BMC Medical Informatics and Decision Making最新文献

{"title":"Prediction of 12-month recurrence of pancreatic cancer using machine learning and prognostic factors.","authors":"Raoof Nopour","doi":"10.1186/s12911-024-02766-y","DOIUrl":"10.1186/s12911-024-02766-y","url":null,"abstract":"<p><strong>Background and aim: </strong>Pancreatic cancer is lethal and prevalent among other cancer types. The recurrence of this tumor is high, especially in patients who did not receive adjuvant therapies. Early prediction of PC recurrence has a significant role in enhancing patients' prognosis and survival. So far, machine learning techniques have given us insight into favorable performance efficiency in various medical domains. So, this study aims to establish a prediction model based on machine learning to achieve better prediction on this topic.</p><p><strong>Materials and methods: </strong>In this retrospective research, we used data from 585 PC patient cases from January 2019 to November 2023 from three clinical centers in Tehran City. Ten chosen ensemble and non-ensemble algorithms were used to establish prediction models on this topic.</p><p><strong>Results: </strong>Random forest and support vector machine with an AU-ROC of approximately 0.9 obtained more performance efficiency regarding PC recurrence. Lymph node metastasis, tumor size, tumor grade, radiotherapy, and chemotherapy were the best factors influencing PC recurrence.</p><p><strong>Conclusion: </strong>Random forest and support vector machine algorithms demonstrated high-performance ability and clinical usability to improve doctors' decisions in achieving different therapeutic and diagnostic measures.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"24 1","pages":"339"},"PeriodicalIF":3.3,"publicationDate":"2024-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11566389/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142615282","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 of a cerebellar ataxia diagnosis model using conditional GAN-based synthetic data generation for visuomotor adaptation task.","authors":"Jinah Kim, Sung-Ho Woo, Taekyung Kim, Won Tae Yoon, Jung Hwan Shin, Jee-Young Lee, Jeh-Kwang Ryu","doi":"10.1186/s12911-024-02720-y","DOIUrl":"10.1186/s12911-024-02720-y","url":null,"abstract":"<p><p>This study proposes a synthetic data generation model to create a classification framework for cerebellar ataxia patients using trajectory data from the visuomotor adaptation task. The classification objectives include patients with cerebellar ataxia, age-matched normal individuals, and young healthy subjects. Synthetic data for the three classes is generated based on class conditions and random noise by leveraging a combination of conditional adversarial generative neural networks and reconstruction networks. This synthetic data, alongside real data, is utilized as training data for the patient classification model to enhance classification accuracy. The fidelity of the synthetic data is assessed visually to measure the validity and diversity of the generated data qualitatively while quantitatively evaluating distribution similarity to real data. Furthermore, the clinical efficacy of the patient classification model employing synthetic data is demonstrated by showcasing improved classification accuracy through a comparative analysis between results obtained using solely real data and those obtained when both real and synthetic data are utilized. This methodological approach holds promise in addressing data insufficiency in the digital healthcare domain, employing deep learning methodologies, and developing early disease diagnosis tools.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"24 1","pages":"336"},"PeriodicalIF":3.3,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11555814/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142615271","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":"Systematic review and meta-analysis of workload among medical records coders in China.","authors":"Yu Liu, Chao Wu, Meiling Cao, Chunyan Lei, Zhiqiang Zhou, Wenjing Ou","doi":"10.1186/s12911-024-02750-6","DOIUrl":"10.1186/s12911-024-02750-6","url":null,"abstract":"<p><strong>Importance: </strong>The homepage of medical records holds significant importance for national performance assessments, DIP settlement lists, and DRG payments. Coders, as auditors of the codes, wield a crucial influence on the quality of the medical records' homepage.</p><p><strong>Objective: </strong>To analyze the general situation of the allocation of medical record full-time coders in China.</p><p><strong>Data source: </strong>CNKI, Wanfang, VIP, PubMed and other databases were searched from database inception to November 31, 2023.</p><p><strong>Main outcomes and measures: </strong>The primary outcome was the allocation of medical records to full-time coders, with the workload of the coders being the primary focus. Secondary outcomes encompassed the professional background of the coders, including their academic qualifications, professional titles, possession of medical coding certificates, and years of experience in coding.