Emanuele Nardone , Claudio De Stefano , Nicole Dalia Cilia , Francesco Fontanella
{"title":"Handwriting strokes as biomarkers for Alzheimer’s disease prediction: A novel machine learning approach","authors":"Emanuele Nardone , Claudio De Stefano , Nicole Dalia Cilia , Francesco Fontanella","doi":"10.1016/j.compbiomed.2025.110039","DOIUrl":"10.1016/j.compbiomed.2025.110039","url":null,"abstract":"<div><div>In recent years, machine learning-based handwriting analysis has emerged as a valuable tool for supporting the early diagnosis of Alzheimer’s disease and predicting its progression. Traditional approaches represent handwriting tasks using a single feature vector, where each feature is computed as the mean over elementary handwriting traits or strokes. We propose a novel approach that analyzes each stroke individually, preserving fine-grained movement information that is critical for detecting subtle handwriting changes that may indicate cognitive decline. We evaluated this method on 34 handwriting tasks collected from 174 participants, extracting dynamic and static features from both on-paper and in-air movements. Using a machine learning framework including classification strategies, feature selection techniques, and ensemble methods like ranking-based and stacking approaches, we were able to effectively model stroke-level variations. The ranking-based ensemble achieved the highest accuracy of 80.18% using all features while stacking performed best for in-air movements with 76.67% accuracy. Feature importance analysis through SHAP revealed that certain tasks, particularly sentence writing under dictation, were consistently more predictive. The experimental results demonstrate the effectiveness of our stroke-level analysis approach, which outperformed aggregated statistical methods on 24 out of 34 handwriting tasks, validating the diagnostic value of examining individual movement patterns.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"190 ","pages":"Article 110039"},"PeriodicalIF":7.0,"publicationDate":"2025-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143734605","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}
Berhanu Boru Bifftu , Susan J. Thomas , Khin Than Win
{"title":"Users’ positive attitudes, perceived usefulness, and intentions to use digital mental health interventions: A systematic literature review and meta-analysis","authors":"Berhanu Boru Bifftu , Susan J. Thomas , Khin Than Win","doi":"10.1016/j.compbiomed.2025.110080","DOIUrl":"10.1016/j.compbiomed.2025.110080","url":null,"abstract":"<div><h3>Background</h3><div>Digital Mental Health Interventions (DMHIs) hold significant potential in addressing gaps in mental health treatment, enhancing mental health literacy, and mitigating associated stigma. However, DMHIs have not been systematically evaluated in terms of potential users’ attitudes, perceived usefulness, and intentions to use. Thus, this study aims to consolidate evidence to ascertain users' attitudes, perceived usefulness, and intentions to utilize DMHIs.</div></div><div><h3>Methods</h3><div>The meta-analysis reports adhere to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline. A comprehensive search of databases: Medline, CINHAL, PsycINFO, SCOPUS, and Web of Science, was conducted. As part of the screening process, Covidence database management software was used. Metaprop command was used to calculate the outcome using a random-effects model. Heterogeneity was assessed using Cochrane chi-square (χ2) and the index of heterogeneity (I<sup>2</sup> statistics) test. Sensitivity test and subgroup analysis were performed. Publication bias was examined by funnel plots and Egger's test.</div></div><div><h3>Results</h3><div>In total, 26 studies were analyzed, including data from 13,923 participants. The overall percentage of users' positive attitudes, perceived usefulness, and intentions to use DHMIs was 0.66 (95 % CI; 0.52, 0.79), 0.73 (95 % CI; 0.64, 0.81), and 0.67 (95 % CI; 0.6, 0.74), respectively. Significant heterogeneity was observed; nonetheless, sensitivity analyses indicated that none of the included individual studies exerted undue influence on the overall pooled prevalence. Assessment of funnel plots and Egger's test (p ≤ 0.895) showed no evidence of publication bias.</div></div><div><h3>Conclusion</h3><div>The results of this meta-analysis indicate that, overall, two-thirds of participants have a positive attitude toward DMHIs, around three-quarters find DMHIs useful, and around two-thirds intend to use them. The findings suggest the need to target users' positive attitudes, perceived utility, and willingness for the improved adoption and sustained use of DMHIs.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"190 ","pages":"Article 110080"},"PeriodicalIF":7.0,"publicationDate":"2025-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143725957","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}
