Journal of Medical Systems最新文献

{"title":"Innovative Limitations and Prospective Recommendations on \"Assessing the Efficiency of Non-Operating Room Anesthesia (NORA) Using Performance Boundaries\".","authors":"Zilin Zhao, Fei Xu, Hejia Wan","doi":"10.1007/s10916-025-02251-7","DOIUrl":"https://doi.org/10.1007/s10916-025-02251-7","url":null,"abstract":"","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"114"},"PeriodicalIF":5.7,"publicationDate":"2025-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145054224","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":"Bimodal ECG and PCG Cardiovascular Disease Detection: Exploring the Potential and Modality Contribution.","authors":"Alessia Calzoni, Mattia Savardi, Marco Silvestri, Sergio Benini, Alberto Signoroni","doi":"10.1007/s10916-025-02245-5","DOIUrl":"10.1007/s10916-025-02245-5","url":null,"abstract":"<p><p>Early detection of cardiovascular diseases (CVDs) is crucial for improving patient outcomes and alleviating healthcare burdens. Electrocardiograms (ECGs) and phonocardiograms (PCGs) offer low-cost, non-invasive, and easily integrable solutions for preventive care settings. In this work, we propose a novel bimodal deep learning model that combines ECG and PCG signals to enhance the early detection of CVDs. To address the challenge of limited bimodal data, we fine-tuned a Convolutional Neural Network (CNN) pre-trained on large-scale audio recordings, leveraging all publicly available unimodal PCG datasets. This PCG branch was then integrated with a 1D-CNN ECG branch via late fusion. Evaluated on an augmented version of MITHSDB, currently the only publicly available bimodal dataset, our approach achieved an AUROC of 96.4%, significantly outperforming ECG-only and PCG-only models by approximately 3%pts and 11%pts, respectively. To interpret the model's decisions, we applied three explainability techniques, quantifying the relative contributions of the electrical and acoustic features. Furthermore, by projecting the learned embeddings into two dimensions using UMAP, we revealed clear separation between normal and pathological samples. Our results conclusively demonstrate that combining ECG and PCG modalities yields substantial performance gains, with explainability and visualization providing critical insights into model behavior. These findings underscore the importance of multimodal approaches for CVDs diagnosis and prevention, and strongly motivate the collection of larger, more diverse bimodal datasets for future research.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"113"},"PeriodicalIF":5.7,"publicationDate":"2025-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12432067/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145040391","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}

{"title":"Prediction Model of Intradialytic Hypertension in Hemodialysis Patients Based on Machine Learning.","authors":"Yu Wang, Hongming Zhou, Qi Guo, Kang Wang, Yehua Luo, Shaodong Luan, Donge Tang, Shuangyong Dong, Lianghong Yin, Yong Dai","doi":"10.1007/s10916-025-02237-5","DOIUrl":"https://doi.org/10.1007/s10916-025-02237-5","url":null,"abstract":"<p><p>The escalating global burden of chronic kidney disease (CKD), particularly end-stage renal disease (ESRD), has intensified reliance on hemodialysis (HD), imposing substantial financial and operational burdens on healthcare systems and patients. Intradialytic hypertension (IDH), a critical complication during HD, is associated with life-threatening cardiovascular and neurological sequelae if unmanaged. This study aims to develop a machine learning (ML)-driven early-alert system for IDH risk prediction by integrating demographic profiles and dialysis session records, enabling clinicians to preemptively identify high-risk patients and prioritize targeted monitoring. Two clinical prediction models (IDH-1 and IDH-2) were developed using Light Gradient Boosting Machine (LGBM), Support Vector Machine (SVM), and TabNet algorithms. IDH-1 estimates immediate hypertension risk by analyzing pre-dialysis vital signs and longitudinal treatment patterns, whereas IDH-2 predicts subsequent session risks by synthesizing real-time dialysis parameters with historical biomarkers. Model performance was rigorously validated using standardized metrics, including AUC-ROC, sensitivity, accuracy, and F1 score, to ensure clinical applicability. 185,125 HD sessions as training set and 71,427 sessions as testing set were used in this study. For IDH-1, the LGBM model demonstrated superior discriminative capacity (AUC: 0.87; recall: 0.73; F1 score: 0.36), outperforming SVM and TabNet. Similarly, LGBM achieved the highest performance for IDH-2 (AUC: 0.74; recall: 0.56; F1 score: 0.26). Most significant parameters in IDH-1 Predictor with LGBM were pre-dialysis diastolic pressures, historical mean arterial pressure, and historical average IDH episodes. For the IDH-2 model with LGBM, historical average IDH episodes and post-dialysis systolic pressures were most important parameters. This study provides two kinds of superior discriminative capacity LGBM model for IDH predicting. The proposed models offer a scalable framework for personalized risk stratification, potentially mitigating adverse outcomes in hemodialysis populations.