</p><p><strong>Results: </strong>Eleven studies, comprising data from 1783 hospitals and 4448 coders, were analyzed. Among the coders, 61% had a medical-related professional background, 62% held a bachelor's degree or higher, 54% possessed an intermediate title or higher, 61% had coding certificates, and 51% had less than 5 years of work experience. The summary findings regarding the number of coders and coded medical records in secondary and tertiary hospitals indicated an average discharge rate of 22,704.0 per hospital in China. The number of coded cases averaged around 11,300. Specifically, coders in tertiary hospitals coded approximately 12,049 medical records on average, while those in secondary hospitals coded around 7,399 medical records.</p><p><strong>Conclusion and relevance: </strong>Our study highlights the shortage of medical record coding personnel in the majority of hospitals, coupled with a significant coding workload, low educational qualifications among staff, short working hours, and an imbalanced title structure. Given these findings, hospitals and relevant management authorities should prioritize the recruitment of highly educated professionals, streamline the assessment process for professional titles, alleviate the coding workload, and enhance coding quality.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"24 1","pages":"335"},"PeriodicalIF":3.3,"publicationDate":"2024-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11555796/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142615284","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 model for predicting the risk of significant liver fibrosis in patients with diabetic retinopathy.","authors":"Gangfeng Zhu, Na Yang, Qiang Yi, Rui Xu, Liangjian Zheng, Yunlong Zhu, Junyan Li, Jie Che, Cixiang Chen, Zenghong Lu, Li Huang, Yi Xiang, Tianlei Zheng","doi":"10.1186/s12911-024-02749-z","DOIUrl":"10.1186/s12911-024-02749-z","url":null,"abstract":"<p><strong>Background: </strong>Diabetic retinopathy (DR), a prevalent complication in patients with type 2 diabetes, has attracted increasing attention. Recent studies have explored a plausible association between retinopathy and significant liver fibrosis. The aim of this investigation was to develop a sophisticated machine learning (ML) model, leveraging comprehensive clinical datasets, to forecast the likelihood of significant liver fibrosis in patients with retinopathy and to interpret the ML model by applying the SHapley Additive exPlanations (SHAP) method.</p><p><strong>Methods: </strong>This inquiry was based on data from the National Health and Nutrition Examination Survey 2005-2008 cohort. Utilizing the Fibrosis-4 index (FIB-4), liver fibrosis was stratified across a spectrum of grades (F0-F4). The severity of retinopathy was determined using retinal imaging and segmented into four discrete gradations. A ten-fold cross-validation approach was used to gauge the propensity towards liver fibrosis. Eight ML methodologies were used: Extreme Gradient Boosting, Random Forest, multilayer perceptron, Support Vector Machines, Logistic Regression (LR), Plain Bayes, Decision Tree, and k-nearest neighbors. The efficacy of these models was gauged using metrics, such as the area under the curve (AUC). The SHAP method was deployed to unravel the intricacies of feature importance and explicate the inner workings of the ML model.</p><p><strong>Results: </strong>The analysis included 5,364 participants, of whom 2,116 (39.45%) exhibited notable liver fibrosis. Following random allocation, 3,754 individuals were assigned to the training set and 1,610 were allocated to the validation cohort. Nine variables were curated for integration into the ML model. Among the eight ML models scrutinized, the LR model attained zenith in both AUC (0.867, 95% CI: 0.855-0.878) and F1 score (0.749, 95% CI: 0.732-0.767). In internal validation, this model sustained its superiority, with an AUC of 0.850 and an F1 score of 0.736, surpassing all other ML models. The SHAP methodology unveils the foremost factors through importance ranking.</p><p><strong>Conclusion: </strong>Sophisticated ML models were crafted using clinical data to discern the propensity for significant liver fibrosis in patients with retinopathy and to intervene early.</p><p><strong>Practice implications: </strong>Improved early detection of liver fibrosis risk in retinopathy patients enhances clinical intervention outcomes.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"24 1","pages":"332"},"PeriodicalIF":3.3,"publicationDate":"2024-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11552118/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142615273","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":"Machine learning-based predictive model for post-stroke dementia.","authors":"Zemin Wei, Mengqi Li, Chenghui Zhang, Jinli Miao, Wenmin Wang, Hong Fan","doi":"10.1186/s12911-024-02752-4","DOIUrl":"10.1186/s12911-024-02752-4","url":null,"abstract":"<p><strong>Background: </strong>Post-stroke dementia (PSD), a common complication, diminishes rehabilitation efficacy and affects disease prognosis in stroke patients. Many factors may be related to PSD, including demographic, comorbidities, and examination characteristics. However, most existing methods are qualitative evaluations of independent factors, which ignore the interaction amongst various factors. Therefore, the purpose of this study is to explore the applicability of machine learning (ML) methods for predicting PSD.