BMC Medical Informatics and Decision Making最新文献

{"title":"ARDSFlag: an NLP/machine learning algorithm to visualize and detect high-probability ARDS admissions independent of provider recognition and billing codes.","authors":"Amir Gandomi, Phil Wu, Daniel R Clement, Jinyan Xing, Rachel Aviv, Matthew Federbush, Zhiyong Yuan, Yajun Jing, Guangyao Wei, Negin Hajizadeh","doi":"10.1186/s12911-024-02573-5","DOIUrl":"10.1186/s12911-024-02573-5","url":null,"abstract":"<p><strong>Background: </strong>Despite the significance and prevalence of acute respiratory distress syndrome (ARDS), its detection remains highly variable and inconsistent. In this work, we aim to develop an algorithm (ARDSFlag) to automate the diagnosis of ARDS based on the Berlin definition. We also aim to develop a visualization tool that helps clinicians efficiently assess ARDS criteria.</p><p><strong>Methods: </strong>ARDSFlag applies machine learning (ML) and natural language processing (NLP) techniques to evaluate Berlin criteria by incorporating structured and unstructured data in an electronic health record (EHR) system. The study cohort includes 19,534 ICU admissions in the Medical Information Mart for Intensive Care III (MIMIC-III) database. The output is the ARDS diagnosis, onset time, and severity.</p><p><strong>Results: </strong>ARDSFlag includes separate text classifiers trained using large training sets to find evidence of bilateral infiltrates in radiology reports (accuracy of 91.9%±0.5%) and heart failure/fluid overload in radiology reports (accuracy 86.1%±0.5%) and echocardiogram notes (accuracy 98.4%±0.3%). A test set of 300 cases, which was blindly and independently labeled for ARDS by two groups of clinicians, shows that ARDSFlag generates an overall accuracy of 89.0% (specificity = 91.7%, recall = 80.3%, and precision = 75.0%) in detecting ARDS cases.</p><p><strong>Conclusion: </strong>To our best knowledge, this is the first study to focus on developing a method to automate the detection of ARDS. Some studies have developed and used other methods to answer other research questions. Expectedly, ARDSFlag generates a significantly higher performance in all accuracy measures compared to those methods.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11250933/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141626067","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":"Improve the efficiency and accuracy of ophthalmologists' clinical decision-making based on AI technology.","authors":"Yingxuan Guo, Changke Huang, Yaying Sheng, Wenjie Zhang, Xin Ye, Hengli Lian, Jiahao Xu, Yiqi Chen","doi":"10.1186/s12911-024-02587-z","DOIUrl":"10.1186/s12911-024-02587-z","url":null,"abstract":"<p><strong>Background: </strong>As global aging intensifies, the prevalence of ocular fundus diseases continues to rise. In China, the tense doctor-patient ratio poses numerous challenges for the early diagnosis and treatment of ocular fundus diseases. To reduce the high risk of missed or misdiagnosed cases, avoid irreversible visual impairment for patients, and ensure good visual prognosis for patients with ocular fundus diseases, it is particularly important to enhance the growth and diagnostic capabilities of junior doctors. This study aims to leverage the value of electronic medical record data to developing a diagnostic intelligent decision support platform. This platform aims to assist junior doctors in diagnosing ocular fundus diseases quickly and accurately, expedite their professional growth, and prevent delays in patient treatment. An empirical evaluation will assess the platform's effectiveness in enhancing doctors' diagnostic efficiency and accuracy.</p><p><strong>Methods: </strong>In this study, eight Chinese Named Entity Recognition (NER) models were compared, and the SoftLexicon-Glove-Word2vec model, achieving a high F1 score of 93.02%, was selected as the optimal recognition tool. This model was then used to extract key information from electronic medical records (EMRs) and generate feature variables based on diagnostic rule templates. Subsequently, an XGBoost algorithm was employed to construct an intelligent decision support platform for diagnosing ocular fundus diseases. The effectiveness of the platform in improving diagnostic efficiency and accuracy was evaluated through a controlled experiment comparing experienced and junior doctors.</p><p><strong>Results: </strong>The use of the diagnostic intelligent decision support platform resulted in significant improvements in both diagnostic efficiency and accuracy for both experienced and junior doctors (P < 0.05). Notably, the gap in diagnostic speed and precision between junior doctors and experienced doctors narrowed considerably when the platform was used. Although the platform also provided some benefits to experienced doctors, the improvement was less pronounced compared to junior doctors.