Journal of Medical Systems最新文献

{"title":"Enhancing Cricothyroidotomy Training for Novices Using Three-Dimensional-Printed Patient-Specific Models of a Patient with Obesity.","authors":"Jaeeun Song, Junhyeok Ock, Wook-Jong Kim, Yong-Seok Park, Namkug Kim, Sung-Hoon Kim","doi":"10.1007/s10916-025-02209-9","DOIUrl":"10.1007/s10916-025-02209-9","url":null,"abstract":"<p><p>This study aimed to enhance cricothyroidotomy training for novice practitioners using three-dimensional-printed patient-specific models based on computed tomography images of a patient with obesity, evaluate these models compared to conventional training phantoms, and suggest possible effective training methods. A prospective, randomised crossover study was conducted with 30 medical students with no prior cricothyroidotomy experience. Participants performed the procedure on a conventional and a patient-specific model. Performance was assessed using time, visual inspections, and a three-dimensional scanner to evaluate the accuracy of the cricothyroidotomy simulation. The correlation between total time and checklist times for procedural step skills was analysed. Furthermore, a post-study survey was conducted to evaluate participants' perceptions of the realism and utility of both simulators. Patient-specific simulators required a longer time (18.63 ± 6.96 s) to confirm tracheal position compared to conventional simulators (15.28 ± 6.96 s; p = 0.034). Conversely, conventional simulators required a longer time (44.86 ± 27.56 s) to intubate than patient-specific simulators (27.96 ± 9.73 s; p < 0.001). Patient-specific simulators exhibited a greater deviation from the intended puncture site (17.14 ± 8.03 mm) compared to conventional simulators (2.95 ± 1.25 mm; p < 0.001), despite high visual success rates for both models. Survey results showed significantly higher ratings for the patient-specific simulator in terms of fidelity, utility, and special features (p < 0.001). This study assessed both time and accuracy in evaluating and enhancing training and procedural outcomes, being the first to incorporate a three-dimensional scanner into assessing outcomes. The findings, along with positive participant feedback from the post-study survey, emphasise the need for specialised training programmes incorporating a three-dimensional-printed, patient-specific models that reflect challenging scenarios particularly involving patients with obesity.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"75"},"PeriodicalIF":3.5,"publicationDate":"2025-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144208673","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":"The Role of Artificial Intelligence Large Language Models in Personalized Rehabilitation Programs for Knee Osteoarthritis: An Observational Study.","authors":"Ömer Alperen Gürses, Anıl Özüdoğru, Figen Tuncay, Caner Kararti","doi":"10.1007/s10916-025-02207-x","DOIUrl":"10.1007/s10916-025-02207-x","url":null,"abstract":"<p><strong>Background: </strong>Large language models (LLMs) can contribute to treatment options and outcomes by assisting physiotherapists for conditions like osteoarthritis.</p><p><strong>Aims: </strong>The objective of this early-stage cross-sectional study is to assess the alignment of large language models with physiotherapists in designing physiotherapy and rehabilitation programs for knee osteoarthritis.</p><p><strong>Methods: </strong>Forty patients diagnosed with knee osteoarthritis were assessed using standardized clinical criteria. For each patient, individualized rehabilitation programs were created by three physiotherapists and by ChatGPT-4o and Gemini Advanced using structured prompts. The presence or absence of 50 clinically relevant rehabilitation parameters was recorded for each program. Chi-square tests were used to evaluate agreement rates between the LLMs and the physiotherapist-generated Consensus programs.</p><p><strong>Results: </strong>ChatGPT-4o achieved a 74% agreement rate with the physiotherapists' Consensus programs, while Gemini Advanced achieved 70%. Although both models showed high compatibility with general rehabilitation components, they demonstrated notable limitations in exercise specificity, including frequency, sets, and progression criteria. ChatGPT-4o performed as well as or better than Gemini in most phases, particularly in Phase 3, while Gemini showed lower consistency in balance and stabilization parameters.</p><p><strong>Conclusions: </strong>ChatGPT-4o and Gemini Advanced demonstrate promising potential in generating personalized rehabilitation programs for knee osteoarthritis. While their outputs generally align with expert recommendations, notable gaps remain in clinical reasoning and the provision of detailed exercise parameters. These findings underscore the importance of ongoing model refinement and the necessity of expert supervision for safe and effective clinical integration.