International Journal of Medical Informatics最新文献

{"title":"Smart data-driven medical decisions through collective and individual anomaly detection in healthcare time series.","authors":"Farbod Khanizadeh, Alireza Ettefaghian, George Wilson, Amirali Shirazibeheshti, Tarek Radwan, Cristina Luca","doi":"10.1016/j.ijmedinf.2024.105696","DOIUrl":"https://doi.org/10.1016/j.ijmedinf.2024.105696","url":null,"abstract":"<p><strong>Background: </strong>Anomalies in healthcare refer to deviation from the norm of unusual or unexpected patterns or activities related to patients, diseases or medical centres. Detecting these anomalies is crucial for timely interventions and efficient decision-making, helping to identify issues like operational inefficiencies, fraud and emerging health complications.</p><p><strong>Objectives: </strong>This study presents a novel method for detecting both collective and individual anomalies in healthcare data through time series analysis using unsupervised machine learning. The dual-strategy approach leverages two methodologies: a 'practice centre-based approach' which monitors changes across different practice centres and a 'process-based approach' which focuses on identifying anomalies within individual centres. The former allows for early detection of systemic issues, while the latter highlights specific irregularities within a centre's operations.</p><p><strong>Methods: </strong>The study utilised a dataset over 500,000 medical records from multiple GP practice centres in the UK collected between 2018-2023. Data are clustered using DBSCAN to identify collective anomalies from deviations from linear trends in consecutive two-month scatterplots. Individual anomalies are identified by examining the SOM-clustered time series of various medical processes within a specific practice centre, where graphs show deviation from the typical pattern.</p><p><strong>Findings: </strong>Our approach addresses some challenges posed by the complexity and sensitivity of healthcare data by not requiring personal information. The method offers accurate visual representations making the data accessible and interpretable for non-technical users. Unlike traditional methods focusing solely on subsequence anomalies, our technique analyses the collective behaviour across multiple time series providing a more comprehensive perspective.</p><p><strong>Conclusion: </strong>This study underscores the importance of integrating unsupervised anomaly detection with clinical expertise to ensure that statistically anomalous patterns align with clinical relevance. The dual-strategy clustering method holds significant potential for enabling timely interventions, proactively identifying potential crises, and ultimately contributing to better decision-making and operational efficiency within the healthcare sector.</p>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"194 ","pages":"105696"},"PeriodicalIF":3.7,"publicationDate":"2024-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142683780","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An interpretable machine learning scoring tool for estimating time to recurrence readmissions in stroke patients.","authors":"Xiao Luo, Xin Cui, Rui Wang, Yi Cheng, Ronghui Zhu, Yaoyong Tai, Cheng Wu, Jia He","doi":"10.1016/j.ijmedinf.2024.105704","DOIUrl":"https://doi.org/10.1016/j.ijmedinf.2024.105704","url":null,"abstract":"<p><strong>Background: </strong>Stroke recurrence readmission poses an additional burden on both patients and healthcare systems. Risk stratification aims to accurately divide patients into groups to provide targeted interventions at reducing readmission. To accurately predict short and intermediate-term risks of readmission and provide information for further temporal risk stratification, we developed and validated an interpretable machine learning risk scoring tool.</p><p><strong>Methods: </strong>In this retrospective study, all stroke admission episodes from January 1st 2015 to December 31st 2019 were obtained from the Shanghai Health and Health Development Research Centre database, which covers medical records of all patients hospitalized in 436 medical institutes in Shanghai. The outcome was time to stroke recurrence readmission within 90 days post discharge. The Score for Stroke Recurrence Readmission Prediction (SSRRP) tool was derived via an interpretable machine learning-based system for time-to-event outcomes. SSRRP as six-variable survival score includes sequelae, length of stay, type of stroke, random plasma glucose, medical expense payment, and number of hospitalizations.