BMC Medical Informatics and Decision Making最新文献

{"title":"A hybrid approach for binary and multi-class classification of voice disorders using a pre-trained model and ensemble classifiers.","authors":"Mehtab Ur Rahman, Cem Direkoglu","doi":"10.1186/s12911-025-02978-w","DOIUrl":"https://doi.org/10.1186/s12911-025-02978-w","url":null,"abstract":"<p><p>Recent advances in artificial intelligence-based audio and speech processing have increasingly focused on the binary and multi-class classification of voice disorders. Despite progress, achieving high accuracy in multi-class classification remains challenging. This paper proposes a novel hybrid approach using a two-stage framework to enhance voice disorders classification performance, and achieve state-of-the-art accuracies in multi-class classification. Our hybrid approach, combines deep learning features with various powerful classifiers. In the first stage, high-level feature embeddings are extracted from voice data spectrograms using a pre-trained VGGish model. In the second stage, these embeddings are used as input to four different classifiers: Support Vector Machine (SVM), Logistic Regression (LR), Multi-Layer Perceptron (MLP), and an Ensemble Classifier (EC). Experiments are conducted on a subset of the Saarbruecken Voice Database (SVD) for male, female, and combined speakers. For binary classification, VGGish-SVM achieved the highest accuracy for male speakers (82.45% for healthy vs. disordered; 75.45% for hyperfunctional dysphonia vs. vocal fold paresis), while VGGish-EC performed best for female speakers (71.54% for healthy vs. disordered; 68.42% for hyperfunctional dysphonia vs. vocal fold paresis). In multi-class classification, VGGish-SVM outperformed other models, achieving mean accuracies of 77.81% for male speakers, 63.11% for female speakers, and 70.53% for combined genders. We conducted a comparative analysis against related works, including the Mel frequency cepstral coefficient (MFCC), MFCC-glottal features, and features extracted using the wav2vec and HuBERT models with SVM classifier. Results demonstrate that our hybrid approach consistently outperforms these models, especially in multi-class classification tasks. The results show the feasibility of a hybrid framework for voice disorder classification, offering a foundation for refining automated tools that could support clinical assessments with further validation.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"177"},"PeriodicalIF":3.3,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12044829/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143972440","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":"Predicting preeclampsia in early pregnancy using clinical and laboratory data via machine learning model.","authors":"Songchang Chen, Jia Li, Xiao Zhang, Wenqiu Xu, Zhixu Qiu, Siyao Yan, Wenrui Zhao, Zhiguang Zhao, Peirun Tian, Qiang Zhao, Qun Zhang, Weiping Chen, Huahua Li, Xiaohong Ruan, Gefei Xiao, Sufen Zhang, Liqing Hu, Jie Qin, Wuyan Huang, Zhongzhe Li, Shunyao Wang, Rui Zhang, Shang Huang, Xin Wang, Yao Yao, Jian Ran, Danling Cheng, Qi Luo, Teng Pan, Ruyun Gao, Jing Zheng, Yuxuan Wang, Cong Liu, Xianling Cao, Xuanyou Zhou, Naixin Xu, Lanlan Zhang, Xu Han, Haolin Wang, Suihua Feng, Shuyuan Li, Jianguo Zhang, Lijian Zhao, Fengxiang Wei","doi":"10.1186/s12911-025-02999-5","DOIUrl":"https://doi.org/10.1186/s12911-025-02999-5","url":null,"abstract":"<p><strong>Background: </strong>This study was performed to characterize the relationship of various laboratory test indicators with clinical information and Preeclampsia (PE) development. Then, prediction models for early-onset preeclampsia (EOPE), late-onset preeclampsia (LOPE), and preterm preeclampsia (Preterm PE) were developed using maternal characteristics and laboratory data.</p><p><strong>Methods: </strong>Between January 2019 and December 2021, we retrospectively recruited 144 EOPE, 363 LOPE, 231 Preterm PE, and 1458 healthy participants from six hospitals. We utilized all available clinical and laboratory data obtained during routine prenatal visits in early pregnancy. The models for EOPE, LOPE, and Preterm PE were created using ensemble machine learning models with patient clinical and laboratory data.</p><p><strong>Results: </strong>By comparing laboratory variables between PE patients and healthy controls, we identified 7, 18, 8, 15, 7,29 laboratory markers for EOPE, LOPE, and Preterm PE, severe PE, superimposed PE, first-time PE respectively. The ensemble EOPE and LOPE models incorporating clinical and laboratory predictors outperformed the clinical factor models respectively. The ensemble EOPE model demonstrated good sensitivity (72.22%,95% confidence interval [CI]: 57.59%-86.85%) and specificity (85.25%,95% CI: 80.54%-89.97%) in distinguishing EOPE from controls in early pregnancy. Similarly, the ensemble LOPE model showed good accuracy in differentiating LOPE from healthy participants (sensitivity: 69.57%, 95% CI: 56.27%-82.86%; specificity: 85.25%, 95% CI: 80.54%-89.97%). The prediction scores demonstrated notable positive correlations with blood pressure at admission, while they showed inverse correlations with 24-hour urine protein levels and fetal growth restriction among PE patients. In conclusion, our study identified key laboratory indicators for forecasting PE. The developed models exhibited good predictive capability for assessing preeclampsia risk and severity based on clinical and laboratory data.