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"112"},"PeriodicalIF":5.7,"publicationDate":"2025-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145033427","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":"Empowering Cancer Patients: A Scoping Review on Gamified Approaches To Health Literacy for Self-Care.","authors":"Filipe Cerqueira, Marta Campos Ferreira, Maria Joana Campos, Carla Silvia Fernandes","doi":"10.1007/s10916-025-02241-9","DOIUrl":"10.1007/s10916-025-02241-9","url":null,"abstract":"<p><p>To address the challenges of self-care in oncology, gamification emerges as an innovative strategy to enhance health literacy and self-care among individuals with oncological disease. This study aims to explore and map how gamification can promote health literacy for self-care of oncological diseases. A scoping review was conducted following the Joanna Briggs Institute guidelines and the PRISMA-ScR Checklist developed for scoping reviews. A comprehensive search strategy was employed across MEDLINE^®, CINAHL^®, Scopus^®, and Web of Science^® databases, with keywords focusing on oncological patients and gamification tools applied to self-management, from inception to December 2023. Thirty studies published between 2011 and 2023 were included, with a total of 1,118 reported participants. Most interventions (n = 21) focused on the development of mobile applications. The most frequent gamification elements included customizable avatars, rewards, social interaction, quizzes, and personalized feedback. The interventions primarily targeted health literacy and patient education, symptom monitoring, management of side effects, pain control, and adherence to medication and nutrition regimens. The integration of gamification elements into digital health solutions for oncology is expanding and holds promises for supporting health literacy and self-care. Further studies, preferably longitudinal, are needed to assess the effectiveness and impact of these interventions across different oncological populations and clinical settings.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"111"},"PeriodicalIF":5.7,"publicationDate":"2025-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12417285/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145023473","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}

{"title":"How Well Do ChatGPT and Claude Perform in Study Selection for Systematic Review in Obstetrics.","authors":"Suppachai Insuk, Kansak Boonpattharatthiti, Chimbun Booncharoen, Panitnan Chaipitak, Muhammed Rashid, Sajesh K Veettil, Nai Ming Lai, Nathorn Chaiyakunapruk, Teerapon Dhippayom","doi":"10.1007/s10916-025-02246-4","DOIUrl":"10.1007/s10916-025-02246-4","url":null,"abstract":"<p><p>The use of generative AI in systematic review workflows has gained attention for enhancing study selection efficiency. However, evidence on its screening performance remains inconclusive, and direct comparisons between different generative AI models are still limited. The objective of this study is to evaluate the performance of ChatGPT-4o and Claude 3.5 Sonnet in the study selection process of a systematic review in obstetrics. A literature search was conducted using PubMed, EMBASE, Cochrane CENTRAL, and EBSCO Open Dissertations from inception till February 2024. Titles and abstracts were screened using a structured prompt-based approach, comparing decisions by ChatGPT, Claude and junior researchers with decisions by an experienced researcher serving as the reference standard. For the full-text review, short and long prompt strategies were applied. We reported title/abstract screening and full-text review performances using accuracy, sensitivity (recall), precision, F1-score, and negative predictive value. In the title/abstract screening phase, human researchers demonstrated the highest accuracy (0.9593), followed by Claude (0.9448) and ChatGPT (0.9138). The F1-score was the highest among human researchers (0.3853), followed by Claude (0.3724) and ChatGPT (0.2755). Negative predictive value (NPV) was high across all screeners: ChatGPT (0.9959), Claude (0.9961), and human researchers (0.9924). In the full-text screening phase, ChatGPT with a short prompt achieved the highest accuracy (0.904), highest F1-score (0.90), and NPV of 1.00, surpassing the performance of Claude and human researchers. Generative AI models perform close to human levels in study selection, as evidenced in obstetrics. Further research should explore their integration into evidence synthesis across different fields.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"110"},"PeriodicalIF":5.7,"publicationDate":"2025-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144992843","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":"Smartphone-Based Digital Health Interventions: A Comprehensive Systematic Review of Efficacy for Cardiovascular and Cerebrovascular Outcomes.","authors":"Swathi Suresh, Sujas Bhardwaj, Hansel Chris Rodrigues, Dhanya Charly, Tufayl Ahmed Mohammed Shekha, Angeline Jessy, Anupa Anirudhan, Abhishek Menesgere, Thomas Gregor Issac","doi":"10.1007/s10916-025-02236-6","DOIUrl":"https://doi.org/10.1007/s10916-025-02236-6","url":null,"abstract":"<p><p>Digital health interventions (DHIs) have the potential to transform the management of cardiovascular disease (CVD) and cerebrovascular disease (CeVD) by addressing their risk factors, including hypertension, diabetes, obesity, dyslipidemia, and physical inactivity. Despite this potential, there remains a need for a comprehensive understanding