</p><p><strong>Results: </strong>A total of 339,212 S admission episodes were finally included in the whole cohort. Among them, 217,393 episodes were included in the training dataset, 54,347 episodes in the internal validation dataset, and 67,472 in the temporal validation dataset. Readmission within 90 days was documented in 33922(9.97 %) episodes, with a median time to emergency readmission of 19 days (Interquartile range: 8-43). In the temporal validation dataset, the SSRRP achieved an integrated area under the curve of 0.730(95 % CI, 0.724-0.737). In addition, SSRRP demonstrated good calibration and clinical benefit rate.</p><p><strong>Conclusions: </strong>In this retrospective cohort study, the SSRRP, a parsimonious and point-based scoring tool, was developed to predict the risk of recurrent readmission for stroke. It also provided accurate information on the time to stroke readmission, enabling further temporal risk stratification and informed clinical decision-making.</p>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"194 ","pages":"105704"},"PeriodicalIF":3.7,"publicationDate":"2024-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142677758","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Recognition of autism in subcortical brain volumetric images using autoencoding-based region selection method and Siamese Convolutional Neural Network.","authors":"Anas Abu-Doleh, Isam F Abu-Qasmieh, Hiam H Al-Quran, Ihssan S Masad, Lamis R Banyissa, Marwa Alhaj Ahmad","doi":"10.1016/j.ijmedinf.2024.105707","DOIUrl":"https://doi.org/10.1016/j.ijmedinf.2024.105707","url":null,"abstract":"<p><strong>Background: </strong>Autism Spectrum Disorder (ASD) is a neurodevelopmental condition that affects social interactions and behavior. Accurate and early diagnosis of ASD is still challenging even with the improvements in neuroimaging technology and machine learning algorithms. It's challenging because of the wide range of symptoms, delayed appearance of symptoms, and the subjective nature of diagnosis. In this study, the aim is to enhance ASD recognition by focusing on brain subcortical regions, which are critical for understanding ASD pathology.</p><p><strong>Methodology: </strong>First, subcortical structures were extracted from a collection of brain MRI datasets using sophisticated processing steps. Next, a 3D autoencoder was trained on these 3D images to help identify brain regions related to ASD. Two distinct feature selection methods were then applied to the features extracted from the encoder. The highest-ranked features were iteratively selected and increased to reconstruct a specific percentage of the brain that represents the most relevant parts for ASD. Finally, a Siamese Convolutional Neural Network (SCNN) was employed as the classifier model.</p><p><strong>Results: </strong>The 3D autoencoder stage helped in identifying and reconstructing the significant subcortical regions related to ASD. Based on the studied dataset, high agreement in regions like the Putamen and Pallidum indicated the critical nature of these structures in distinguishing Autism from controls cases. Subsequently, applying SCNN on these selected subcortical regions yielded promising results. For example, using the classifier on the output regions identified by the Mutual Information (MI) features selection method achieved the highest accuracy of 0.66.</p><p><strong>Conclusions: </strong>This study shows that using a two-stage model involving autoencoder and SCNN can notably improve the classification of ASD from brain MRI volumetric images. Applying an iterative feature extraction approach allowed to achieve a more accurate identification of ASD-related brain areas. This two-stage approach not only improved classification performance but also enhanced the interpretability of the neuroimaging data.</p>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"194 ","pages":"105707"},"PeriodicalIF":3.7,"publicationDate":"2024-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142677760","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predicting Fear of Breast Cancer Recurrence in women five years after diagnosis using Machine Learning and healthcare reimbursement data from the French nationwide VICAN survey","authors":"Mamoudou Koume ,&nbsp;Lorène Seguin ,&nbsp;Julien Mancini ,&nbsp;Marc-Karim Bendiane ,&nbsp;Anne-Déborah Bouhnik ,&nbsp;Raquel Urena","doi":"10.1016/j.ijmedinf.2024.105705","DOIUrl":"10.1016/j.ijmedinf.2024.105705","url":null,"abstract":"<div><h3>Objective</h3><div>A major concern for cancer survivors after treatment is the Fear of Cancer Recurrence (FCR), which is the fear that cancer will reappear or progress. This fear can be exacerbated by medical uncertainty about the future, leading to harmful obsession and having a negative impact on quality of life. This study aims to develop a predictive Machine Learning (ML) model using healthcare reimbursement data to better predict FCR and understand the factors influencing FCR in women with breast cancer five years after their diagnosis.