram was used to predict the risk of brain invasion, and a decision curve was used to verify the clinical utility of the nomogram.</div></div><div><h3>Results</h3><div>According to stepwise logistic regression analysis, we found that sex, maximum diameter, peritumoral edema and ADC_min were closely related to brain invasion in meningioma. The model of the above four variables has the optimal discriminative ability to predict brain invasion, with an AUC of 0.924 (95 % CI, 0.879–0.969) and a sensitivity of 92.2 % (95 % CI, 74.5%–98.0 %).</div></div><div><h3>Conclusions</h3><div>Combining clinical features, MRI morphological characteristics and ADC_min, the model exhibits excellent discriminatory ability and high sensitivity, which can be used for predicting the risk of brain invasion of meningiomas.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"187 ","pages":"Article 109763"},"PeriodicalIF":7.0,"publicationDate":"2025-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143180253","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Ten Meter Walk Test for motor function assessment with technological devices based on lower members’ movements: A systematic review","authors":"Maykol Santos ,&nbsp;Eftim Zdravevski ,&nbsp;Carlos Albuquerque ,&nbsp;Paulo Jorge Coelho ,&nbsp;Ivan Miguel Pires","doi":"10.1016/j.compbiomed.2025.109734","DOIUrl":"10.1016/j.compbiomed.2025.109734","url":null,"abstract":"<div><h3>Objective:</h3><div>The Ten Meter Walk Test (10MWT) is a vital diagnostic tool for identifying neuromuscular and neurodegenerative conditions. This systematic review explores the potential of wearables, mobile devices, and sensors to enhance the 10MWT’s use in medical gait analysis based on lower limb movements.</div></div><div><h3>Methods:</h3><div>This systematic review explores the use of wearables, mobile devices, and sensors to improve the 10MWT in medical gait analysis based on lower limb movements. The study uses the PRISMA approach to assess literature from January 2010 to October 2023, highlighting the importance of new technologies like machine learning and artificial intelligence in improving the accuracy and efficiency of the 10MWT.</div></div><div><h3>Results:</h3><div>The findings demonstrate how technology-enabled 10MWT can help develop specialized treatment strategies and provide a more accurate understanding of disease pathophysiology.</div></div><div><h3>Conclusions:</h3><div>The paper reviews 17 studies on lower limb movements during the 10MWT, highlighting their importance in assessing medical diseases and gait analysis as a diagnostic tool. It emphasizes the role of technology in rehabilitation and physical therapy, where some studies combine Transcranial Direct Current Stimulation with robotic or wearable technologies.</div></div><div><h3>Significance:</h3><div>The review comprehensively explains these technologies’ advantages and current use in therapeutic contexts.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"187 ","pages":"Article 109734"},"PeriodicalIF":7.0,"publicationDate":"2025-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143181788","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The effects of ventricle geometries and boundary conditions on computational modeling of ventriculoperitoneal catheters","authors":"Bryan C. Good ,&nbsp;James A. Killeffer ,&nbsp;Stephanie C. TerMaath","doi":"10.1016/j.compbiomed.2025.109776","DOIUrl":"10.1016/j.compbiomed.2025.109776","url":null,"abstract":"<div><div>Hydrocephalus is a condition where an excess amount of cerebrospinal fluid (CSF) accumulates within the ventricles of the brain, leading to elevated intracranial pressure. The most common treatment is the surgical placement of a brain shunt to drain CSF from the ventricles. However, brain shunts have extremely high failure rates and improved devices are needed to minimize device obstruction and failure. To help establish modeling standards for this complex scenario, this work computationally investigates geometric and fluid dynamic variables to determine their effects on shunt performance, in addition to evaluating physiological modeling requirements. Catheter performance metrics included total catheter flow, drainage hole flows, and wall shear stresses (WSSs), which are all known to influence catheter obstruction.</div><div>It was determined that ventricle and choroid plexus parameters (size, shape, and location) did not play a significant role in catheter performance (&lt;2 % drainage hole flow differences, &lt;12 % WSS differences). Further, patient-specific ventricle models were found to not affect catheter performance compared to a simplified cylinder model (&lt;1 % drainage hole flow differences, &lt;10 % WSS differences). Intracranial pressure boundary conditions, both static and pulsatile, were also applied. It was found that drainage hole flows and WSSs averaged over time were not significantly different between the waveforms and from comparable static pressure simulations. These results show that simplified geometric models and pressure boundary conditions can be used to computationally study ventriculoperitoneal (VP) catheter performance. The results of this research will enhance overall knowledge of CSF dynamics, provide modeling guidance, and contribute to the development of improved CSF shunts.