Yijun Wang, Tongjian Zhu, Tong Zhou, Bing Wu, Wuping Tan, Kezhong Ma, Zhuoya Yao, Jian Wang, Siyang Li, Fanglin Qin, Yannan Xu, Liguo Tan, Jinjun Liu, Jun Wang
{"title":"Hyper-DREAM, a Multimodal Digital Transformation Hypertension Management Platform Integrating Large Language Model and Digital Phenotyping: Multicenter Development and Initial Validation Study.","authors":"Yijun Wang, Tongjian Zhu, Tong Zhou, Bing Wu, Wuping Tan, Kezhong Ma, Zhuoya Yao, Jian Wang, Siyang Li, Fanglin Qin, Yannan Xu, Liguo Tan, Jinjun Liu, Jun Wang","doi":"10.1007/s10916-025-02176-1","DOIUrl":"10.1007/s10916-025-02176-1","url":null,"abstract":"<p><p>Within the mHealth framework, systematic research that collects and analyzes patient data to establish comprehensive digital health archives for hypertensive patients, and leverages large language models (LLMs) to assist clinicians in health management and Blood Pressure (BP) control remains limited. In this study, our aims to describe the design, development and usability evaluation process of a management platform (Hyper-DREAM) for hypertension. Our multidisciplinary team employed an iterative design approach over the course of a year to develop the Hyper-DREAM platform. This platform's primary functionalities encompass multimodal data collection (personal hypertensive digital phenotype archive), multimodal interventions (BP measurement, medication assistance, behavior modification, and hypertension education) and multimodal interactions (clinician-patient engagement and BP Coach component). In August 2024, the mHealth App Usability Questionnaire (MAUQ) was conducted involving 51 hypertensive patients recruited from three distinct centers. In parallel, six clinicians engaged in management activities and contributed feedback via the Doctor's Software Satisfaction Questionnaire (DSSQ). Concurrently, a real-world comparative experiment was conducted to evaluate the usability of the BP Coach, ChatGPT-4o Mini, ChatGPT-4o and clinicians. The comparative experiment demonstrated that the BP Coach achieved significantly higher scores in utility (mean scores 4.05, SD 0.87) and completeness (mean scores 4.12, SD 0.78) when compared to ChatGPT-4o Mini, ChatGPT-4o, and clinicians. In terms of clarity, the BP Coach was slightly lower than clinicians (mean scores 4.03, SD 0.88). In addition, the BP Coach exhibited lower performance in conciseness (mean scores 3.00, SD 0.96). Clinicians reported a marked improvement in work efficiency (2.67 vs. 4.17, P < .001) and experienced faster and more effective patient interactions (3.0 vs. 4.17, P = .004). Furthermore, the Hyper-DREAM platform significantly decreased work intensity (2.5 vs. 3.5, P = .01) and minimized disruptions to daily routines (2.33 vs. 3.55, P = .004). The Hyper-DREAM platform demonstrated significantly greater overall satisfaction compared to the WeChat-based standard management (3.33 vs. 4.17, P = .01). Additionally, clinicians exhibited a markedly higher willingness to integrate the Hyper-DREAM platform into clinical practice (2.67 vs. 4.17, P < .001). Furthermore, patient management time decreased from 11.5 min (SD 1.87) with Wechat-based standard management to 7.5 min (SD 1.84, P = .01) with Hyper-DREAM. Hypertensive patients reported high satisfaction with the Hyper-DREAM platform, including ease of use (mean scores 1.60, SD 0.69), system information arrangement (mean scores 1.69, SD 0.71), and usefulness (mean scores 1.57, SD 0.58). In conclusion, our study presents Hyper-DREAM, a novel artificial intelligence-driven platform for hypertension management, designed to alleviate clinicia","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"42"},"PeriodicalIF":3.5,"publicationDate":"2025-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143764140","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Feature Selection in Breast Cancer Gene Expression Data Using KAO and AOA with SVM Classification.","authors":"Abrar Yaqoob, Navneet Kumar Verma","doi":"10.1007/s10916-025-02171-6","DOIUrl":"10.1007/s10916-025-02171-6","url":null,"abstract":"<p><p>Breast cancer classification using gene expression data presents significant challenges due to high dimensionality and complexity. This study introduces a novel hybrid framework integrating the Kashmiri Apple Optimization Algorithm (KAO) and the Armadillo Optimization Algorithm (AOA) for effective feature selection, coupled with Support Vector Machines (SVM) for precise classification. The dual-stage approach leverages KAO for global exploration of informative genes and AOA for refining the selection through local optimization, addressing issues of redundancy and premature convergence. Applied to breast cancer datasets, the proposed method achieved a classification accuracy of 98.97%, precision of 98.46%, recall of 100%, and an F1-score of 99.22% using a subset of 15 genes. The robustness of the framework was validated across varying subset sizes, demonstrating consistent high performance. By optimizing feature relevance and redundancy, the KAO-AOA framework provides a promising tool for gene-based cancer prediction with potential applications to other cancer datasets and real-world clinical use.