</p><p><strong>Methods: </strong>9 acceptable features were screened out by the Spearman correlation analysis and Boruta algorithm. We developed and evaluated 8 ML models: logistic regression, elastic net, k-nearest neighbors, decision tree, extreme gradient boosting, support vector machine, random forest, and multilayer perceptron.</p><p><strong>Results: </strong>A total of 539 stroke patients were included in this study. Among the 8 models used to predict PSD, extreme gradient boosting and random forest showed the highest area under the curve (AUC) of the receiver operating characteristic curve (ROC), with values of 0.7287 and 0.7285, respectively. The most important features for predicting PSD included age, high sensitivity C-reactive protein, stroke side and location, and the occurrence of cerebral hemorrhage.</p><p><strong>Conclusion: </strong>Our findings suggest that ML models, especially extreme gradient boosting, can best predict the risk of PSD.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"24 1","pages":"334"},"PeriodicalIF":3.3,"publicationDate":"2024-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11555950/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142615279","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":"An interactive dashboard for analyzing user interaction patterns in the i2b2 clinical data warehouse.","authors":"Lena Baum, Armin Müller, Marco Johns, Hammam Abu Attieh, Mehmed Halilovic, Vladimir Milicevic, Diogo Telmo Neves, Karen Otte, Anna Pasquier, Felix Nikolaus Wirth, Patrick Segelitz, Katharina Schönrath, Joachim E Weber, Fabian Prasser","doi":"10.1186/s12911-024-02748-0","DOIUrl":"10.1186/s12911-024-02748-0","url":null,"abstract":"<p><strong>Background: </strong>Clinical data warehouses provide harmonized access to healthcare data for medical researchers. Informatics for Integrating Biology and the Bedside (i2b2) is a well-established open-source solution with the major benefit that data representations can be tailored to support specific use cases. These data representations can be defined and improved via an iterative approach together with domain experts and the medical researchers using the platform. To facilitate these discussions, it is important to understand how users interact with the system.</p><p><strong>Objective: </strong>The objective of this work was to develop metrics for describing user interactions with clinical data warehouses in general and i2b2 in particular. Moreover, we aimed to develop a dashboard featuring interactive visualizations that inform data engineers and data stewards about potential improvements.</p><p><strong>Methods: </strong>We first identified metrics for different data usage dimensions and extracted the relevant metadata about previous user queries from the i2b2 database schema for further analysis. We then implemented associated visualizations in Python and integrated the results into an interactive dashboard using Dash.</p><p><strong>Results: </strong>The identified categories of metrics include frequency of use, session duration, and use of functionality and features. We created a dashboard that extends our local i2b2 data warehouse platform, focusing on the latter category, further broken down into the number of queries, frequently queried concepts, and query complexity. The implementation is available as open-source software.</p><p><strong>Conclusion: </strong>A range of metrics can be derived from metadata logged in the i2b2 database schema to provide data engineers and data stewards with a comprehensive understanding of how users interact with the platform. This can help to identify the strengths and limitations of specific instances of the platform for specific use cases and aid their iterative improvement.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"24 1","pages":"333"},"PeriodicalIF":3.3,"publicationDate":"2024-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11556165/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142615269","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 national platform for advancing self-care processes for the most common illnesses and conditions: designing, evaluating, and implementing.","authors":"Khadijeh Moulaei, Somayeh Salehi, Masoud Shahabian, Babak Sabet, Farshid Rezaei, Adrina Habibzadeh, Mohammad Reza Afrash","doi":"10.1186/s12911-024-02744-4","DOIUrl":"10.1186/s12911-024-02744-4","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Effective self-care practices are crucial for maintaining health and well-being, as inadequate self-care can lead to increased health risks and decreased overall quality of life. To address these issues, one promising approach involves leveraging progressive web app (PWA) platforms to educate and empower individuals with necessary self-care services. This study aims to design, implement, and evaluate a national self-care PWA platform, aiming to enhance accessibility and effectiveness in promoting health and self-care practices. The platform designed to improve self-care processes can be utilized by mothers, children, adolescents, youth, adults, and patients with emotional and mental disorders.