</p><p><strong>Conclusion: </strong>The diagnostic intelligent decision support platform established in this study, based on the XGBoost algorithm and NER, effectively enhances the diagnostic efficiency and accuracy of junior doctors in ocular fundus diseases. This has significant implications for optimizing clinical diagnosis and treatment.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11234671/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141562636","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":"Impact of linkage level on inferences from big data analyses in health and medical research: an empirical study.","authors":"Bora Lee, Young-Kyun Lee, Sung Han Kim, HyunJin Oh, Sungho Won, Suk-Yong Jang, Ye Jin Jeon, Bit-Na Yoo, Jean-Kyung Bak","doi":"10.1186/s12911-024-02586-0","DOIUrl":"10.1186/s12911-024-02586-0","url":null,"abstract":"<p><strong>Background: </strong>Linkage errors that occur according to linkage levels can adversely affect the accuracy and reliability of analysis results. This study aimed to identify the differences in results according to personally identifiable information linkage level, sample size, and analysis methods through empirical analysis.</p><p><strong>Methods: </strong>The difference between the results of linkage in directly identifiable information (DII) and indirectly identifiable information (III) linkage levels was set as III linkage based on name, date of birth, and sex and DII linkage based on resident registration number. The datasets linked at each level were named as database_III (DB_III) and database_DII (DB_DII), respectively. Considering the analysis results of the DII-linked dataset as the gold standard, descriptive statistics, group comparison, incidence estimation, treatment effect, and moderation effect analysis results were assessed.</p><p><strong>Results: </strong>The linkage rates for DB_DII and DB_III were 71.1% and 99.7%, respectively. Regarding descriptive statistics and group comparison analysis, the difference in effect in most cases was \"none\" to \"very little.\" With respect to cervical cancer that had a relatively small sample size, analysis of DB_III resulted in an underestimation of the incidence in the control group and an overestimation of the incidence in the treatment group (hazard ratio [HR] = 2.62 [95% confidence interval (CI): 1.63-4.23] in DB_III vs. 1.80 [95% CI: 1.18-2.73] in DB_DII). Regarding prostate cancer, there was a conflicting tendency with the treatment effect being over or underestimated according to the surveillance, epidemiology, and end results summary staging (HR = 2.27 [95% CI: 1.91-2.70] in DB_III vs. 1.92 [95% CI: 1.70-2.17] in DB_DII for the localized stage; HR = 1.80 [95% CI: 1.37-2.36] in DB_III vs. 2.05 [95% CI: 1.67-2.52] in DB_DII for the regional stage).</p><p><strong>Conclusions: </strong>To prevent distortion of the analyses results in health and medical research, it is important to check that the patient population and sample size by each factor of interest (FOI) are sufficient when different data are linked using DB_DII. In cases involving a rare disease or with a small sample size for FOI, there is a high likelihood that a DII linkage is unavoidable.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11234607/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141562735","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":"Optimization of vision transformer-based detection of lung diseases from chest X-ray images.","authors":"Jinsol Ko, Soyeon Park, Hyun Goo Woo","doi":"10.1186/s12911-024-02591-3","DOIUrl":"10.1186/s12911-024-02591-3","url":null,"abstract":"<p><strong>Background: </strong>Recent advances in Vision Transformer (ViT)-based deep learning have significantly improved the accuracy of lung disease prediction from chest X-ray images. However, limited research exists on comparing the effectiveness of different optimizers for lung disease prediction within ViT models. This study aims to systematically evaluate and compare the performance of various optimization methods for ViT-based models in predicting lung diseases from chest X-ray images.</p><p><strong>Methods: </strong>This study utilized a chest X-ray image dataset comprising 19,003 images containing both normal cases and six lung diseases: COVID-19, Viral Pneumonia, Bacterial Pneumonia, Middle East Respiratory Syndrome (MERS), Severe Acute Respiratory Syndrome (SARS), and Tuberculosis. Each ViT model (ViT, FastViT, and CrossViT) was individually trained with each optimization method (Adam, AdamW, NAdam, RAdam, SGDW, and Momentum) to assess their performance in lung disease prediction.