of their clinical efficacy, implementation challenges, and usability features to integrate them effectively into healthcare frameworks. This systematic review aims to evaluate the clinical efficacy, functionalities, and barriers associated with smartphone-based DHIs for managing CVD and CeVD. The protocol was first registered in PROSPERO, with the registration number CRD42024570866. A comprehensive search was conducted across PubMed, ScienceDirect, Web of Science, and DOAJ. The study selection adhered to the PICO framework, and methodological quality was assessed using Cochrane's RoB 2 tool for randomized controlled trials (RCTs) and ROBINS-I for non-RCTs. A total of 35 studies were included, primarily consisting of 2-arm RCTs (57%). The studies were largely conducted in high-income countries, with limited representation from low- and middle-income regions. Significant improvements in physical activity were observed, while clinical outcomes like blood pressure, glucose levels, lipid profiles, and body weight showed mixed results, with several changes not reaching statistical significance. This review underscores the potential of DHIs in managing CVD and CeVD risk factors. However, the variability in clinical outcomes highlights the need for tailored, personalized interventions, longer study durations, and strategies to improve retention and engagement for optimizing efficacy and scalability.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"109"},"PeriodicalIF":5.7,"publicationDate":"2025-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144957490","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":"Applying the Model for Assessing the Value of AI (MAS-AI) Framework To Organizational AI: A Case Study of Surgical Scheduling Assessment in Italy.","authors":"Valentina Bellini, Francesco Calabrò, Elena Bignami, Tudor Mihai Haja, Iben Fasterholdt, Benjamin Sb Rasmussen, Rossana Cecchi","doi":"10.1007/s10916-025-02235-7","DOIUrl":"https://doi.org/10.1007/s10916-025-02235-7","url":null,"abstract":"<p><p>This work aims to explore the transferability of the Model for Assessing the value of Artificial Intelligence in medical imaging (MAS-AI) in the Italian context through a case-study.We applied the MAS-AI, a model for assessing AI in healthcare, to fulfil a technology assessment of an AI model developed within our institution. The model, called New organization model for the surgical unit (BLOC-OP), uses AI to improve the schedule efficiency of the surgical unit. The analysis of BLOC-OP's features, as they were described in the project presentation, was conducted through the requirements for the assessment contained in the MAS-AI model.The methodological framework of MAS-AI was fully followed, allowing us to conduct a comprehensive assessment of the BLOC-OP model in all its aspects. We provided a detailed description of each domain within the framework, along with a summary table.The case study demonstrates the feasibility of applying MAS-AI to organizational AI models in a national context different from where the framework was originally developed. Rather than proposing a new model, we tested the adaptability of MAS-AI in evaluating a non-imaging AI system. This confirms its flexibility beyond its original scope and supports its potential as a generalizable tool for AI evaluation in healthcare.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"108"},"PeriodicalIF":5.7,"publicationDate":"2025-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12373541/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144957560","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}

{"title":"Symptom Recognition in Medical Conversations Via multi- Instance Learning and Prompt.","authors":"Hua Wang, Xue-Feng Bai, Xiu-Tao Cui, Gang Chen, Guo-Ming Fan, Guo-Lian Wei, Ye-Ping Zheng, Jing-Jing Wu, Sheng-Sheng Gao","doi":"10.1007/s10916-025-02240-w","DOIUrl":"https://doi.org/10.1007/s10916-025-02240-w","url":null,"abstract":"<p><p>With the widespread adoption of electronic health record (EHR) systems, there is a crucial need for automatic extraction of key symptom information from medical dialogue to support intelligent medical record generation. However, symptom recognition in such dialogues remains challenging because (a) symptom clues are scattered across multi-turn, unstructured conversations, (b) patient descriptions are often informal and deviate from standardized terminology, and (c) many symptom statements are ambiguous or negated, making them difficult for conventional models to interpret. To address these challenges, we propose a novel symptom identification approach that combines multi-instance learning (MIL) with prompt-guided attention for fine-grained symptom identification. In our framework, each conversation is treated as a bag of utterances. A MIL-based model aggregates information across utterances to improve recall and pinpoints which specific utterances mention each symptom, thus enabling sentence-level symptom recognition. Concurrently, a prompt-guided attention strategy leverages standardized symptom terminology as prior knowledge to guide the model in recognizing synonyms, implicit symptom mentions, and negations, thereby improving precision. We further employ R-Drop regularization to enhance robustness against noisy inputs. Experiments on public medical-dialogue datasets demonstrate that our method significantly outperforms