</div></div><div><h3>Materials and Methods</h3><div>We used data from the VICAN (VIe après le CANcer) survey to propose an interpretable model to identify patients at risk of developing clinical FCR. The reimbursement data for each patient were analyzed beyond the first two years of treatment, excluding the initial phase influenced by the cancer curative therapeutic process. Data experiments were conducted, including the use of algorithms such as Random Forest, Support Vector Machines, Gradient Boosting, eXtreme Gradient Boosting, and Multilayer Perceptron. The AUC was used to choose the optimal model.</div></div><div><h3>Results</h3><div>The dataset is composed of 918 patients classified regarding FCR. The experimental design incorporated classification levels of medications, biological and medical procedures. Subsequently, data was generated for two experiments, facilitating examination at the ultimate healthcare classification level in Experiment 1, while Experiment 2 targeted the penultimate classification level. Overall, the best-performing model achieved an AUC of 66%. This demonstrates some effectiveness of the algorithms in detecting patients at risk of developing clinical FCR and encourages further investigations to enhance the model's performance and assess its generalizability.</div></div><div><h3>Conclusion</h3><div>ML applied to reimbursement data has shown promise in predicting FCR, aiding in the identification of patients at risk of developing it. The results pave the way for personalized prevention and intervention strategies, representing a significant advancement in postcancer care focusing on the needs of survivors.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"193 ","pages":"Article 105705"},"PeriodicalIF":3.7,"publicationDate":"2024-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142640310","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep learning-driven ultrasound equipment quality assessment with ATS-539 phantom data","authors":"Dong Hoon Jang ,&nbsp;Ji Won Heo ,&nbsp;Kyu Hong Lee ,&nbsp;Ro Woon Lee ,&nbsp;Tae Ran Ahn ,&nbsp;Hyun Gyu Lee","doi":"10.1016/j.ijmedinf.2024.105698","DOIUrl":"10.1016/j.ijmedinf.2024.105698","url":null,"abstract":"<div><h3>Introduction</h3><div>Ultrasound equipment provides real-time visualization of internal organs, essential for early disease detection and diagnosis. However, poor-quality ultrasound images can compromise diagnostic accuracy and increase the risk of misdiagnosis. Quality assessments are often subjective, relying on the evaluator's experience and interpretation, which can vary.</div></div><div><h3>Methods</h3><div>This study introduces a two-stage deep learning framework designed to objectively assess ultrasound image quality using phantom data across three key parameters: ‘Dead zone’, ‘Axial/lateral resolution’, and ‘Gray scale and dynamic range’. Stage 1 automatically extracts regions of interest for each parameter, while Stage 2 employs detection or classification models to evaluate image quality within these regions. To generate an overall equipment quality score, a logistic regression model combines the weighted results from each parameter.</div></div><div><h3>Results</h3><div>The classification model demonstrated high performance across datasets, achieving AUC scores of 98.6% for ‘Dead zone’, 87.7% for ‘Axial/lateral resolution’, and 96.0% for ‘Gray scale and dynamic range’. Further analysis using guideline-compliant images of individual devices showed AUC scores of 98.2%, 92.8%, and 100%, respectively. These findings highlight deep learning's potential for quantitative and objective assessments of ultrasound image quality. Ultimately, this framework provides a streamlined approach to quality management, enabling consistent quality control and efficient scoring-based evaluation of ultrasound equipment.