</p><p><strong>Results: </strong>When tested with ViT on the dataset with balanced-sample sized classes, RAdam demonstrated superior accuracy compared to other optimizers, achieving 95.87%. In the dataset with imbalanced sample size, FastViT with NAdam achieved the best performance with an accuracy of 97.63%.</p><p><strong>Conclusions: </strong>We provide comprehensive optimization strategies for developing ViT-based model architectures, which can enhance the performance of these models for lung disease prediction from chest X-ray images.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11232177/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141558120","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":"Measuring electronic health literacy in the context of diabetes care: psychometric evaluation of a Persian version of the condition-specific eHealth literacy scale for diabetes.","authors":"Maryam Peimani, Mozhgan Tanhapour, Fatemeh Bandarian, Ensieh Nasli-Esfahani, Afshin Ostovar","doi":"10.1186/s12911-024-02594-0","DOIUrl":"10.1186/s12911-024-02594-0","url":null,"abstract":"<p><strong>Background: </strong>The rise of the internet and social media has led to increased interest among diabetes patients in using technology for information gathering and disease management. However, adequate eHealth literacy is crucial for protecting patients from unreliable diabetes-related information online.</p><p><strong>Objective: </strong>To examine the psychometric characteristics and explore the preliminary validity of the Persian version of the Condition-specific eHealth Literacy Scale for Diabetes (Persian CeHLS-D) to assess eHealth literacy in the context of diabetes care.</p><p><strong>Methods: </strong>After adapting, translating, examining content validity, and pilot testing the questionnaire, it was administered to 300 patients with type 2 diabetes mellitus (T2DM). Construct validity was assessed through confirmatory factor analysis, convergent and known-groups validity. The internal consistency (Cronbach's alpha), composite reliability and maximum reliability, and test-retest correlation were assessed.</p><p><strong>Results: </strong>Factor analysis supported the hypothesized two-factor model with 10 items, and the standardized factor loadings ranged from 0.44 to 0.86 (P-values < 0.001). Cronbach's alpha and test-retest correlation were good for each factor. Convergent validity was confirmed by significant correlations of Persian CeHLS-D with diabetes health literacy, perceived usefulness and importance of using the internet for health information, internet anxiety, and perceived physical and mental health. Know-groups validity determined using groups with different internet-use frequencies, and different attitudes towards providing online healthcare services, were satisfied.</p><p><strong>Conclusion: </strong>This study demonstrated the Persian CeHLS-D as a reliable and valid measure of eHealth literacy among patients with T2DM in Iran. Its satisfactory psychometric properties support its use in research and clinical settings to assess eHealth literacy and inform interventions.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11225123/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141537592","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":"Effect of the data-informed platform for health intervention on the culture of data use for decision-making among district health office staff in North Shewa Zone, Ethiopia: a cluster-randomised controlled trial.","authors":"Girum Taye Zeleke, Bilal Iqbal Avan, Mehret Amsalu Dubale, Joanna Schellenberg","doi":"10.1186/s12911-024-02597-x","DOIUrl":"10.1186/s12911-024-02597-x","url":null,"abstract":"<p><strong>Background: </strong>Similar to other low and middle-income countries, Ethiopia faces limitations in using local health data for decision-making.We aimed to assess the effect of an intervention, namely the data-informed platform for health, on the culture of data-based decision making as perceived by district health office staff in Ethiopia's North Shewa Zone.</p><p><strong>Methods: </strong>By designating district health offices as 'clusters', a cluster-randomised controlled trial was implemented. Out of a total of 24 districts in the zone, 12 districts were allocated to intervention arm and the other 12 in the control group arms. In the intervention arm district health office teams were supported in four-monthly cycles of data-driven decision-making over 20 months. This support included: (a) defining problems using a health system framework; (b) reviewing data; (c) considering possible solutions; (d) value-based prioritizing; and (e) a consultative process to develop, commit to, and follow up on action plans. To measure the culture of data use for decision-making in both intervention and control arms, we interviewed 120 health management staff (5 per district office). Using a Likert scale based standard Performance of Routine Information System Management tool, the information is categorized into six domains:- evidence-based decision making, emphasis on data quality, use of information, problem solving, responsibility and motivation. After converting the Likert scale responses into percentiles, difference-in-difference methods were applied to estimate the net effect of the intervention. In intervention districts, analysis of variance was used to summarize variation by staff designation.