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"73"},"PeriodicalIF":3.5,"publicationDate":"2025-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12134017/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144208728","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":"Performance of DeepSeek-R1 and ChatGPT-4o on the Chinese National Medical Licensing Examination: A Comparative Study.","authors":"Jin Wu, Zhiheng Wang, Yifan Qin","doi":"10.1007/s10916-025-02213-z","DOIUrl":"https://doi.org/10.1007/s10916-025-02213-z","url":null,"abstract":"<p><p>Large Language Models (LLMs) have a significant impact on medical education due to their advanced natural language processing capabilities. ChatGPT-4o (Chat Generative Pre-trained Transformer), a mainstream Western LLM, demonstrates powerful multimodal abilities. DeepSeek-R1, a newly released free and open-source LLM from China, demonstrates capabilities on par with ChatGPT-4o across various domains. This study aims to evaluate the performance of DeepSeek-R1 and ChatGPT-4o on the Chinese National Medical Licensing Examination (CNMLE) and explore the performance differences of LLMs from distinct linguistic environments in Chinese medical education. We evaluated both LLMs using 600 multiple-choice questions from the written part of 2024 CNMLE, covering four units. The questions were categorized into low- and high-difficulty groups according to difficulty. The primary outcome was the overall accuracy rate of each LLM. The secondary outcomes included accuracy within each of the four units and within the two difficulty-level groups. DeepSeek-R1 achieved a statistically significantly higher overall accuracy of 92.0% compared to ChatGPT-4o's 87.2% (P < 0.05). In the low-difficulty group, DeepSeek-R1 demonstrated an accuracy rate of 95.9%, which was significantly higher than ChatGPT-4o's 92.0% (P < 0.05). No statistically significant differences were observed between the models in any of the four units or in the high-difficulty group (P > 0.05). DeepSeek-R1 demonstrated a performance advantage on CNMLE.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"74"},"PeriodicalIF":3.5,"publicationDate":"2025-06-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144208687","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":"Comparative Analysis of Feature Extraction Methods and Machine Learning Models for Predicting Osteoporosis Prevalence.","authors":"Danni Zhang, Xingyu Yang, Fangying Wang, Cifang Qiu, Yanfu Chai, Danruo Fang","doi":"10.1007/s10916-025-02203-1","DOIUrl":"https://doi.org/10.1007/s10916-025-02203-1","url":null,"abstract":"<p><p>This study systematically examined the impact of three feature selection techniques (Boruta, Extreme gradient boosting (XGBoost), and Lasso) for optimizing four machine learning models (Random forest (RF), XGBoost, Logistic regression (LR), and Support vector machine (SVM)) in predicting bone density prevalence. Our findings revealed that varying data partitioning ratios (training and test sets: 0.6:0.4; 0.7:0.3; 0.8:0.2; 0.9:0.1) minimally impacted the prediction accuracy across all four models, a conclusion reinforced by 10-fold cross validation. Besides, principal component analysis (PCA) led to substantial accuracy degradation (0.6-0.8 range), suggesting incompatibility with this study's requirements due to the inherent complex decision boundaries in the original high-dimensional data. Comparative analysis demonstrated that the Boruta-XGBoost combination achieved superior performance (accuracy: 0.9083 ± 0.0146), significantly outperforming the Lasso-LR combination (0.7480 ± 0.0157) across all evaluation frameworks. Regarding model evaluation metrics, the RF model exhibited enhanced discriminative capacity with Area under the receiver operating characteristic (AUROC) values of 0.85, 0.81, and 0.80 under different feature selection approaches, surpassing the SVM model (0.78, 0.76, and 0.76). This advantage likely stems from RF's native capability to capture non-linear relationships and feature interactions.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"72"},"PeriodicalIF":3.5,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144173965","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":"Architecting a National Provider Directory for Healthcare.","authors":"Jacob T Kannarkat, Joseph T Kannarkat, William H Shrank","doi":"10.1007/s10916-025-02204-0","DOIUrl":"10.1007/s10916-025-02204-0","url":null,"abstract":"<p><p>The Department of Health and Human Services' vision of interoperability of health information systems promises improved efficiency and patient care quality. This endeavor hinges on solving the challenge of reliable user authentication within this interoperable system. Consequently, the Centers for Medicare and Medicaid Services (CMS) and Oklahoma are building a centralized provider directory pilot to inform development of a National Directory of Health (NDH). In this commentary, the authors identify barriers to implementing a national provider directory based on past efforts to build reliable provider directories and offer important considerations for this undertaking. Key areas to study include motivating directory users to self-validate their information, assisting healthcare stakeholders in complying with upcoming CMS regulations, and exploring a modular design for future expansion of directory functions. Given the likely multi-billion-dollar investment into a NDH and potentially significant returns in administrative efficiency and patient care quality, learning from past efforts is paramount for an effective and sustainable solution.