BMC Medical Informatics and Decision Making最新文献

{"title":"Anesthesia depth prediction from drug infusion history using hybrid AI.","authors":"Liang Wang, Yiqi Weng, Wenli Yu","doi":"10.1186/s12911-025-02986-w","DOIUrl":"https://doi.org/10.1186/s12911-025-02986-w","url":null,"abstract":"<p><strong>Background: </strong>Accurately predicting the depth of anesthesia is essential for ensuring patient safety and optimizing surgical outcomes. Traditional regression-based approaches often struggle to model the complex and dynamic nature of patient responses to anesthetic agents. Machine learning techniques offer a promising alternative by capturing intricate relationships within physiological data. This study proposes a hybrid model integrating Long Short-Term Memory (LSTM) networks, Transformer architectures, and Kolmogorov-Arnold Networks (KAN) to improve the predictive accuracy of anesthesia depth.</p><p><strong>Methods: </strong>The proposed model combines multiple deep learning techniques to address different aspects of anesthesia prediction. The LSTM component captures the sequential nature of drug administration and physiological responses. The Transformer architecture utilizes attention mechanisms to enhance contextual understanding of patient data. The KAN models nonlinear relationships between drug infusion histories and anesthesia depth. The model was trained and evaluated on patient data from a publicly available anesthesia monitoring database. Performance was assessed using Mean Squared Error (MSE) and compared against other models.</p><p><strong>Results: </strong>The hybrid model demonstrated superior predictive performance compared to conventional regression approaches. Tested on the VitalDB database, the proposed framework achieved a MSE of 0.0062, which is lower than other methods. The inclusion of attention mechanisms and nonlinear modeling contributed to improved accuracy and robustness. The results indicate that the combined approach effectively captures the temporal and nonlinear characteristics of anesthesia depth, offering a more reliable predictive tool for clinical use.</p><p><strong>Conclusions: </strong>This study presents a novel deep learning framework for anesthesia depth prediction, integrating sequential, attention-based, and nonlinear modeling techniques. The results suggest that this hybrid approach enhances prediction reliability and provides anesthesiologists with a more comprehensive analysis of factors influencing anesthesia depth. Future research will focus on refining model robustness, exploring real-time applications, and addressing potential biases in predictive analytics to further improve clinical decision-making.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"158"},"PeriodicalIF":3.3,"publicationDate":"2025-04-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143810505","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":"Detecting cognitive impairment in cerebrovascular disease using gait, dual tasks, and machine learning.","authors":"Vânia Guimarães, Inês Sousa, Miguel Velhote Correia","doi":"10.1186/s12911-025-02979-9","DOIUrl":"10.1186/s12911-025-02979-9","url":null,"abstract":"<p><strong>Background: </strong>Cognitive impairment is common after a stroke, but it can often go undetected. In this study, we investigated whether using gait and dual tasks could help detect cognitive impairment after stroke.</p><p><strong>Methods: </strong>We analyzed gait and neuropsychological data from 47 participants who were part of the Ontario Neurodegenerative Disease Research Initiative. Based on neuropsychological criteria, participants were categorized as impaired (n = 29) or cognitively normal (n = 18). Nested cross-validation was used for model training, hyperparameter tuning, and evaluation. Grid search with cross-validation was used to optimize the hyperparameters of a set of feature selectors and classifiers. Different gait tests were assessed separately.</p><p><strong>Results: </strong>The best classification performance was achieved using a comprehensive set of gait metrics, measured by the electronic walkway, that included dual-task costs while performing subtractions by ones. Using a Support Vector Machine (SVM), we could achieve a sensitivity of 96.6%, and a specificity of 61.1%. An optimized threshold of 27 in the Montreal Cognitive Assessment (MoCA) revealed lower classification performance than the gait metrics, although differences in classification results were not significant. Combining the classifications provided by MoCA with those provided by gait metrics in a majority voting approach resulted in a higher specificity of 72.2%, and a high sensitivity of 93.1%.