Yan Liu , Chen Li , Long-Chen Shen , He Yan , Guo Wei , Robin B. Gasser , Xiaohua Hu , Jiangning Song , Dong-Jun Yu
{"title":"scRCA: A Siamese network-based pipeline for annotating cell types using noisy single-cell RNA-seq reference data","authors":"Yan Liu , Chen Li , Long-Chen Shen , He Yan , Guo Wei , Robin B. Gasser , Xiaohua Hu , Jiangning Song , Dong-Jun Yu","doi":"10.1016/j.compbiomed.2025.110068","DOIUrl":"10.1016/j.compbiomed.2025.110068","url":null,"abstract":"<div><div>Accurate cell type annotation is fundamentally critical for single-cell sequencing (scRNA-seq) data analysis to provide insightful knowledge of tissue-specific cell heterogeneity and cell state transition tracking. Cell type annotation is usually conducted by comparative analysis with known data (i.e., reference) – which contains a presumably accurate representation of cell types. However, this assumption is often problematic, as factors such as human errors in wet-lab experiments and methodological limitations can introduce annotation errors in the reference dataset. As current pipelines for single-cell transcriptomic analysis do not adequately consider this challenge, there is a major demand for constructing a computational pipeline that achieves high-quality cell type annotation using reference datasets containing inherent errors (referred to as “noise” in this study). Here, we built a Siamese network-based pipeline, termed scRCA, to accurately annotate cell types based on noisy reference data. To help users evaluate the reliability of scRCA annotations, an interpreter was also developed to explore the factors underlying the model's predictions. Our experiments demonstrate that, across 14 datasets, scRCA outperformed other widely adopted reference-based methods for cell type annotation. Using an independent dataset of four multiple myeloma patients, we further illustrated that scRCA can distinguish cancerous cells based on gene expression levels and identify genes closely associated with multiple myeloma through scRCA's interpretable module, providing significant information for subsequent clinical treatments. With these advancements, we anticipate that scRCA will serve as a practical reference-based approach for accurate annotating cell type annotation.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"190 ","pages":"Article 110068"},"PeriodicalIF":7.0,"publicationDate":"2025-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143725431","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}
Margarita Zachariou , Eleni M. Loizidou , George M. Spyrou
{"title":"Topological influence of immediate-early genes in brain genetic networks and their link to Alzheimer's disease","authors":"Margarita Zachariou , Eleni M. Loizidou , George M. Spyrou","doi":"10.1016/j.compbiomed.2025.110043","DOIUrl":"10.1016/j.compbiomed.2025.110043","url":null,"abstract":"<div><div>Immediate-early genes (IEGs), a subset of activity-regulated genes (ARGs), are rapidly and transiently activated by neuronal activity independent of protein synthesis. While extensively researched, the role of IEGs within genetic networks and their potential as drug targets for brain diseases remain underexplored. This study aimed to investigate the topological influence of IEGs within genetic networks and explore their relevance to Alzheimer's disease (AD).</div><div>To achieve this, we employed a multi-step approach: mouse ARG data were analysed and mapped to human genes to identify the topological properties that distinguish IEGs from other ARGs; the involvement of ARGs in biological pathways and diseases and their mutational constraints were examined; ARG-related variants in AD were assessed using genome-wide association study (GWAS) summary statistics and functional analysis; and network and GWAS findings were integrated to identify ARG-AD-associated genes.</div><div>Our key findings were: (1) IEGs exhibit significantly higher topological influence across human and mouse gene networks compared to other ARGs; (2) ARGs are less frequently involved in diseases and exhibit higher mutational constraint than non-ARGs; (3) Several AD-associated variants are located in ARG regions, particularly in <em>MARK4</em> near <em>FOSB</em>, with an AD risk eQTL that increases <em>MARK4</em> expression in cortical areas; (4) <em>MARK4</em> emerges as a key node in a dense AD multi-omic network and exhibits a high druggability score.</div><div>These findings underscore the influential role of IEGs within genetic networks, providing valuable insights into their potential as intervention points for diseases characterised by downstream dysregulation, with <em>MARK4</em> emerging as a promising and underexplored target for AD.