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"71"},"PeriodicalIF":3.5,"publicationDate":"2025-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144142266","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":"Interpretable Machine Learning in Retinal Vein Occlusion: A Step Toward Precision Medicine.","authors":"Cheng Xue","doi":"10.1007/s10916-025-02205-z","DOIUrl":"https://doi.org/10.1007/s10916-025-02205-z","url":null,"abstract":"","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"70"},"PeriodicalIF":3.5,"publicationDate":"2025-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144127933","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":"Hand Motion Analysis of Different Segments of a Procedure: Is One Segment Enough?","authors":"Jeffrey Weinstein, Hamza Ali, Oussama Metrouh, Ammar Sarwar, John D Mitchell, Vincent Baribeau, Vanessa T Wong, Robina Matyal, Matthew R Palmer, Christopher MacLellan, Muneeb Ahmed","doi":"10.1007/s10916-025-02198-9","DOIUrl":"https://doi.org/10.1007/s10916-025-02198-9","url":null,"abstract":"<p><strong>Purpose: </strong>This study aims to examine if the hand motions of operators associated with certain parts of central venous access are more important than others in distinguishing between experts and non-experts.</p><p><strong>Materials and methods: </strong>Experts (n = 10) and Trainees (PGY2; n = 18) performed central venous access on a phantom 4 times each as their needle hand and ultrasound probe motions were tracked. Path length-time graphs were used to divide the procedure into three phases: (1) the access phase: visualizing the internal jugular vein on ultrasound and needle placement; (2) the wire phase: passing a wire through the needle; and (3) the confirmation phase: confirming the intravascular wire position and threading a dilator on the wire. Comparisons between trainees and experts were made for the complete trial, and each phase using Mann-Whitney U tests with Benjamini-Hochberg correction. Receiver Operating Characteristic analysis was performed to compare the performance of each phase in differentiating between experts and trainees.</p><p><strong>Results: </strong>Motion data from 10 experts and 18 trainees was analyzed. Experts and trainees differed significantly for all the motion metrics (p < 0.001). A comparison of the phases showed that the access phase (AUC = 0.96; R2 = 0.79) and the wire phase (AUC = 0.95; R2 = 0.59) were able to distinguish between experts and trainees with an accuracy comparable to the complete trial (AUC = 0.94; R2 = 0.69).</p><p><strong>Conclusions: </strong>The access phase of simulated central venous access can best differentiate between experts and trainees. This sample of hand motion performance may be able to simplify motion analysis of technical performance and obviate the need for recording hand motion for the entire procedure.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"69"},"PeriodicalIF":3.5,"publicationDate":"2025-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144127931","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":"From Admission to Discharge: Leveraging NLP for Upstream Primary Coding with SNOMED CT.","authors":"Elisa Asensio Blasco, Xavier Borrat Frigola, Xavier Pastor Duran, Artur Conesa González, Narcís Macià, David Sánchez Barcenilla, Ricardo Garrido Bejar, Santiago Frid","doi":"10.1007/s10916-025-02200-4","DOIUrl":"10.1007/s10916-025-02200-4","url":null,"abstract":"<p><p>This study aims to describe implementing a SNOMED CT-coded health problem (HP) list at Hospital Clínic de Barcelona. The project focuses on enhancing the accuracy and efficiency of clinical coding by automating the process from patient admission, while simultaneously enabling the reuse of coded data for research and management purposes. SNOMED CT was selected as the reference terminology for recording HPs. A subset of terms (our Health Problems Catalogue -HPC-) was created to meet local needs. An NLP tool was integrated into the clinical workstation to assist in primary coding HPs from natural language inputs. The system architecture included four servers (Coder, Reviewer, Manager, and Terminology Server) supporting real-time coding and review processes. Clinical and operational data from April to October 2024 were analyzed to evaluate the system's performance. Between April 9 and October 4, 2024, a total of 118,534 HPs were recorded. Of these, 74.2% were coded in real-time using the NLP tool, 23.3% were coded by documentation specialists, and 2.5% remained uncoded. The system significantly reduced coding delays and enriched the institutional data warehouse, facilitating real-time research and management activities. Implementing a SNOMED CT-coded HP list supported by NLP and terminology services improved coding accuracy and clinician efficiency. This system enhances clinical understanding, enables evidence-based recommendations, and supports data-driven decision-making in healthcare management and research. Clinical Trial Number Not applicable.