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;This study was conducted in three phases. In the first phase, during 35 meetings with 19 health care providers including physicians and another group of professionals, the most common illnesses and conditions that require self-care were identified. Platform capabilities were then assessed based on stakeholder opinions. Subsequently, during 15 meetings 19 health care providers identified a comprehensive list of conditions benefiting from dedicated decision aids to enhance individuals' self-care processes. In the second phase, a progressive web app platform was designed based on these common illnesses and conditions and capabilities and subsequently evaluated. To usability evaluation the platform, 26 evaluators utilized the system for two weeks. The QUIS 5.5 questionnaire was employed for evaluation, and the results were analyzed using SPSS 23. In the final phase, the system was implemented at the Smart University of Medical Sciences (SMUMS), affiliated with the Ministry of Health and Medical Education in Iran.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;Based on the most common illnesses and conditions (n = 87) and identified capabilities, the national self-care platform was designed with eight sections catering to 'Maternal and child health services,' 'Mothers,' 'Infants,' 'Teenagers,' 'Adults,' 'Elderly,' 'Health of All Age Groups,' 'Patients with Mental and Emotional Health Disorders,' and 'General Information' for user education. Furthermore, the platform features 54 decision aids (DA), teleconsultation services, and a self-care magazine (Access link: https://khodmoragheb.ir/ ). These features were integrated to provide comprehensive support and resources for self-care. A mean exceeding 7 was attained across all evaluated dimensions, indicating that evaluators generally agreed the platform performed well.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusion: &lt;/strong&gt;The designed national self-care platform offers a promising solution for managing healthcare challenges. This innovative approach addresses the specific needs of individuals and extends its reach to Persian-speaking patients worldwide, fostering a global impact. By embracing self-care practices on an international scale, this platform contributes to a more inclusive a","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"24 1","pages":"331"},"PeriodicalIF":3.3,"publicationDate":"2024-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11542451/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142589396","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":"Predictive modeling of preoperative acute heart failure in older adults with hypertension: a dual perspective of SHAP values and interaction analysis.","authors":"Qili Yu, Zhiyong Hou, Zhiqian Wang","doi":"10.1186/s12911-024-02734-6","DOIUrl":"10.1186/s12911-024-02734-6","url":null,"abstract":"<p><strong>Background: </strong>In older adults with hypertension, hip fractures accompanied by preoperative acute heart failure significantly elevate surgical risks and adverse outcomes, necessitating timely identification and management to improve patient outcomes.</p><p><strong>Research objective: </strong>This study aims to enhance the early recognition of acute heart failure in older hypertensive adults prior to hip fracture surgery by developing a predictive model using logistic regression (LR) and machine learning methods, optimizing preoperative assessment and management.</p><p><strong>Methods: </strong>Employing a retrospective study design, we analyzed hypertensive older adults who underwent hip fracture surgery at Hebei Medical University Third Hospital from January 2018 to December 2022. Predictive models were constructed using LASSO regression and multivariable logistic regression, evaluated via nomogram charts. Five additional machine learning methods were utilized, with variable importance assessed using SHAP values and the impact of key variables evaluated through multivariate correlation analysis and interaction effects.</p><p><strong>Results: </strong>The study included 1,370 patients. LASSO regression selected 18 key variables, including sex, age, coronary heart disease, pulmonary infection, ventricular arrhythmias, acute myocardial infarction, and anemia. The logistic regression model demonstrated robust performance with an AUC of 0.753. Although other models outperformed it in sensitivity and F1 score, logistic regression's discriminative ability was significant for clinical decision-making. The Gradient Boosting Machine model, notable for a sensitivity of 95.2%, indicated substantial capability in identifying patients at risk, crucial for reducing missed diagnoses.</p><p><strong>Conclusion: </strong>We developed and compared efficacy of predictive models using logistic regression and machine learning, interpreting them with SHAP values and analyzing key variable interactions. This offers a scientific basis for assessing preoperative heart failure risk in older adults with hypertension and hip fractures, providing significant guidance for individualized treatment strategies and underscoring the value of applying machine learning in clinical settings.