</p><p><strong>Conclusions: </strong>Our results suggest that gait analysis can be a useful tool for detecting cognitive impairment in patients with cerebrovascular disease, serving as a suitable alternative or complement to MoCA in the screening for cognitive impairment.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"157"},"PeriodicalIF":3.3,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11963529/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143763046","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":"Development and multicentric external validation of a prognostic COVID-19 severity model based on thoracic CT.","authors":"Ine Dirks, Matías Nicolás Bossa, Abel Díaz Berenguer, Tanmoy Mukherjee, Hichem Sahli, Nikos Deligiannis, Emma Verelst, Bart Ilsen, Simon Van Eyndhoven, Lucie Seyler, Arne Witdouck, Gilles Darcis, Julien Guiot, Athanasios Giannakis, Jef Vandemeulebroucke","doi":"10.1186/s12911-025-02983-z","DOIUrl":"10.1186/s12911-025-02983-z","url":null,"abstract":"<p><strong>Background: </strong>Risk stratification of COVID-19 patients can support therapeutic decisions, planning and resource allocation in the hospital. In times of high incidence, a prognostic model based on data efficiently retrieved from one source can enable fast decision support.</p><p><strong>Methods: </strong>A model was developed to identify patients at risk of developing severe COVID-19 within one month based on their age, sex and imaging features extracted from the thoracic computed tomography (CT). The model was trained on publicly available data from the Study of Thoracic CT in COVID-19 (STOIC) challenge and validated on unseen data from the same study and an external, multicentric dataset. The model, trained on data acquired before any variant of concern dominated, was assessed separately on data collected at later stages of the pandemic when the delta and omicron variants were most prevalent.</p><p><strong>Results: </strong>A logistic regression based on handcrafted features was found to perform on par with a direct deep learning approach, and the former was selected for simplicity. Volumetric and intensity-based features of lesions and healthy lung parenchyma proved most predictive, in addition to patient age and sex. The model reached an area under the curve of 0.78 on the challenge test set and 0.74 on the external test set. The performance did not drop for the subset acquired at a later stage of the pandemic.</p><p><strong>Conclusions: </strong>A logistic regression utilizing features from thoracic CT and its metadata can provide rapid decision support by estimating short-term COVID-19 severity. Its stable performance underscores its potential for real-world clinical integration. By enabling rapid risk stratification using readily available imaging data, this approach can support early clinical decision-making, optimize resource allocation, and improve patient management, particularly during surges in COVID-19 cases. Furthermore, this study provides a foundation for future research on prognostic modelling in respiratory infections.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"156"},"PeriodicalIF":3.3,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11963321/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143762948","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":"Reusing data from HL7 CDA-based shared EHR systems for clinical trial conduct: a method for analyzing feasibility.","authors":"Georg Duftschmid, Florian Katsch, Gabriela Ciortuz, Dipak Kalra, Christoph Rinner","doi":"10.1186/s12911-025-02980-2","DOIUrl":"10.1186/s12911-025-02980-2","url":null,"abstract":"<p><strong>Background: </strong>Electronic health record (EHR) systems have been shown to represent a valuable source of data reuse in the design and conduct of clinical trials. Earlier work has mostly focused on institutional EHR systems. Shared EHR systems have been neglected so far, even though they are highly prevalent today and their characteristics (integrated data across a patient's care providers, standardized information model) make them attractive for the task. However, as they typically focus on a limited data set for the most common care situations, it remains unclear, whether shared EHR systems actually cover the data elements required for clinical trial conduct. In this paper we present a method, which allows shared EHR systems to be analyzed in this regard.</p><p><strong>Methods: </strong>We focus on shared EHR systems using HL7 CDA as this is currently the most-widely used content standard. For the data elements that are commonly used in clinical trials we refer to the EHR4CR reference list. The latter is semiautomatically mapped to the EHR system's information model using the open source tool ART-DECOR. For the final automatic analysis of the mappings, another open source tool is provided.