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"40"},"PeriodicalIF":3.5,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143719889","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Large Language Models' Responses to Spinal Cord Injury: A Comparative Study of Performance.","authors":"Jinze Li, Chao Chang, Yanqiu Li, Shengyu Cui, Fan Yuan, Zhuojun Li, Xinyu Wang, Kang Li, Yuxin Feng, Zuowei Wang, Zhijian Wei, Fengzeng Jian","doi":"10.1007/s10916-025-02170-7","DOIUrl":"10.1007/s10916-025-02170-7","url":null,"abstract":"<p><p>With the increasing application of large language models (LLMs) in the medical field, their potential in patient education and clinical decision support is becoming increasingly prominent. Given the complex pathogenesis, diverse treatment options, and lengthy rehabilitation periods of spinal cord injury (SCI), patients are increasingly turning to advanced online resources to obtain relevant medical information. This study analyzed responses from four LLMs-ChatGPT-4o, Claude-3.5 sonnet, Gemini-1.5 Pro, and Llama-3.1-to 37 SCI-related questions spanning pathogenesis, risk factors, clinical features, diagnostics, treatments, and prognosis. Quality and readability were assessed using the Ensuring Quality Information for Patients (EQIP) tool and Flesch-Kincaid metrics, respectively. Accuracy was independently scored by three senior spine surgeons using consensus scoring. Performance varied among the models. Gemini ranked highest in EQIP scores, suggesting superior information quality. Although the readability of all four LLMs was generally low, requiring a college-level reading comprehension ability, they were all able to effectively simplify complex content. Notably, ChatGPT led in accuracy, achieving significantly higher \"Good\" ratings (83.8%) compared to Claude (78.4%), Gemini (54.1%), and Llama (62.2%). Comprehensiveness scores were high across all models. Furthermore, the LLMs exhibited strong self-correction abilities. After being prompted for revision, the accuracy of ChatGPT and Claude's responses improved by 100% and 50%, respectively; both Gemini and Llama improved by 67%. This study represents the first systematic comparison of leading LLMs in the context of SCI. While Gemini excelled in response quality, ChatGPT provided the most accurate and comprehensive responses.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"39"},"PeriodicalIF":3.5,"publicationDate":"2025-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143700643","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mehmet Ali Gelen, Turker Tuncer, Mehmet Baygin, Sengul Dogan, Prabal Datta Barua, Ru-San Tan, U R Acharya
{"title":"TQCPat: Tree Quantum Circuit Pattern-based Feature Engineering Model for Automated Arrhythmia Detection using PPG Signals.","authors":"Mehmet Ali Gelen, Turker Tuncer, Mehmet Baygin, Sengul Dogan, Prabal Datta Barua, Ru-San Tan, U R Acharya","doi":"10.1007/s10916-025-02169-0","DOIUrl":"10.1007/s10916-025-02169-0","url":null,"abstract":"<p><strong>Background and purpose: </strong>Arrhythmia, which presents with irregular and/or fast/slow heartbeats, is associated with morbidity and mortality risks. Photoplethysmography (PPG) provides information on volume changes of blood flow and can be used to diagnose arrhythmia. In this work, we have proposed a novel, accurate, self-organized feature engineering model for arrhythmia detection using simple, cost-effective PPG signals.</p><p><strong>Method: </strong>We have drawn inspiration from quantum circuits and employed a quantum-inspired feature extraction function /named the Tree Quantum Circuit Pattern (TQCPat). The proposed system consists of four main stages: (i) multilevel feature extraction using discrete wavelet transform (MDWT) and TQCPat, (ii) feature selection using Chi-squared (Chi2) and neighborhood component analysis (NCA), (iii) classification using k-nearest neighbors (kNN) and support vector machine (SVM) and (iv) information fusion.</p><p><strong>Results: </strong>Our proposed TQCPat-based feature engineering model has yielded a classification accuracy of 91.30% using 46,827 PPG signals in classifying six classes with ten-fold cross-validation.</p><p><strong>Conclusion: </strong>Our results show that the proposed TQCPat-based model is accurate for arrhythmia classification using PPG signals and can be tested with a large database and more arrhythmia classes.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"38"},"PeriodicalIF":3.5,"publicationDate":"2025-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11933173/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143700651","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Valentina Bellini, Tania Domenichetti, Elena Giovanna Bignami