</p><p><strong>Clinical trial number: </strong>Not applicable.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"178"},"PeriodicalIF":3.3,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12044989/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143954825","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":"Auto-expansion software prompting reduces abbreviation use in electronic hospital discharge letters: an observational pre- and post-intervention study.","authors":"Shamus Toomath, Emily J Hibbert","doi":"10.1186/s12911-025-03005-8","DOIUrl":"https://doi.org/10.1186/s12911-025-03005-8","url":null,"abstract":"<p><strong>Background: </strong>Abbreviation use remains a significant cause of miscommunication among healthcare practitioners worldwide, creating uncertainty in interpretation and leading to poorer patient outcomes. This study aimed to assess the effectiveness of implementing auto-expansion prompts to reduce abbreviation use in electronic discharge letters (eDLs).</p><p><strong>Methods: </strong>Observational pre- and post-intervention study conducted in 2019 at a tertiary referral hospital in Western Sydney.</p><p><strong>Participants: </strong>Junior medical officers (JMOs) in postgraduate years 1 and 2.</p><p><strong>Intervention: </strong>The intervention consisted of an email invitation to JMOs, outlining the risks of abbreviation use in eDLs, and providing instructions on how to use auto-expand prompts for 11 commonly used abbreviations in Cerner Powerchart.</p><p><strong>Primary outcome measure: </strong>Reduction in the frequency of use of 11 commonly used abbreviations selected for auto-expansion, measured by a 200 eDL audit pre- and post-intervention.</p><p><strong>Secondary outcome measures: </strong>Reduction in the total number of abbreviations used and the mean number of abbreviations per eDL in the post-intervention audit compared to pre-intervention.</p><p><strong>Results: </strong>The baseline audit identified 1668 abbreviation uses in 200 eDLs, consisting of 350 different abbreviations. In the post-intervention audit, use of the 11 auto-expand abbreviations decreased by 43.6%, with decreased frequency of use for 9 of the 11 abbreviations. Post-intervention there was a 34.4% reduction in the total number of abbreviations used, with 1093 abbreviations identified in 200 eDLs.</p><p><strong>Conclusions: </strong>Advising JMOs to implement auto-expansion prompts for specific abbreviations, in combination with education on the risks of abbreviation use, is a cheap and effective solution to reducing abbreviation use in eDLs. This approach could significantly improve clarity of communication between hospital doctors and community healthcare professionals during patient care transition, potentially reducing medical errors.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"180"},"PeriodicalIF":3.3,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12045010/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143980917","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":"Virtual body image exercises for people with obesity - results on eating behavior and body perception of the ViTraS pilot study.","authors":"Kathrin Gemesi, Nina Döllinger, Natascha-Alexandra Weinberger, Erik Wolf, David Mal, Sebastian Keppler, Stephan Wenninger, Emily Bader, Carolin Wienrich, Claudia Luck-Sikorski, Marc Erich Latoschik, Johann Habakuk Israel, Mario Botsch, Christina Holzapfel","doi":"10.1186/s12911-025-02993-x","DOIUrl":"https://doi.org/10.1186/s12911-025-02993-x","url":null,"abstract":"<p><strong>Background: </strong>A negative body image can have an impact on developing and maintaining obesity. Using virtual reality (VR) to conduct cognitive behavioral therapy (CBT) is an innovative approach to treat people with obesity. This multicenter non-randomized pilot study examined the feasibility and the effect on eating behavior and body perception of a newly developed VR system to conduct body image exercises.</p><p><strong>Methods: </strong>Participants with a body mass index (BMI) ≥ 30.0 kg/m² without severe mental diseases attended three study visits in an interval of one to four weeks to receive virtual (VR intervention) or traditional (non-VR intervention) body image exercises. Data on anthropometrics, eating behavior (Dutch Eating Behavior Questionnaire, DEBQ), body perception (Body Shape Questionnaire, BSQ; Multidimensional Assessment of Interoceptive Awareness, MAIA), and satisfaction (standardized interview and questionnaire) were collected.