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"190 ","pages":"Article 110043"},"PeriodicalIF":7.0,"publicationDate":"2025-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143725428","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":"Unsupervised domain adaptation with multi-level distillation boost and adaptive mask for medical image segmentation","authors":"Yongze Wang, Lei Fan, Maurice Pagnucco, Yang Song","doi":"10.1016/j.compbiomed.2025.110055","DOIUrl":"10.1016/j.compbiomed.2025.110055","url":null,"abstract":"<div><div>The mean-teacher (MT) framework has emerged as a commonly used approach in unsupervised domain adaptation (UDA) tasks. Existing methods primarily focus on aligning the outputs of the student and teacher networks by using guidance from the teacher network’s multi-layer features. To build upon the potential of the MT framework, we propose a framework named <em>Multi-Level Distillation Boost (MLDB)</em>. It combines Self-Knowledge Distillation and Dual-Directional Knowledge Distillation to align predictions between the intermediate and high-level features of the student and teacher networks. Additionally, considering the complex variability in anatomical structures, foregrounds, and backgrounds across different domains of medical images, we introduce an <em>Adaptive Masked Image Consistency (AMIC)</em> approach. It provides a customized masking strategy to augment images for source and target domain datasets, using varying mask ratios and sizes to improve the adaptability and efficacy of data augmentation. Our experiments on fundus and polyp datasets indicate that the proposed methods achieve competitive performances of 95.2%/86.1% and 97.3%/89.0% Dice scores for optic disc/cup on REFUGE<span><math><mo>→</mo></math></span>RIM, REFUGE<span><math><mo>→</mo></math></span>Drishti-GS, and 78.3% and 86.2% for polyp on Kvasir<span><math><mo>→</mo></math></span>ETIS and Kvasir<span><math><mo>→</mo></math></span>Endo, respectively. The code is available at <span><span>https://github.com/Yongze/MLDB_AMIC</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"190 ","pages":"Article 110055"},"PeriodicalIF":7.0,"publicationDate":"2025-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143734606","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}
Asmaa Sabet Anwar , Khaled Amin , Mohiy M. Hadhoud , Mina Ibrahim
{"title":"ResTransUNet: A hybrid CNN-transformer approach for liver and tumor segmentation in CT images","authors":"Asmaa Sabet Anwar , Khaled Amin , Mohiy M. Hadhoud , Mina Ibrahim","doi":"10.1016/j.compbiomed.2025.110048","DOIUrl":"10.1016/j.compbiomed.2025.110048","url":null,"abstract":"<div><h3>Background and Objective:</h3><div>Accurate medical tumor segmentation is critical for early diagnosis and treatment planning, significantly improving patient outcomes. This study aims to enhance liver and tumor segmentation from CT and liver images by developing a novel model, ResTransUNet, which combines convolutional and transformer blocks to improve segmentation accuracy.</div></div><div><h3>Methods:</h3><div>The proposed ResTransUNet model is a custom implementation inspired by the TransUNet architecture, featuring a Standalone Transformer Block and ResNet50 as the backbone for the encoder. The hybrid architecture leverages the strengths of Convolutional Neural Networks (CNNs) and Transformer blocks to capture both local features and global context effectively. The encoder utilizes a pre-trained ResNet50 to extract rich hierarchical features, with key feature maps to preserved it as skip connections. The Standalone Transformer Block, integrated into the model, employs multi-head attention mechanisms to capture long-range dependencies across the image, enhancing segmentation performance in complex cases. The decoder reconstructs the segmentation mask by progressively upsampling encoded features while integrating skip connections, ensuring both semantic information and spatial details are retained. This process culminates in a precise binary segmentation mask that effectively distinguishes liver and tumor regions.</div></div><div><h3>Results:</h3><div>The ResTransUNet model achieved superior Dice Similarity Coefficient (DSC) for liver segmentation (98.3% on LiTS and 98.4% on 3D-IRCADb-01) and for tumor segmentation from CT images (94.7% on LiTS and 89.8% on 3D-IRCADb-01) as well as from liver images (94.6% on LiTS and 91.1% on 3D-IRCADb-01). The model also demonstrated high precision, sensitivity, and specificity, outperforming current state-of-the-art methods in these tasks.</div></div><div><h3>Conclusions:</h3><div>The ResTransUNet model demonstrates robust and accurate performance in complex medical image segmentation tasks, particularly in liver and tumor segmentation. These findings suggest that ResTransUNet has significant potential for improving the precision of surgical interventions and therapy planning in clinical settings.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"190 ","pages":"Article 110048"},"PeriodicalIF":7.0,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143725427","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}
Jisun Hong , Jihun Lee , Daegil Choi , Jaehyo Jung