{"title":"Innovative Technologies for Smarter and Efficient Operating Room Scheduling.","authors":"Valentina Bellini, Tania Domenichetti, Elena Giovanna Bignami","doi":"10.1007/s10916-025-02168-1","DOIUrl":"10.1007/s10916-025-02168-1","url":null,"abstract":"<p><p>An optimized scheduling system for surgical procedures is considered fundamental for maximizing hospital resource utilization and improving patient outcomes. The integration of Artificial Intelligence (AI) tools and New Technologies is paramount in this project to enable personalized patient care and optimize perioperative clinical pathways. We read with interest the manuscript by Parks et al., which developed a predictive model of surgical case durations. The model appears to adopt a pragmatic approach by analyzing tangible variables and undergoing validation across various types of surgical procedures, which suggests potential avenues for enhancing efficiency and sustainability in healthcare practices. However, we have some observations, particularly regarding the feasibility and practical implementation of the proposed model. A key limitation of the model is the precise definition of surgical duration, which requires further specification. To effectively translate the model into a practical scheduling approach, it is essential to consider total Operating Room (OR) occupancy time as a critical determinant of surgical planning and resource allocation. This includes not only the actual procedural time but also preoperative preparation, anesthesia induction and recovery, cleaning, and material restocking, all of which significantly impact overall scheduling efficiency. Another critical aspect concerns the quality and reliability of the input data, which is fundamental for ensuring the accuracy and effectiveness of the model. Furthermore, the adoption of new technologies should be regarded not merely as an innovation but as a means to develop high-performance, efficient tools that enhance current clinical practice. In this context, machine learning models should not only serve as analytical instruments but also as actionable tools, enabling the transition from predictive insights to strategic planning and optimized scheduling, ultimately improving decision-making and resource allocation. While making accurate predictions is a good starting point, maintaining an active AI model requires investment in resources, such as an increase in the number of surgical cases compared to the current organizational system. It may be beneficial to consider the creation of a multidisciplinary group that could promote the integration of AI with other emerging technologies.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"37"},"PeriodicalIF":3.5,"publicationDate":"2025-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143674211","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Using Generative AI to Extract Structured Information from Free Text Pathology Reports.","authors":"Fahad Shahid, Min-Huei Hsu, Yung-Chun Chang, Wen-Shan Jian","doi":"10.1007/s10916-025-02167-2","DOIUrl":"10.1007/s10916-025-02167-2","url":null,"abstract":"<p><p>Manually converting unstructured text pathology reports into structured pathology reports is very time-consuming and prone to errors. This study demonstrates the transformative potential of generative AI in automating the analysis of free-text pathology reports. Employing the ChatGPT Large Language Model within a Streamlit web application, we automated the extraction and structuring of information from 33 unstructured breast cancer pathology reports from Taipei Medical University Hospital. Achieving a 99.61% accuracy rate, the AI system notably reduced the processing time compared to traditional methods. This not only underscores the efficacy of AI in converting unstructured medical text into structured data but also highlights its potential to enhance the efficiency and reliability of medical text analysis. However, this study is limited to breast cancer pathology reports and was conducted using data obtained from hospitals associated with a single institution. In the future, we plan to expand the scope of this research to include pathology reports for other cancer types incrementally and conduct external validation to further substantiate the robustness and generalizability of the proposed system. Through this technological integration, we aimed to substantiate the capabilities of generative AI in improving both the speed and reliability of data processing. The outcomes of this study affirm that generative AI can significantly transform the handling of pathology reports, promising substantial advancements in biomedical research by facilitating the structured analysis of complex medical data.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"36"},"PeriodicalIF":3.5,"publicationDate":"2025-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11906504/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143624956","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Sadia Azmin Anisha, Arkendu Sen, Badariah Ahmad, Chris Bain