</p><p><strong>Results: </strong>A stepwise approach was developed to analyze HL7 CDA-based shared EHR systems for their coverage of data elements that are relevant for clinical trials. All tools used in this work as well as all mappings are publicly accessible to make the method reusable and the results reproducible. We applied our approach to the Austrian nation-wide EHR system ELGA and showed that the latter allows the recording of 88% of all EHR4CR data elements, 77% in structured format.</p><p><strong>Conclusions: </strong>Our method allows HL7 CDA-based shared EHR systems to be easily analyzed to what extent their content could be reused in the context of clinical trials. The results for ELGA indicate that it has a substantial corresponding potential.</p><p><strong>Clinical trial number: </strong>Not applicable.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"155"},"PeriodicalIF":3.3,"publicationDate":"2025-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11963467/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143762949","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":"Men's experiences of decision-making in life-prolonging treatments of metastatic castration-resistant prostate cancer - wishing for a process adapted to personal preferences: a prospective interview study.","authors":"Sandra Doveson, Agneta Wennman-Larsen, Per Fransson, Lena Axelsson","doi":"10.1186/s12911-025-02985-x","DOIUrl":"10.1186/s12911-025-02985-x","url":null,"abstract":"<p><strong>Background: </strong>In the fast-expanding field of life-prolonging-treatment of metastatic, castration-resistant prostate cancer, treatment decision-making is very complex - both for patients and healthcare professionals since there is no \"one size that fits all\" in choosing treatment in this phase. Little research has been conducted about men's experiences of treatment decision-making in this advanced, incurable, phase. Hence, this study aimed to describe men's experiences of decision-making in life-prolonging treatments of metastatic castration-resistant prostate cancer.</p><p><strong>Methods: </strong>Seventeen men were recruited from four oncology clinics in Sweden and interviewed at baseline. Qualitative interviews (n = 31) were conducted over two years, the timepoints for subsequent interviews (10 men were interviewed twice or more) adhered to when each man switched or terminated life-prolonging treatment. Data was analysed with qualitative content analysis.</p><p><strong>Results: </strong>Initially, the men were adamant about proceeding with treatment. As their illness continued to progress, they gradually turned their focus more towards their well-being. They wished for continuity regarding treating physicians and constantly being assigned new physicians compromised the quality of care and complicated decision-making. In their decision-making, the men adapted their own approach to the approach taken by their physician, even if it was not an approach they had originally preferred. They wished for their role preferences to be respected. Most men had made treatment decisions collaboratively with their physician, but some described having taken on a more, or less, driving role in decision-making than they really wished for. Navigating healthcare was perceived as difficult and for some it thus felt necessary to pursue and coordinate their own care by e.g. using personal connections or contacting clinics ahead of referral. A part of treatment decision-making was forming a basis for a decision, in which the need for personalized information (quality, quantity and timing) came forth as important.</p><p><strong>Conclusions: </strong>When diagnosed with metastatic castration-resistant prostate cancer, men's preferences for their decision-making role, and perspectives on the treatment outcome need to be continuously addressed throughout their disease course. Improved continuity of care and a more personalised care approach should meet these patients' wishes and needs in this phase.</p><p><strong>Trial registration: </strong>Clinical trial number: Not applicable.