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"24 1","pages":"329"},"PeriodicalIF":3.3,"publicationDate":"2024-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11539738/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142589405","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":"Experiences and needs of older patients with stroke in China involved in rehabilitation decision-making: a qualitative study.","authors":"Zining Guo, Sining Zeng, Keyu Ling, Shufan Chen, Ting Yao, Haihan Li, Ling Xu, Xiaoping Zhu","doi":"10.1186/s12911-024-02735-5","DOIUrl":"10.1186/s12911-024-02735-5","url":null,"abstract":"<p><strong>Background: </strong>Shared decision-making is recommended for stroke rehabilitation. However, the complexity of the rehabilitation modalities exposes patients to decision-making conflicts, exacerbates their disabilities, and diminishes their quality of life. This study aimed to explore the experiences and needs of older patients with stroke in China during rehabilitation decision-making, providing a reference for developing decision-support strategies.</p><p><strong>Methods: </strong>A qualitative phenomenological design was used to explore the experiences and needs of older patients with stroke in China. Purposive sampling was used to recruit 31 older Chinese patients with stroke. The participants participated in face-to-face, semi-structured, and in-depth interviews. Data were analyzed using inductive thematic analysis.</p><p><strong>Results: </strong>The key themes identified include (1) mixed feelings in shared decision-making, (2) multiple barriers hinder the possibility of participating in shared decision-making, (3) Delegating rehabilitation decisions to surrogates, (4) gaps between reality and expectation, and (5) decision fatigue from lack of continuity in the rehabilitation health care system.</p><p><strong>Conclusions: </strong>Older patients with stroke in China have complex rehabilitation decision-making experiences and needs and face multiple obstacles when participating in shared decision-making. They lack an effective shared decision-making support system to assist them. Providing patients with comprehensive support (such as emotional and informational), strengthening the construction of a continuous rehabilitation system, alleviating economic pressure, and promoting patient participation in rehabilitation decision-making are necessary.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"24 1","pages":"330"},"PeriodicalIF":3.3,"publicationDate":"2024-11-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11539789/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142589400","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":"Analysis and knowledge extraction of newborn resuscitation activities from annotation files.","authors":"Mohanad Abukmeil, Øyvind Meinich-Bache, Trygve Eftestøl, Siren Rettedal, Helge Myklebust, Thomas Bailey Tysland, Hege Ersdal, Estomih Mduma, Kjersti Engan","doi":"10.1186/s12911-024-02736-4","DOIUrl":"10.1186/s12911-024-02736-4","url":null,"abstract":"<p><p>Deprivation of oxygen in an infant during and after birth leads to birth asphyxia, which is considered one of the leading causes of death in the neonatal period. Adequate resuscitation activities are performed immediately after birth to save the majority of newborns. The primary resuscitation activities include ventilation, stimulation, drying, suction, and chest compression. While resuscitation guidelines exist, little research has been conducted on measured resuscitation episodes. Objective data collected for measuring and registration of the executed resuscitation activities can be used to generate temporal timelines. This paper is primarily aimed to introduce methods for analyzing newborn resuscitation activity timelines, through visualization, aggregation, redundancy and dimensionality reduction. We are using two datasets: 1) from Stavanger University Hospital with 108 resuscitation episodes, and 2) from Haydom Lutheran Hospital with 76 episodes. The resuscitation activity timelines were manually annotated, but in future work we will use the proposed method on automatically generated timelines from video and sensor data. We propose an encoding generator with unique codes for combination of activities. A visualization of aggregated episodes is proposed using sparse nearest neighbor graph, shown to be useful to compare datasets and give insights. Finally, we propose a method consisting of an autoencoder trained for reducing redundancy in encoded resuscitation timeline descriptions, followed by a neighborhood component analysis for dimensionality reduction. Visualization of the resulting features shows very good class separability and potential for clustering the resuscitation files according to the outcome of the newborns as dead, admitted to NICU or normal. This shows great potential for extracting important resuscitation patterns when tested on larger datasets.</p>","PeriodicalId":9340,"journal":{"name":"BMC Medical Informatics and Decision Making","volume":"24 1","pages":"327"},"PeriodicalIF":3.3,"publicationDate":"2024-11-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11539679/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142582333","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}