{"title":"Exploring Acceptance of Digital Health Technologies for Managing Non-Communicable Diseases Among Older Adults: A Systematic Scoping Review.","authors":"Sadia Azmin Anisha, Arkendu Sen, Badariah Ahmad, Chris Bain","doi":"10.1007/s10916-025-02166-3","DOIUrl":"10.1007/s10916-025-02166-3","url":null,"abstract":"<p><p>This review explores the acceptance of digital health (DH) technologies for managing non-communicable diseases (NCDs) among older adults (≥ 50 years), with an extended focus on artificial intelligence (AI)-powered conversational agents (CAs) as an emerging notable subset of DH. A systematic literature search was conducted in June 2024 using PubMed, Web of Science, Scopus, and ACM Digital Library. Eligible studies were empirical and published in English between January 2010 and May 2024. Covidence software facilitated screening and data extraction, adhering to PRISMA-ScR guidelines. The screening process finally yielded 20 studies. Extracted data from these selected studies included interventions, participant demographics, technology types, sample sizes, study designs and locations, technology acceptance measures, key outcomes, and methodological limitations. A narrative synthesis approach was used for analysis, revealing four key findings: (1) overall positive attitudes of older adults towards DH acceptance; (2) the Technology Acceptance Model (TAM) and the Unified Theory of Acceptance and Use of Technology (UTAUT) are the most frequently used standard frameworks for evaluating technology acceptance; (3) the key facilitators of technology acceptance include perceived usefulness, ease of use, social influence, and digital or e-health literacy, while barriers involve technical challenges, usability issues, and privacy concerns; (4) the acceptance of AI-based CAs for NCD management among older adults remains inadequately evaluated, possibly due to limited adaptation of established frameworks to specific healthcare contexts and technology innovations. This review significantly contributes to the DH field by providing a comprehensive analysis of technology acceptance for NCD management among older adults, extending beyond feasibility and usability. The findings offer stakeholders valuable insights into how to better integrate these technologies to improve health outcomes and quality of life for older adults. Protocol Registration: PROSPERO (Registration ID: CRD42024540035).</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"35"},"PeriodicalIF":3.5,"publicationDate":"2025-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11897087/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143604884","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Xina Liu, Jun Xie, Junjun Hou, Xinying Xu, Yan Guo
{"title":"D-GET: Group-Enhanced Transformer for Diabetic Retinopathy Severity Classification in Fundus Fluorescein Angiography.","authors":"Xina Liu, Jun Xie, Junjun Hou, Xinying Xu, Yan Guo","doi":"10.1007/s10916-025-02165-4","DOIUrl":"10.1007/s10916-025-02165-4","url":null,"abstract":"<p><p>Early detection of Diabetic Retinopathy (DR) is vital for preserving vision and preventing deterioration of eyesight. Fundus Fluorescein Angiography (FFA), recognized as the gold standard for diagnosing DR, effectively reveals abnormalities in retinal vasculature. Given the labor-intensive and costly nature of manual DR diagnosis, along with its low accuracy, developing a DR classification model based on FFA using deep learning techniques is crucial. Furthermore, DR classification faces challenges such as minimal lesion variance between different disease stages and significant size variations of lesions within the same stage, with small lesions often overlooked by existing models. We propose a deep learning model, D-GET, utilizing a Group-Enhanced Transformer for classifying DR lesion severity in FFA images. The D-GET model incorporates a Full-Scale Transformer Block, where the Group-Focal module captures feature information at multiple scales, from fine details to broader patterns, and adaptively integrates contextual information, enhancing the model's ability to detect small-scale lesions. The model also includes a Channel Adaptive Attention Module (CAAM) that synthesizes channel and spatial information to improve feature detection and localization. Experimental findings indicate that the D-GET method we developed surpasses existing methods on a custom dataset. The D-GET model, developed for DR classification using FFA images, significantly improves the detection of small-scale lesions. This advancement enhances the diagnosis and treatment of DR, establishing a solid foundation for its broader application across various domains of ophthalmic and general medical imaging.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"34"},"PeriodicalIF":3.5,"publicationDate":"2025-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143567243","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