</p><p><strong>Results: </strong>In total, 66 participants (VR intervention: 31, non-VR intervention: 35) were included. The majority was female (52/66, 78.8 %), the mean age was 45.0 ± 12.8 years, and the mean BMI was 36.8 ± 4.3 kg/m². Both intervention groups showed non-significant body weight reduction (VR intervention: 1.7 ± 3.3 %, non-VR intervention: 0.9 ± 3.0 %) and showed no statistically significant difference between the groups (p = 0.35). Scores of DEBQ, BSQ, and MAIA showed over time no statistically significant changes neither between the two groups nor within the groups (all p ≥ 0.05). The overall satisfaction of the VR group with the two virtual body image exercises was high (4.1 ± 0.8 on a 5-point Likert scale).</p><p><strong>Conclusions: </strong>The intervention with the developed VR system was feasible and the virtual and traditional body image exercises resulted in statistically non-significant weight loss. It seems that single focus on body image is not successful in improving eating behavior and body perception in people with obesity. Long-term human intervention studies with larger sample sizes are necessary to examine the efficacy of integrating this kind of VR system into standard obesity therapy.</p><p><strong>Trial registration: </strong>This study was registered in the German Clinical Trials Register (Registration number: DRKS00027906, Date of registration: 8^th February 2022).</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"176"},"PeriodicalIF":3.3,"publicationDate":"2025-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12023441/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143977246","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":"SEM model analysis of diabetic patients' acceptance of artificial intelligence for diabetic retinopathy.","authors":"Luchang Jin, Yanmin Tao, Ya Liu, Gang Liu, Lin Lin, Zixi Chen, Sihan Peng","doi":"10.1186/s12911-025-03008-5","DOIUrl":"https://doi.org/10.1186/s12911-025-03008-5","url":null,"abstract":"<p><strong>Aims: </strong>This study aimed to investigate diabetic patients' acceptance of artificial intelligence (AI) devices for diabetic retinopathy screening and the related influencing factors.</p><p><strong>Methods: </strong>An integrated model was proposed, and structural equation modeling was used to evaluate items and construct reliability and validity via confirmatory factor analysis. The model's path effects, significance, goodness of fit, and mediation and moderation effects were analyzed.</p><p><strong>Results: </strong>Intention to Use (IU) is significantly affected by Subjective Norms (SN), Resistance Bias (RB), and Uniqueness Neglect (UN). Perceived Usefulness (PU) and Perceived Ease of Use (PEOU) were significant mediators between IU and other variables. The moderating effect of trust (TR) is non-significant on the path of PU to IU.</p><p><strong>Conclusions: </strong>The significant positive impact of SN may be caused by China's collectivist and authoritarian cultures. Both PU and PEOU had a significant mediation effect, which suggests that impressions influence acceptance. Although the moderating effect of TR was not significant, the unstandardized factor loading remained positive in this study. We presume that this may be due to an insufficient sample size, and the public was unfamiliar with AI medical devices.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"175"},"PeriodicalIF":3.3,"publicationDate":"2025-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12023383/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143963535","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":"Simulating the overload of medical processes due to system failures during a cyberattack.","authors":"Markus Willing, Simon Ebbers, Christian Dresen, Marc Czolbe, Christoph Saatjohann, Sebastian Schinzel","doi":"10.1186/s12911-025-02988-8","DOIUrl":"https://doi.org/10.1186/s12911-025-02988-8","url":null,"abstract":"<p><p>Today's medical IT is more and more connected and network or IT system outages may impact the quality of patient treatment. IT outages from cyberattacks are particularly worrisome if attackers focus on those medical IT devices that are critical for medical processes. However, medical processes are primarily documented for the hospital employees and not for analyzing the criticality of any given human or medical IT resource. This paper presents a generic model for realistic, patient-focused simulation of medical processes. The model allows the simulation of cyber incidents, focusing on device outages or overload situations like mass casualty incidents. Furthermore, we present a proof-of-concept tool that implements the described model, enabling end-users to simulate their processes. The tool offers the ability to run with low detailed data for overview purposes and highly detailed data for fine-grained simulation results. We perform different scenario simulations for a sample hospital, including the acute phase of a ransomware attack, negative performance impacts due to the implementation of cybersecurity measures, and emergency plans for mass casualty incidents. In each scenario, the respective simulation resulted in a quantitative statement of how these scenarios affect overall process performance and show possible key factors supporting decision-making. We use real-world data from a German trauma room to optimize and evaluate the process simulation.