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"153"},"PeriodicalIF":3.3,"publicationDate":"2025-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11960004/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143751230","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":"Association between patient characteristics and recommendations by medical on-call service 116117 in Germany: a cross sectional observational study.","authors":"Heike Hansen, Agata Menzel, Jan Hendrik Oltrogge-Abiry, Dagmar Lühmann, Martin Scherer, Ingmar Schäfer","doi":"10.1186/s12911-025-02970-4","DOIUrl":"10.1186/s12911-025-02970-4","url":null,"abstract":"<p><strong>Background: </strong>Use of emergency departments has increased in recent years. Different efforts address this problem, eg, medical on-call services. The basis of the DEMAND intervention is computer-assisted initial telephone assessment implemented at regional associations of statutory health insurance physicians in Germany. In this intervention, recommendations for healthcare settings were given over the telephone by medical staff. Recommendations were provided using the software SmED which calculates neural networks. This study aimed to analyse if patient characteristics are associated with the output of the intervention, ie, specific setting recommendations.</p><p><strong>Methods: </strong>Between January 2020 and March 2021, patients aged 18 years and older of the DEMAND intervention from eight intervention sites received a standardised postal survey. Recommended and used settings, and data on sociodemography, health status at the time of the emergency call, past health service use, and health literacy were collected by self-report. Multilevel, multivariable logistic regression models adjusted for random effects at the level of regions and months of observation within regions were conducted.</p><p><strong>Results: </strong>Of 9473 contacted individuals, 1756 (18.5 %) participated in the survey. Median age was 66 years, 59.0% were women and 30.2% living alone. The most frequently recommended service was emergency home visits (40.1%). Recommendations for this setting were associated with worse self-rated health (odds ratio 0.67, 95% confidence interval: 0.55/0.81, p < 0.001). Telephone counselling was associated with lower age (0.71, 0.59/0.85, p < 0.001), lower subjective treatment urgency (0.65, 0.51/0.82, p < 0.001) and health problems not classified as symptoms and complaints (0.41, 0.25/0.68, p = 0.001) or infections (0.22, 0.09/0.57, p = 0.002). Emergency departments were associated with better self-rated health (1.37, 1.11/1.70, p = 0.003) and health problems classified as injuries (3.12, 1.67/5.83, p < 0.001). Rescue service were associated with higher age (1.44, 1.15/1.81, p = 0.002) and higher subjective treatment urgency (2.51, 1.83/3.43, p < 0.001). General practices were associated with lower subjective treatment urgency (0.58, 0.44/0.76, p < 0.001) and health problems not classified as injuries (0.26, 0.10/0.68, p = 0.006). Emergency practices were associated with lower age (0.60, 0.48/0.74, p < 0.001), and specialist practices were associated with health problems classified as symptoms or complaints (3.75, 1.49/9.45, p = 0.005).</p><p><strong>Conclusions: </strong>Most associations between patient characteristics and recommendations were comprehensible and in line with the aim of the intervention. However, it should be clarified why patients with better self-rated health were more likely to receive recommendations for emergency departments.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"151"},"PeriodicalIF":3.3,"publicationDate":"2025-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11959809/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143751217","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":"Harness machine learning for multiple prognoses prediction in sepsis patients: evidence from the MIMIC-IV database.","authors":"Su-Zhen Zhang, Hai-Yi Ding, Yi-Ming Shen, Bing Shao, Yuan-Yuan Gu, Qiu-Hua Chen, Hai-Dong Zhang, Ying-Hao Pei, Hua Jiang","doi":"10.1186/s12911-025-02976-y","DOIUrl":"10.1186/s12911-025-02976-y","url":null,"abstract":"<p><strong>Background: </strong>Sepsis, a severe systemic response to infection, frequently results in adverse outcomes, underscoring the urgency for prompt and accurate prognostic tools. Machine learning methods such as logistic regression, random forests, and CatBoost, have shown potential in early sepsis prediction. The study aimed to create and verify a machine learning model capable of early prognostic identification of patients with sepsis in intensive care units (ICUs).</p><p><strong>Methods: </strong>Patients adhering to inclusion and exclusion criteria from the MIMIC-IV v2.2 database were divided into a training set and a validation set in a 7:3 ratio. Initially, we employed difference analysis to assess the significance of each variable and subsequently screened relevant features with multinomial logistic regression analysis. Logistic regression, random forest, and CatBoost algorithms were used to construct machine learning models to predict rapid recovery, chronic critical illness, and mortality in sepsis. The models were compared through several evaluation indexes including precision, accuracy, recall, F1 score, and the area under the receiver-operating-characteristic curve(AUC) in the validation set to select the optimal model. The best model was visualized and interpreted utilizing the Shapley Additive explanations method.