{"title":"Depression level prediction via textual and acoustic analysis","authors":"Jisun Hong , Jihun Lee , Daegil Choi , Jaehyo Jung","doi":"10.1016/j.compbiomed.2025.110009","DOIUrl":"10.1016/j.compbiomed.2025.110009","url":null,"abstract":"<div><div>Extensive research on automatic depression diagnosis has utilized video data to capture related cues, but data collection is challenging because of privacy concerns. By contrast, voice data offer a less-intrusive assessment method and can be analyzed for features such as simple tones, the expression of negative emotions, and a focus on oneself. Recent advancements in multimodal depression-level prediction using speech and text data have gained traction, but most studies overlook the temporal alignment of these modalities, limiting their analysis of the interaction between speech content and intonation. To overcome these limitations, this study introduces timestamp-integrated multimodal encoding for depression (TIMEX-D) which synchronizes the acoustic features of human speech with corresponding text data to predict depression levels on the basis of their relationship. TIMEX-D comprises three main components: a timestamp extraction block that extracts timestamps from speech and text, a multimodal encoding block that extends positional encoding from transformers to mimic human speech recognition, and a depression analysis block that predicts depression levels while reducing model complexity compared with existing transformers. In experiments using the DAIC-WOZ and EDAIC datasets, TIMEX-D achieved accuracies of 99.17 % and 99.81 %, respectively, outperforming previous methods by approximately 13 %. The effectiveness of TIMEX-D in predicting depression levels can enhance mental health diagnostics and monitoring across various contexts.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"190 ","pages":"Article 110009"},"PeriodicalIF":7.0,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143714282","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}
C. Sreedhar , K. Mahesh babu , Suresh Kallam , G.S. Pradeep Ghantasala , S. Anthoniraj , S. Kumarganesh , K. Martin Sagayam , Binay Kumar Pandey , Digvijay Pandey
{"title":"Enhancing healthcare data security using RFE and CRHSM for big data","authors":"C. Sreedhar , K. Mahesh babu , Suresh Kallam , G.S. Pradeep Ghantasala , S. Anthoniraj , S. Kumarganesh , K. Martin Sagayam , Binay Kumar Pandey , Digvijay Pandey","doi":"10.1016/j.compbiomed.2025.110063","DOIUrl":"10.1016/j.compbiomed.2025.110063","url":null,"abstract":"<div><div>Providing security to the medical big data stored in healthcare cloud systems is the most exciting and demanding task in the present day. Many researchers use cryptographic techniques to protect big data against malicious users/attacks in cloud environments. Still, they face the problems of high complexity in operations, increased time consumption, storage overhead, and lack of efficiency. Hence, this paper aims to develop a new extensive data security framework for improving the reliability of healthcare systems. The main contribution of this work is to introduce a novel Cohesive Random-Hash based Security Model (CRHSM) for securing both user authentication and medical records. Here, the hospital environment is considered with the entities of patients, healthcare professionals, and hospital servers. In this environment, the registered entities can only participate in communication, including system initialization, registration, login, authentication, encryption, and decryption modules. First, the entities must register with the hospital server to obtain the smart card for further communications. Here, the Squirrel Search Optimization (SSO) technique generates the random number used for the registration phase. During the login and authentication module, the legitimacy of patients and healthcare professionals is validated based on the authentication parameters. Moreover, the medical records are encrypted before storing them in the cloud systems using an efficient Reformist Feistel Encryption (RFE) mechanism. Moreover, various evaluation parameters are considered for assessing the performance of the proposed model, and the evaluated values are compared with security approaches to validate the effectiveness of the proposed scheme.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"190 ","pages":"Article 110063"},"PeriodicalIF":7.0,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143714276","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":"Development and Multi-center validation of a machine learning Model for advanced colorectal neoplasms screening","authors":"Mingqing Zhang , Yongdan Zhang , Lizhong Zhao , Haoren Jing , Xinyu Gao , Tianhao Li , Zhicheng Pu , Shiwu Zhang , Xipeng Zhang","doi":"10.1016/j.compbiomed.2025.110066","DOIUrl":"10.1016/j.compbiomed.2025.110066","url":null,"abstract":"<div><h3>Background</h3><div>In colorectal cancer (CRC) screening programs, accurately identifying individuals at high risk for advanced colorectal neoplasia (ACN) is essential as they require further colonoscopy, early intervention, and monitoring follow-up. This study aimed to develop a machine learning (ML)-based risk prediction model, serving as an effective tool for the early identification of high-risk individuals for ACN.