existing techniques, achieving an 85.93% F1-score (with 85.09% precision and 86.83% recall) - about 8% points higher than a strong multi-label classification baseline. Notably, our model accurately identifies the specific utterances corresponding to each symptom mention (symptom-utterance pairs), highlighting its fine-grained extraction capability. Ablation studies confirm that the MIL component boosts recall, while the prompt-guided attention component reduces false positives. By precisely locating symptom information within conversations, our approach effectively tackles the issues of dispersed data and inconsistent expressions. This fine-grained symptom documentation capability represents a promising advancement for automated medical information extraction, more intelligent EHR systems, and diagnostic decision support.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"107"},"PeriodicalIF":5.7,"publicationDate":"2025-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144957508","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":"Automated Characterization of Sudden Cardiac Death Using Locality Preserving Projection and Fuzzy Entropy Based on Empirical Mode Decomposition from ECG Signals.","authors":"Manhong Shi, Yinuo Shi, Wenkang Zhou, Xue Qi","doi":"10.1007/s10916-025-02239-3","DOIUrl":"10.1007/s10916-025-02239-3","url":null,"abstract":"<p><p>The early prediction of sudden cardiac death (SCD) has garnered considerable global attention as a potentially life-saving intervention for at-risk individuals. While various strategies have been proposed, many are constrained by prediction time resolution (typically analyzing 1- to 2-min ECG segments) and early prediction time windows not exceeding 20 min. In this study, we propose a novel yet straightforward methodology that combines locality preserving projection (LPP) features and fuzzy entropy (FuEn) based on empirical mode decomposition (EMD) from individual ECG beats containing 1000 data points. Specifically, 15 features were extracted: 14 discriminative LPP features selected from the training data using the feature ranking method, along with one FuEn feature calculated from the first intrinsic mode function (IMF1) of the EMD. These selected features are applied to test data to differentiate between normal subjects and those at risk of SCD. A distinguishing aspect of our approach is that it analyzes each single ECG beat for SCD prediction, rather than relying on 1- or 2-min segments. Additionally, we incorporate group-based fivefold cross-validation to ensure a robust evaluation of prediction performance. Our method successfully predicts SCD 30 min in advance with an accuracy of 97.6%. In principle, the features extracted from this methodology can be integrated into portable medical sensors for real-time SCD risk assessment, suitable for use both in medical facilities and at home under the supervision of healthcare providers.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"105"},"PeriodicalIF":5.7,"publicationDate":"2025-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144859265","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":"Nurse Staffing Management in the Context of Emergency Departments and Seasonal Respiratory Diseases: An Artificial Intelligence and Discrete-Event Simulation Approach.","authors":"Miguel Ortiz-Barrios, Llanos Cuenca, Sebastián Arias-Fonseca, Sally McClean, Armando Pérez-Aguilar","doi":"10.1007/s10916-025-02242-8","DOIUrl":"10.1007/s10916-025-02242-8","url":null,"abstract":"<p><p>Emergency Departments (EDs) usually experience nursing shortages during Seasonal Respiratory Diseases (SRDs). As a result, patient waiting times for medical treatment increase with the consequent overcrowding, high intra-hospital infection rates, and no-shows. Therefore, the nurse staffing must be balanced with the projected volume of SRD-related ED admissions to EDs. In this article, we propose merging Artificial Intelligence (AI) and Discrete-Event Simulation (DES) to build remedies that diminish the waiting times for nursing care in mild and severe respiratory-affected patients. We first implemented Extreme Gradient Boosting (XGBoost) to calculate the probability of treatment within the ED wards. Afterwards, we plugged the XGBoost predictions into a simulation model to evaluate whether the current nurse staff was sufficient to ensure the timely treatment of the expected respiratory-affected patients. Ultimately, we pretested three improvement scenarios recommended by the hospital administrators to tackle the imbalance problem. A Spanish ED was involved in the project to validate the suggested approach. The specificity of the predictive AI-based model was 95.97% (CI 95% 93.07% - 97.90%), while the specificity was 82.0% (CI 95% 73.05% - 88.96%). On a different tack, the positive and negative predictive scores corresponded to 87.23% (CI 95% 78.76% - 93.22%) and 94.08% (95% CI 90.80% - 96.45%). Furthermore, the Area Under Receiver Operator Characteristic (AU-ROC) curve was 89.00% (CI 95% 84.46% - 94.78%). Ultimately, the median waiting time for respiratory support use was lessened between 0.88 and 7.51 h after using a new nurse staffing configuration.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"106"},"PeriodicalIF":5.7,"publicationDate":"2025-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144859279","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}