</p><p><strong>Results: </strong>The overall decision-making culture in health management staff showed a net improvement of 13% points (95% C.I:9, 18) in intervention districts. The net effect of each of the six domains in turn was an 11% point increase (95% C.I:7, 15) on culture of evidence based decision making, a 16% point increase (95% C.I:8, 24) on emphasis on data quality, a 20% point increase (95% C.I:12, 28) on use of information, a 21% point increase (95% C.I:13, 29) on problem solving, and a 10% point increase (95% C.I:4, 16) on responsibility and motivation. In terms of variation by staff designation within intervention districts, statistically significant differences were observed only for problem solving and responsibility.</p><p><strong>Conclusion: </strong>The data-informed platform for health strategy resulted in a measurable improvement in data use and structured decision-making culture by using existing systems, namely the Performance Monitoring Team meetings. The intervention supported district health offices in identifying and solving problems through a structured process. After further research, DIPH intervention could also be applied to other health administration and facility levels.</p><p><strong>Trial registration: </strong>Cli","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11225382/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141537591","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 intervention studies measure the effectiveness of medication safety-related clinical decision support systems in primary and long-term care: a systematic review.","authors":"David Lampe, John Grosser, Dennis Grothe, Birthe Aufenberg, Daniel Gensorowsky, Julian Witte, Wolfgang Greiner","doi":"10.1186/s12911-024-02596-y","DOIUrl":"10.1186/s12911-024-02596-y","url":null,"abstract":"<p><strong>Background: </strong>Medication errors and associated adverse drug events (ADE) are a major cause of morbidity and mortality worldwide. In recent years, the prevention of medication errors has become a high priority in healthcare systems. In order to improve medication safety, computerized Clinical Decision Support Systems (CDSS) are increasingly being integrated into the medication process. Accordingly, a growing number of studies have investigated the medication safety-related effectiveness of CDSS. However, the outcome measures used are heterogeneous, leading to unclear evidence. The primary aim of this study is to summarize and categorize the outcomes used in interventional studies evaluating the effects of CDSS on medication safety in primary and long-term care.</p><p><strong>Methods: </strong>We systematically searched PubMed, Embase, CINAHL, and Cochrane Library for interventional studies evaluating the effects of CDSS targeting medication safety and patient-related outcomes. We extracted methodological characteristics, outcomes and empirical findings from the included studies. Outcomes were assigned to three main categories: process-related, harm-related, and cost-related. Risk of bias was assessed using the Evidence Project risk of bias tool.</p><p><strong>Results: </strong>Thirty-two studies met the inclusion criteria. Almost all studies (n = 31) used process-related outcomes, followed by harm-related outcomes (n = 11). Only three studies used cost-related outcomes. Most studies used outcomes from only one category and no study used outcomes from all three categories. The definition and operationalization of outcomes varied widely between the included studies, even within outcome categories. Overall, evidence on CDSS effectiveness was mixed. A significant intervention effect was demonstrated by nine of fifteen studies with process-related primary outcomes (60%) but only one out of five studies with harm-related primary outcomes (20%). The included studies faced a number of methodological problems that limit the comparability and generalizability of their results.