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"187 ","pages":"Article 109776"},"PeriodicalIF":7.0,"publicationDate":"2025-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143181787","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Towards real-world clinical data standardization: A modular FHIR-driven transformation pipeline to enhance semantic interoperability in healthcare","authors":"Alberto Marfoglia ,&nbsp;Filippo Nardini ,&nbsp;Valerio Antonio Arcobelli ,&nbsp;Serena Moscato ,&nbsp;Sabato Mellone ,&nbsp;Antonella Carbonaro","doi":"10.1016/j.compbiomed.2025.109745","DOIUrl":"10.1016/j.compbiomed.2025.109745","url":null,"abstract":"<div><h3>Background:</h3><div>Given the exponential increase in clinical data, which accounts for around 30% of global data volume, effective information management has become crucial to ensuring robust interoperability. This trend is further expedited by implementing consumer-oriented Internet of Things platforms, contributing to the growth of the $8.3 trillion healthcare industry. These advancements, combined with challenges such as heterogeneous data formats and a lack of incentives, necessitate the development of pragmatic infrastructures and tools that harness contemporary clinical standards like Fast Healthcare Interoperability Resources (FHIR).</div></div><div><h3>Objective:</h3><div>This study aims to present a modular conversion pipeline employing a templating strategy for translating clinical data into the FHIR model. Emphasis is placed on utilizing a standard mapping specification like FHIR Mapping Language. This ensures essential properties such as platform independence, portability, and code reusability.</div></div><div><h3>Methods:</h3><div>The pipeline was developed incrementally, dividing its core functionalities into five modules: Input, Refinement, Mapping, Validation, and Export. These were subsequently validated by converting a dataset from a prosthetic fitting and rehabilitation center to demonstrate the approach’s validity in a real-world data context.</div></div><div><h3>Results:</h3><div>A total of 1962 hospital stay records of 1006 unique patients were converted successfully to 15 distinct types of FHIR resources. The successful conversion states the pipeline’s effectiveness, additionally showcasing its capabilities for enhancing semantic interoperability and facilitating the reuse of real-world data.</div></div><div><h3>Conclusion:</h3><div>Our approach emerges as a modular data conversion framework that addresses the limitations of existing solutions, making significant contributions to the creation of standardized, interoperable, and high-quality clinical datasets that serve as a foundation for further work.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"187 ","pages":"Article 109745"},"PeriodicalIF":7.0,"publicationDate":"2025-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143181790","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"S4Sleep: Elucidating the design space of deep-learning-based sleep stage classification models","authors":"Tiezhi Wang,&nbsp;Nils Strodthoff","doi":"10.1016/j.compbiomed.2025.109735","DOIUrl":"10.1016/j.compbiomed.2025.109735","url":null,"abstract":"<div><div>Machine-learning-based automatic sleep stage scoring is a promising approach to enhance the time-consuming manual annotation process of polysomnography recordings. Although numerous algorithms have been proposed for this purpose, systematic exploration of architectural design decisions remains limited. This study conducts a comprehensive investigation into these design choices within the broad category of encoder–predictor architectures. The methodology identifies robust architectures applicable to both time series and spectrogram input representations, both of which leverage structured state space models as integral components. Without further hyperparameter adjustments, the proposed models S4Sleep(spec) and S4Sleep(ts) consistently surpass all existing approaches on the most commonly used benchmark datasets: Sleep EDF, the Montreal Archive of Sleep Studies, and, most notably, the extensive Sleep Heart Health Study dataset. The architectural insights derived from this research, along with the refined methodology for architecture search demonstrated herein, are expected to not only advance future research in sleep staging but also be beneficial for other time series annotation tasks.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"187 ","pages":"Article 109735"},"PeriodicalIF":7.0,"publicationDate":"2025-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143180257","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}