</p><p><strong>Results: </strong>13174 sepsis patients were included. Post the screening process,26 clinical features were obtained to develop three distinct machine learning models. CatBoost exhibited superior performance among the three models with a weighted AUC of 0.771. The prognosis with the highest predictive performance was mortality (AUC = 0.804), followed by the prognoses of rapid recovery (AUC = 0.773) and chronic critical illness(AUC = 0.737). Urine output, respiratory rate, and temperature were the top three important features for the whole model prediction.</p><p><strong>Conclusion: </strong>The machine learning model developed leveraging the CatBoost algorithm demonstrates the latent capacity to identify sepsis prognosis early. It also suggests that interventions targeting factors such as urine output, respiratory status, and temperature in the early stage may potentially alter the adverse prognosis of sepsis patients. However, the model will still require further external validation in the future.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"152"},"PeriodicalIF":3.3,"publicationDate":"2025-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11959728/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143751220","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 novel network-level fused deep learning architecture with shallow neural network classifier for gastrointestinal cancer classification from wireless capsule endoscopy images.","authors":"Muhammad Attique Khan, Usama Shafiq, Ameer Hamza, Anwar M Mirza, Jamel Baili, Dina Abdulaziz AlHammadi, Hee-Chan Cho, Byoungchol Chang","doi":"10.1186/s12911-025-02966-0","DOIUrl":"10.1186/s12911-025-02966-0","url":null,"abstract":"<p><p>Deep learning has significantly contributed to medical imaging and computer-aided diagnosis (CAD), providing accurate disease classification and diagnosis. However, challenges such as inter- and intra-class similarities, class imbalance, and computational inefficiencies due to numerous hyperparameters persist. This study aims to address these challenges by presenting a novel deep-learning framework for classifying and localizing gastrointestinal (GI) diseases from wireless capsule endoscopy (WCE) images. The proposed framework begins with dataset augmentation to enhance training robustness. Two novel architectures, Sparse Convolutional DenseNet201 with Self-Attention (SC-DSAN) and CNN-GRU, are fused at the network level using a depth concatenation layer, avoiding the computational costs of feature-level fusion. Bayesian Optimization (BO) is employed for dynamic hyperparameter tuning, and an Entropy-controlled Marine Predators Algorithm (EMPA) selects optimal features. These features are classified using a Shallow Wide Neural Network (SWNN) and traditional classifiers. Experimental evaluations on the Kvasir-V1 and Kvasir-V2 datasets demonstrate superior performance, achieving accuracies of 99.60% and 95.10%, respectively. The proposed framework offers improved accuracy, precision, and computational efficiency compared to state-of-the-art models. The proposed framework addresses key challenges in GI disease diagnosis, demonstrating its potential for accurate and efficient clinical applications. Future work will explore its adaptability to additional datasets and optimize its computational complexity for broader deployment.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"150"},"PeriodicalIF":3.3,"publicationDate":"2025-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11956435/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143751214","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":"Leveraging large language models to mimic domain expert labeling in unstructured text-based electronic healthcare records in non-english languages.","authors":"Izzet Turkalp Akbasli, Ahmet Ziya Birbilen, Ozlem Teksam","doi":"10.1186/s12911-025-02871-6","DOIUrl":"10.1186/s12911-025-02871-6","url":null,"abstract":"<p><strong>Background: </strong>The integration of big data and artificial intelligence (AI) in healthcare, particularly through the analysis of electronic health records (EHR), presents significant opportunities for improving diagnostic accuracy and patient outcomes. However, the challenge of processing and accurately labeling vast amounts of unstructured data remains a critical bottleneck, necessitating efficient and reliable solutions. This study investigates the ability of domain specific, fine-tuned large language models (LLMs) to classify unstructured EHR texts with typographical errors through named entity recognition tasks, aiming to improve the efficiency and reliability of supervised learning AI models in healthcare.