José Antonio Rivas-Navarrete, Humberto Pérez-Espinosa, A L Padilla-Ortiz, Ansel Y Rodríguez-González, Diana Cristina García-Cambero
{"title":"Edge Computing System for Automatic Detection of Chronic Respiratory Diseases Using Audio Analysis.","authors":"José Antonio Rivas-Navarrete, Humberto Pérez-Espinosa, A L Padilla-Ortiz, Ansel Y Rodríguez-González, Diana Cristina García-Cambero","doi":"10.1007/s10916-025-02154-7","DOIUrl":"10.1007/s10916-025-02154-7","url":null,"abstract":"<p><p>Chronic respiratory diseases affect people worldwide, but conventional diagnostic methods may not be accessible in remote locations far from population centers. Sounds from the human respiratory system have displayed potential in autonomously detecting these diseases using artificial intelligence (AI). This article outlines the development of an audio-based edge computing system that automatically detects chronic respiratory diseases (CRDs). The system utilizes machine learning (ML) techniques to analyze audio recordings of respiratory sounds (cough and breath) and classify the presence or absence of these diseases, using features such as Mel frequency cepstral coefficients (MFCC) and chromatic attributes (chromagram) to capture the relevant acoustic features of breath sounds. The system was trained and tested using a dataset of respiratory sounds collected from 86 individuals. Among them, 53 had chronic respiratory conditions, including asthma and chronic obstructive pulmonary disease (COPD), while the remaining 33 were healthy. The system's final evaluation was conducted with a group of 13 patients and 22 healthy individuals. Our approach demonstrated high sensitivity and specificity in the classification of sounds on edge devices, including smartphone and Raspberry Pi. Our best results for CRDs reached a sensitivity of 90.0%, a specificity of 93.55%, and a balanced accuracy of 91.75% for accurately identifying individuals with both healthy and diseased. These results showcase the potential of edge computing and machine learning systems in respiratory disease detection. We believe this system can contribute to developing efficient and cost-effective screening tools.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"33"},"PeriodicalIF":3.5,"publicationDate":"2025-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143542270","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Paul G Mayo, Kenneth I Vaden, Lois J Matthews, Judy R Dubno
{"title":"Feature-Based Audiogram Value Estimator (FAVE): Estimating Numerical Thresholds from Scanned Images of Handwritten Audiograms.","authors":"Paul G Mayo, Kenneth I Vaden, Lois J Matthews, Judy R Dubno","doi":"10.1007/s10916-025-02146-7","DOIUrl":"10.1007/s10916-025-02146-7","url":null,"abstract":"<p><p>Hearing loss is a public health concern that affects millions of people globally. Clinically, a person's hearing sensitivity is often measured using pure-tone audiometry and visualized as a pure-tone audiogram, a plot of hearing sensitivity as a function of frequency. Digital test equipment allows clinicians to store audiograms as numerical values, though some practices write audiograms by hand and store them as digital images in electronic health records systems. This leaves the numerical values inaccessible to public-health researchers unless manually interpreted. Therefore, this study developed machine-learning models for estimating numerical threshold values from scanned images of handwritten audiograms. Training data were a novel set of 556 handwritten audiograms from a longitudinal cohort study of age-related hearing loss. The models were sliding-window, single-class object detectors based on Aggregate Channel Features, altogether called Feature-based Audiogram Value Estimator or \"FAVE\". Model accuracy was determined using symbol location accuracy and comparing estimated numerical threshold values to known values from an electronic database. FAVE resulted in an average of 87.0% recall and 96.2% precision for symbol locations. The numerical threshold values were less accurate, with 78.3% of estimations having no error, though threshold estimates were not significantly different from true thresholds. Threshold estimation was more accurate than pre-trained deep learning approaches for the current test set. Future work should consider implementing detectors with similar image channels and identify limitations on symbol and axis tick label detection.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"32"},"PeriodicalIF":3.5,"publicationDate":"2025-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12034326/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143515980","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}