</p><p><strong>Conclusions: </strong>Evidence on the effectiveness of CDSS is currently inconclusive due in part to inconsistent outcome definitions and methodological problems in the literature. Additional high-quality studies are therefore needed to provide a comprehensive account of CDSS effectiveness. These studies should follow established methodological guidelines and recommendations and use a comprehensive set of harm-, process- and cost-related outcomes with agreed-upon and consistent definitions.</p><p><strong>Prospero registration: </strong>CRD42023464746.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11225126/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141533668","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":"Real-time non-invasive hemoglobin prediction using deep learning-enabled smartphone imaging.","authors":"Yuwen Chen, Xiaoyan Hu, Yiziting Zhu, Xiang Liu, Bin Yi","doi":"10.1186/s12911-024-02585-1","DOIUrl":"10.1186/s12911-024-02585-1","url":null,"abstract":"<p><strong>Background: </strong>Accurate measurement of hemoglobin concentration is essential for various medical scenarios, including preoperative evaluations and determining blood loss. Traditional invasive methods are inconvenient and not suitable for rapid, point-of-care testing. Moreover, current models, due to their complex parameters, are not well-suited for mobile medical settings, which limits the ability to conduct frequent and rapid testing. This study aims to introduce a novel, compact, and efficient system that leverages deep learning and smartphone technology to accurately estimate hemoglobin levels, thereby facilitating rapid and accessible medical assessments.</p><p><strong>Methods: </strong>The study employed a smartphone application to capture images of the eye, which were subsequently analyzed by a deep neural network trained on data from invasive blood test data. Specifically, the EGE-Unet model was utilized for eyelid segmentation, while the DHA(C3AE) model was employed for hemoglobin level prediction. The performance of the EGE-Unet was evaluated using statistical metrics including mean intersection over union (MIOU), F1 Score, accuracy, specificity, and sensitivity. The DHA(C3AE) model's performance was assessed using mean absolute error (MAE), mean-square error (MSE), root mean square error (RMSE), and R^2.</p><p><strong>Results: </strong>The EGE-Unet model demonstrated robust performance in eyelid segmentation, achieving an MIOU of 0.78, an F1 Score of 0.87, an accuracy of 0.97, a specificity of 0.98, and a sensitivity of 0.86. The DHA(C3AE) model for hemoglobin level prediction yielded promising outcomes with an MAE of 1.34, an MSE of 2.85, an RMSE of 1.69, and an R^2 of 0.34. The overall size of the model is modest at 1.08 M, with a computational complexity of 0.12 FLOPs (G).</p><p><strong>Conclusions: </strong>This system presents a groundbreaking approach that eliminates the need for supplementary devices, providing a cost-effective, swift, and accurate method for healthcare professionals to enhance treatment planning and improve patient care in perioperative environments. The proposed system has the potential to enable frequent and rapid testing of hemoglobin levels, which can be particularly beneficial in mobile medical settings.</p><p><strong>Trial registration: </strong>The clinical trial was registered on the Chinese Clinical Trial Registry (No. ChiCTR2100044138) on 20/02/2021.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11218390/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141475960","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":"Medical-informed machine learning: integrating prior knowledge into medical decision systems.","authors":"Christel Sirocchi, Alessandro Bogliolo, Sara Montagna","doi":"10.1186/s12911-024-02582-4","DOIUrl":"10.1186/s12911-024-02582-4","url":null,"abstract":"<p><strong>Background: </strong>Clinical medicine offers a promising arena for applying Machine Learning (ML) models. However, despite numerous studies employing ML in medical data analysis, only a fraction have impacted clinical care. This article underscores the importance of utilising ML in medical data analysis, recognising that ML alone may not adequately capture the full complexity of clinical data, thereby advocating for the integration of medical domain knowledge in ML.</p><p><strong>Methods: </strong>The study conducts a comprehensive review of prior efforts in integrating medical knowledge into ML and maps these integration strategies onto the phases of the ML pipeline, encompassing data pre-processing, feature engineering, model training, and output evaluation. The study further explores the significance and impact of such integration through a case study on diabetes prediction. Here, clinical knowledge, encompassing rules, causal networks, intervals, and formulas, is integrated at each stage of the ML pipeline, resulting in a spectrum of integrated models.