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"174"},"PeriodicalIF":3.3,"publicationDate":"2025-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12016121/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143978636","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":"ABiMed: An intelligent and visual clinical decision support system for medication reviews and polypharmacy management.","authors":"Mouazer Abdelmalek, Léguillon Romain, Boudegzdame Nada, Levrard Thibaud, Le Bars Yoann, Simon Christian, Séroussi Brigitte, Grosjean Julien, Lelong Romain, Letord Catherine, Darmoni Stéfan, Schuers Matthieu, Belmin Joël, Sedki Karima, Dubois Sophie, Falcoff Hector, Tsopra Rosy, Lamy Jean-Baptiste","doi":"10.1186/s12911-025-03002-x","DOIUrl":"https://doi.org/10.1186/s12911-025-03002-x","url":null,"abstract":"<p><strong>Background: </strong>Polypharmacy can be both a public health and an economic issue. Medication reviews are structured interviews of the patient by the pharmacist, aiming at optimizing the drug treatment and deprescribing potentially inappropriate medications. However, they remain difficult to perform and time-consuming. Several clinical decision support systems were developed for helping clinicians to reduce inappropriate polypharmacy. However, most were limited to the implementation of clinical practice guidelines. In this work, our objective is to design an innovative clinical decision support system for medication reviews and polypharmacy management, named ABiMed.</p><p><strong>Methods: </strong>ABiMed associates several approaches: guidelines implementation, but also the automatic extraction of patient data from the GP's electronic health record and its transfer to the pharmacist, and the visual presentation of contextualized drug knowledge using visual analytics. We performed an ergonomic assessment and qualitative evaluations involving pharmacists and GPs during focus groups and workshops.</p><p><strong>Results: </strong>We describe the proposed architecture, which allows a collaborative multi-user usage. We present the various screens of ABiMed for entering or verifying patient data, for accessing drug knowledge (posology, adverse effects, interactions), for viewing STOPP/START rules and for suggesting modification to the treatment. Qualitative evaluations showed that health professionals were highly interested in our approach, associating the automatic guidelines execution with the visual presentation of drug knowledge.</p><p><strong>Conclusions: </strong>The association of guidelines implementation with visual presentation of knowledge is a promising approach for managing polypharmacy. Future works will focus on the improvement and the evaluation of ABiMed.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"173"},"PeriodicalIF":3.3,"publicationDate":"2025-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12016315/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143965789","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":"Explainable machine learning algorithm to predict cardiovascular event in patients undergoing peritoneal dialysis.","authors":"Qiqi Yan, Guiling Liu, Ruifeng Wang, Dandan Li, Xiaoli Chen, Jingjing Cong, Deguang Wang","doi":"10.1186/s12911-025-03003-w","DOIUrl":"https://doi.org/10.1186/s12911-025-03003-w","url":null,"abstract":"<p><strong>Objective: </strong>To compare the performance of predictive models for cardiovascular event (CVE) in patients undergoing peritoneal dialysis (PD) based on machine learning algorithm and Cox proportional hazard regression.</p><p><strong>Methods: </strong>This study included patients underwent PD catheterization in our center from January 1, 2010, to July 31, 2022. The patients were randomly divided into training and validation sets in a 7:3 ratio. Cox regression, extreme gradient boosting (XGBoost), and random survival forest (RSF) models were developed using the training set and validated using the validation set. The time-dependent area under the curve (AUC) and concordance index (C-index) were used to evaluate the discriminative ability of predictive models.</p><p><strong>Results: </strong>A total of 318 patients were enrolled in this study. 110 (34.6%) patients developed CVE during the median follow-up of 31(16,56) months. The RSF model had better predictive performance, with a C-index of 0.725 and 1-, 3-, and 5-year time-dependent AUC of 0.812, 0.836, and 0.706 in the validation set, respectively. The top 5 important variables identified were platelet count, age, 4 hD/Pcr, left atrium diameter, and left ventricular diameter. Patients were classified into high-risk and low-risk groups based on the cut-off risk score calculated using the maximally selected rank statistics in the validation set. The log-rank test showed a significant difference in cumulative CVE-free survival probability between the two groups.</p><p><strong>Conclusion: </strong>The RSF model may be a useful method for evaluating CVE risk in PD patients.