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"68"},"PeriodicalIF":3.5,"publicationDate":"2025-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12095342/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144119984","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":"Forecasting Surgical Bed Utilization: Architectural Design of a Machine Learning Pipeline Incorporating Predicted Length of Stay and Surgical Volume.","authors":"Arjun Singh, Patrick E Farmer, Jeffrey L Tully, Ruth S Waterman, Rodney A Gabriel","doi":"10.1007/s10916-025-02201-3","DOIUrl":"10.1007/s10916-025-02201-3","url":null,"abstract":"<p><p>The objective of this study was to develop a machine learning model utilizing data from the electronic health record (EHR) to model length of stay and daily surgical volume, in order to subsequently predict daily surgical inpatient bed utilization. Machine learning is increasingly used to aid healthcare decision-making and resource allocation. Surgical inpatient bed utilization is a key metric of hospital efficiency and an ideal target for optimization. EHR data from all surgical cases over one year at a single institution was obtained. Data from the first 32 weeks of the year were used to train the model with the remaining data used to validate and test the models. Various machine learning approaches were explored to predict hospital length of stay and surgical volume. Seasonal Autoregressive Integrated Moving Average (SARIMA) was used to forecast daily surgical bed requirements. The root mean squared error (RMSE) was reported. For predicting bed utilization > 2 weeks in the future, our optimized models improved prediction from an RMSE of 43.1 to 24.4 beds. For predicting bed utilization in 2 weeks, our optimized models improved prediction from an RMSE of 42.6 to 24.8 beds. Finally, predicting bed utilization same day demonstrated an RMSE of 22.7 beds. We described the architecture of a machine learning approach to forecast surgical bed utilization. Forecasting use of surgical resources may decrease stress on a hospital system through more accurate predicting of the ebbs and flows of hospital needs.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"67"},"PeriodicalIF":3.5,"publicationDate":"2025-05-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12095433/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144111056","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":"Wearable Sensors and Artificial Intelligence for Sleep Apnea Detection: A Systematic Review.","authors":"Ainhoa Osa-Sanchez, Javier Ramos-Martinez-de-Soria, Amaia Mendez-Zorrilla, Ibon Oleagordia Ruiz, Begonya Garcia-Zapirain","doi":"10.1007/s10916-025-02199-8","DOIUrl":"10.1007/s10916-025-02199-8","url":null,"abstract":"<p><p>Sleep apnea, a prevalent disorder affecting millions of people worldwide, has attracted increasing attention in recent years due to its significant impact on public health and quality of life. The integration of wearable devices and artificial intelligence technologies has revolutionized the treatment and diagnosis of sleep apnea. Leveraging the portability and sensors of wearable devices, coupled with AI algorithms, has enabled real-time monitoring and accurate analysis of sleep patterns, facilitating early detection and personalized interventions for people suffering from sleep apnea. This article presents a systematic review of the current state of the art in identifying the latest artificial intelligence techniques, wearable devices, data types, and preprocessing methods employed in the diagnosis of sleep apnea. Four databases were used and the results before screening report 249 studies published between 2020 and 2024. After screening, 28 studies met the inclusion criteria. This review reveals a trend in recent years where methodologies involving patches, clocks and rings have been increasingly integrated with convolutional neural networks, producing promising results, particularly when combined with transfer learning techniques. We observed that the outcomes of various algorithms and their combinations also rely on the quantity and type of data utilized for training. The findings suggest that employing multiple combinations of different neural networks with convolutional layers contributes to the development of a more precise system for early diagnosis of sleep apnea.</p>","PeriodicalId":16338,"journal":{"name":"Journal of Medical Systems","volume":"49 1","pages":"66"},"PeriodicalIF":3.5,"publicationDate":"2025-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12089203/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144093627","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}