</p><p><strong>Results: </strong>The findings highlight the benefits of integration in terms of accuracy, interpretability, data efficiency, and adherence to clinical guidelines. In several cases, integrated models outperformed purely data-driven approaches, underscoring the potential for domain knowledge to enhance ML models through improved generalisation. In other cases, the integration was instrumental in enhancing model interpretability and ensuring conformity with established clinical guidelines. Notably, knowledge integration also proved effective in maintaining performance under limited data scenarios.</p><p><strong>Conclusions: </strong>By illustrating various integration strategies through a clinical case study, this work provides guidance to inspire and facilitate future integration efforts. Furthermore, the study identifies the need to refine domain knowledge representation and fine-tune its contribution to the ML model as the two main challenges to integration and aims to stimulate further research in this direction.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11212227/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141466351","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":"Design, implementation and usability analysis of patient empowerment in ADLIFE project via patient reported outcome measures and shared decision making.","authors":"Gokce B Laleci Erturkmen, Natassia Kamilla Juul, Irati Erreguerena Redondo, Ana Ortega Gil, Dolores Verdoy Berastegui, Esteban de Manuel, Mustafa Yuksel, Bunyamin Sarigul, Gokhan Yilmaz, Sarah N L I M Choi Keung, Theodoros N Arvanitis, Thea Damkjaer Syse, Janika Bloemeke-Cammin, Rachelle Kaye, Anne Dichmann Sorknæs","doi":"10.1186/s12911-024-02588-y","DOIUrl":"https://doi.org/10.1186/s12911-024-02588-y","url":null,"abstract":"<p><strong>Introduction: </strong>This paper outlines the design, implementation, and usability study results of the patient empowerment process for chronic disease management, using Patient Reported Outcome Measurements and Shared Decision-Making Processes.</p><p><strong>Background: </strong>The ADLIFE project aims to develop innovative, digital health solutions to support personalized, integrated care for patients with severe long-term conditions such as Chronic Obstructive Pulmonary Disease, and/or Chronic Heart Failure. Successful long-term management of patients with chronic conditions requires active patient self-management and a proactive involvement of patients in their healthcare and treatment. This calls for a patient-provider partnership within an integrated system of collaborative care, supporting self-management, shared-decision making, collection of patient reported outcome measures, education, and follow-up.</p><p><strong>Methods: </strong>ADLIFE follows an outcome-based and patient-centered approach where PROMs represent an especially valuable tool to evaluate the outcomes of the care delivered. We have selected 11 standardized PROMs for evaluating the most recent patients' clinical context, enabling the decision-making process, and personalized care planning. The ADLIFE project implements the \"SHARE approach' for enabling shared decision-making via two digital platforms for healthcare professionals and patients. We have successfully integrated PROMs and shared decision-making processes into our digital toolbox, based on an international interoperability standard, namely HL7 FHIR. A usability study was conducted with 3 clinical sites with 20 users in total to gather feedback and to subsequently prioritize updates to the ADLIFE toolbox.</p><p><strong>Results: </strong>User satisfaction is measured in the QUIS7 questionnaire on a 9-point scale in the following aspects: overall reaction, screen, terminology and tool feedback, learning, multimedia, training material and system capabilities. With all the average scores above 6 in all categories, most respondents have a positive reaction to the ADLIFE PEP platform and find it easy to use. We have identified shortcomings and have prioritized updates to the platform before clinical pilot studies are initiated.</p><p><strong>Conclusions: </strong>Having finalized design, implementation, and pre-deployment usability studies, and updated the tool based on further feedback, our patient empowerment mechanisms enabled via PROMs and shared decision-making processes are ready to be piloted in clinal settings. Clinical studies will be conducted based at six healthcare settings across Spain, UK, Germany, Denmark, and Israel.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":null,"pages":null},"PeriodicalIF":3.3,"publicationDate":"2024-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11212241/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141466326","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}