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"172"},"PeriodicalIF":3.3,"publicationDate":"2025-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12016290/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143974685","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":"Hierarchical embedding attention for overall survival prediction in lung cancer from unstructured EHRs.","authors":"Domenico Paolo, Carlo Greco, Alessio Cortellini, Sara Ramella, Paolo Soda, Alessandro Bria, Rosa Sicilia","doi":"10.1186/s12911-025-02998-6","DOIUrl":"https://doi.org/10.1186/s12911-025-02998-6","url":null,"abstract":"<p><p>The automated processing of Electronic Health Records (EHRs) poses a significant challenge due to their unstructured nature, rich in valuable, yet disorganized information. Natural Language Processing (NLP), particularly Named Entity Recognition (NER), has been instrumental in extracting structured information from EHR data. However, existing literature primarly focuses on extracting handcrafted clinical features through NLP and NER methods without delving into their learned representations. In this work, we explore the untapped potential of these representations by considering their contextual richness and entity-specific information. Our proposed methodology extracts representations generated by a transformer-based NER model on EHRs data, combines them using a hierarchical attention mechanism, and employs the obtained enriched representation as input for a clinical prediction model. Specifically, this study addresses Overall Survival (OS) in Non-Small Cell Lung Cancer (NSCLC) using unstructured EHRs data collected from an Italian clinical centre encompassing 838 records from 231 lung cancer patients. Whilst our study is applied on EHRs written in Italian, it serves as use case to prove the effectiveness of extracting and employing high level textual representations that capture relevant information as named entities. Our methodology is interpretable because the hierarchical attention mechanism highlights the information in EHRs that the model considers the most crucial during the decision-making process. We validated this interpretability by measuring the agreement of domain experts on the importance assigned by the hierarchical attention mechanism to EHRs information through a questionnaire. Results demonstrate the effectiveness of our method, showcasing statistically significant improvements over traditional manually extracted clinical features.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"169"},"PeriodicalIF":3.3,"publicationDate":"2025-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12007135/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143974160","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":"A machine learning-based severity stratification tool for high altitude pulmonary edema.","authors":"Luobu Gesang, Yangzong Suona, Zhuoga Danzeng, Bai Ci, Quzhen Gesang, WangJiu Cidan, Qiangba Dingzeng, Zhuoga Baima, Quzhen Zhaxi","doi":"10.1186/s12911-025-02992-y","DOIUrl":"https://doi.org/10.1186/s12911-025-02992-y","url":null,"abstract":"<p><p>This study aimed to identify key predictors for the severity of High Altitude Pulmonary Edema (HAPE) to assist clinicians in promptly recognizing severely affected patients in the emergency department, thereby reducing associated mortality rates. Multinomail logistic regression, random forest, and decision tree methods were utilized to determine important predictor variables and evaluate model performance. A total of 508 patients diagnosed with HAPE were included in the study, with 53 variables analyzed. Lung rales, sputum sputuming, heart rate, and oxygen saturation were identified as the most relevant predictors for the LASSO model. Subsequently, Multinomail logistic regression, decision tree, and random forest models were trained and evaluated using these factors on a test set. The random forest model showed the highest performance, with an accuracy of 77.94%, precision of 70.27%, recall of 68.22%, and F1 score of 68.96%, outperforming the other models. Further analysis revealed significant differences in predictive capabilities among the models for HAPE patients at varying severity levels. The random forest model demonstrated high predictive accuracy across all severity levels of HAPE, particularly excelling in identifying severely ill patients with an impressive AUC of 0.86. The study assessed the reliability and effectiveness of the HAPE severity scoring model by validating Multinomail logistic regression and random forest models. This study introduces a valuable screening tool for categorizing the severity of HAPE, aiding healthcare providers in recognizing individuals with severe HAPE, enabling prompt treatment and the formulation of suitable therapeutic approaches.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"171"},"PeriodicalIF":3.3,"publicationDate":"2025-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12008950/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143977243","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}