</p><p><strong>Methods: </strong>Turkish clinical notes from pediatric emergency room admissions at Hacettepe University İhsan Doğramacı Children's Hospital from 2018 to 2023 were analyzed. The data were preprocessed with open source Python libraries and categorized using a pretrained GPT-3 model, \"text-davinci-003,\" before and after fine-tuning with domain-specific data on respiratory tract infections (RTI). The model's predictions were compared against ground truth labels established by pediatric specialists.</p><p><strong>Results: </strong>Out of 24,229 patient records classified as poorly labeled, 18,879 were identified without typographical errors and confirmed for RTI through filtering methods. The fine-tuned model achieved a 99.88% accuracy, significantly outperforming the pretrained model's 78.54% accuracy in identifying RTI cases among the remaining records. The fine-tuned model demonstrated superior performance metrics across all evaluated aspects compared to the pretrained model.</p><p><strong>Conclusions: </strong>Fine-tuned LLMs can categorize unstructured EHR data with high accuracy, closely approximating the performance of domain experts. This approach significantly reduces the time and costs associated with manual data labeling, demonstrating the potential to streamline the processing of large-scale healthcare data for AI applications.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"154"},"PeriodicalIF":3.3,"publicationDate":"2025-03-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11959812/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143751225","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":"Developing a high-performance AI model for spontaneous intracerebral hemorrhage mortality prediction using machine learning in ICU settings.","authors":"Xiao-Han Vivian Yap, Kuan-Chi Tu, Nai-Ching Chen, Che-Chuan Wang, Chia-Jung Chen, Chung-Feng Liu, Tee-Tau Eric Nya, Ching-Lung Kuo","doi":"10.1186/s12911-025-02984-y","DOIUrl":"https://doi.org/10.1186/s12911-025-02984-y","url":null,"abstract":"<p><strong>Background: </strong>Spontaneous intracerebral hemorrhage (SICH) is a devastating condition that significantly contributes to high mortality rates. This study aims to construct a mortality prediction model for patients with SICH using four various artificial intelligence (AI) machine learning algorithms.</p><p><strong>Method: </strong>A retrospective analysis was conducted on electronic medical records of SICH patients aged 20 and above, admitted to Chi Mei Medical Center's intensive care unit between January 2016 and December 2021. The study utilized 37 features related to mortality. Predictive models were developed using logistic regression, Random forest, LightGBM, XGBoost, and Multi-layer Perceptron (MLP), with assessments of feature importance, and Area under the curve (AUC).</p><p><strong>Results: </strong>A total of 1451 SICH patients were enrolled. Factors associated with mortality included lower initial GCS scores (p < 0.001), pupillary changes (P < 0.001), kidney disease (p < 0.001), and respiratory failure requiring intubation (p < 0.001). Negative correlations were observed between mortality and pupil light reflexes, as well as GCS components E(r=-0.4602), V (r=-0.4132), M(r=-0.4082). Positive correlations were identified with vasopressors (r = 0.4464), FiO2 (r = 0.3901), and sedative-hypnotic drugs (r = 0.1178). XGBoost demonstrated the best predictive performance (AUC = 0.913), outperforming LR (0.899), RF (0.905), LightGBM (0.909), and MLP (0.892). The XGBoost model, utilizing both 18 and 36 features, continues to outperform both the Acute Physiology and Chronic Health Evaluation (APACHE II) (p < 0.001) and Sequential Organ Failure Assessment (SOFA) scoring systems (p < 0.001).</p><p><strong>Conclusion: </strong>This study successfully developed an AI mortality prediction model for SICH patients, with XGBoost exhibiting superior performance. The model, incorporating 18 key features, has been integrated into clinical practice assisting clinicians in treatment decisions and communication with patients' families.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"25 1","pages":"149"},"PeriodicalIF":3.3,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11951645/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143742451","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}