</div></div><div><h3>Methods</h3><div>This study analyzed data from the Tianjin CRC screening program. The dataset from 2012 to 2022 was divided into a training set and 11 validation sets across 12 medical centers. The 2023 data was used as an independent temporal external validation set. First, the least absolute shrinkage and selection operator and logistic regression (LR) analysis were used to select significant features. Next, six ML models were constructed to predict ACN and validate its predictive capability on the validation sets. Among the classifiers, the best-performing model, Tianjin ML (TML), was selected, and its performance was compared with the Asia Pacific Colorectal Screening (APCS) and LR. Moreover, we developed a stacked ensemble learning model to improve the prediction performance for ACN. Finally, we conducted an interpretability analysis using SHapley Additive exPlanations (SHAP) values and deployed a web application tool based on the Streamlit framework.</div></div><div><h3>Findings</h3><div>Among the trained models, the TML achieved the best performance, with an area under the curve (AUC) of 0.690, a sensitivity of 0.649, a specificity of 0.626, an F1 score of 0.320, and an accuracy of 0.629. Furthermore, the TML performed well in 11 validation sets and the independent temporal external validation set. A predictive probability threshold of 0.140 was identified for stratifying individuals into low- and high-risk groups. The TML exhibited superior performance compared to APCS and LR. The stacked ensemble learning model, S-TML, further improved the AUC to 0.709. Additionally, SHAP analysis identified age, gender, and fecal immunochemical tests as the top three predictive factors for ACN.</div></div><div><h3>Interpretation</h3><div>The TML outperformed traditional models, including APCS and LR, in predicting ACN and could serve as a screening decision support tool to identify high–risk individuals for ACN.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"190 ","pages":"Article 110066"},"PeriodicalIF":7.0,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143714281","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}
Griffani Megiyanto Rahmatullah , Shanq-Jang Ruan , I. Wayan Wiprayoga Wisesa , Lieber Po-Hung Li
{"title":"Enhancing visual speech perception through deep automatic lipreading: A systematic review","authors":"Griffani Megiyanto Rahmatullah , Shanq-Jang Ruan , I. Wayan Wiprayoga Wisesa , Lieber Po-Hung Li","doi":"10.1016/j.compbiomed.2025.110019","DOIUrl":"10.1016/j.compbiomed.2025.110019","url":null,"abstract":"<div><div>Communication involves exchanging information between individuals or groups through various media sources. However, limitations such as hearing loss can make it difficult for some individuals to understand the information delivered during speech communication. Conventional methods, including sign language, written text, and manual lipreading, offer some solutions; however, emerging software-based tools using artificial intelligence (AI) are introducing more effective approaches. Many approaches rely on AI to improve communication quality, with the current trend of leveraging deep learning being a particularly effective tool. This paper presents a comprehensive Systematic Literature Review (SLR) of research trends in automatic lipreading technologies, a critical field in enhancing communication among individuals with hearing impairments. The SLR, which followed the Preferred Reporting Items for Systematic Literature Review and Meta-Analysis (PRISMA) protocol, identified 114 original research articles published between 2014 and mid-2024. The essential information from these articles was summarized, including the trends in automatic lipreading research, dataset availability, task categories, existing approaches, and architectures for automatic lipreading systems. The results showed that various techniques and advanced deep learning models achieved convincing performance to become state-of-the-art in automatic lipreading tasks. However, several challenges, such as insufficient data quantity, inadequate environmental conditions, and language diversity, must be resolved in the future. Furthermore, many improvements have been made to the deep learning models to overcome these challenges and become a massive solution, particularly for automatic lipreading tasks in the near future.</div></div>","PeriodicalId":10578,"journal":{"name":"Computers in biology and medicine","volume":"190 ","pages":"Article 110019"},"PeriodicalIF":7.0,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143714284","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}