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"193 ","pages":"Article 105698"},"PeriodicalIF":3.7,"publicationDate":"2024-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142633144","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Usefulness of self-guided digital services among mental health patients: The role of health confidence and sociodemographic characteristics","authors":"Iiris Hörhammer ,&nbsp;Johanna Suvanto ,&nbsp;Maarit Kinnunen ,&nbsp;Sari Kujala","doi":"10.1016/j.ijmedinf.2024.105693","DOIUrl":"10.1016/j.ijmedinf.2024.105693","url":null,"abstract":"<div><h3>Background</h3><div>Remote services provided via telephone or the internet have become an essential part of mental health provision. Alongside services involving healthcare personnel (HCP), self-guided digital services hold great promise for improved self-management and health outcomes without increasing the burden on HCP. Therefore, better understanding of patients’ use and experienced benefits of these services are needed. This study investigated how health confidence and sociodemographic background are associated with mental health patients’ experiences of self-guided digital services.</div></div><div><h3>Methods</h3><div>This cross-sectional survey study was performed in 2022 at a Finnish Mental Health and Substance Abuse Services (MHSAS) unit of a regional public service provider that serves a population of about 163<!--> <!-->000 people. All patients who had visited the unit up to 6 months before the study were invited to respond to an online survey on their experiences with the remote MHSAS. We report the average subjective usefulness of telephone, guided digital and self-guided digital services. Regression models were fitted to study the associations of patient characteristics with use of any digital service, and with experienced usefulness of self-guided digital services.</div></div><div><h3>Findings</h3><div>The respondents (n = 438) rated the usefulness of telephone, guided digital and self-guided digital services similarly (4.0/5.0, 3.9/5.0, and 3.9/5.0, respectively). Health confidence was associated with not using digital services at all as well as with high perceived usefulness of self-guided services. While elderly patients were more likely to avoid using digital services, age was not associated with experienced usefulness of self-guided digital services. No association between unemployment status and experiences of digital services was found.</div></div><div><h3>Conclusions</h3><div>Different types of remote services are perceived as beneficial by mental health patients. To ensure effectiveness and equity, patients’ health confidence should be considered when directing them to self-guided services. Elderly mental health patients who use digital services are equally able as younger patients to benefit from self-guided services.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"194 ","pages":"Article 105693"},"PeriodicalIF":3.7,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142656282","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Data-driven explainable machine learning for personalized risk classification of myasthenic crisis.","authors":"Sivan Bershan, Andreas Meisel, Philipp Mergenthaler","doi":"10.1016/j.ijmedinf.2024.105679","DOIUrl":"https://doi.org/10.1016/j.ijmedinf.2024.105679","url":null,"abstract":"<p><strong>Objective: </strong>Myasthenic crisis (MC) is a critical progression of Myasthenia gravis (MG), requiring intensive care treatment and invasive therapies. Classifying patients at high-risk for MC facilitates treatment decisions such as changes in medication or the need for mechanical ventilation and helps prevent disease progression by decreasing treatment-induced stress on the patient. Here, we investigated whether it is possible to reliably classify MG patients into groups at low or high risk of MC based entirely on routine medical data using explainable machine learning (ML).</p><p><strong>Methods: </strong>In this single-center pseudo-prospective cohort study, we investigated the precision of ML models trained with real-world routine clinical data to identify MG patients at risk for MC, and identified explainable distinctive features for the groups. 51 MG patients, including 13 MC, were used for model training based on real-world clinical data available from the hospital management system. Patients were classified to high or low risk for MC using Lasso regression or random forest ML models.</p><p><strong>Results: </strong>The mean cross-validated AUC classifying MG patients as high or low risk for MC based on simple or compound features derived from real-world clinical data showed a predictive accuracy of 68.8% for a regularized Lasso regression and 76.5% for a random forest model. Studying feature importance across 5100 model runs identified explainable features to distinguish MG patients at high or low risk for MC. Feature importance scores suggested that multimorbidity may play a role in risk classification.</p><p><strong>Conclusion: </strong>This study establishes feasibility and proof-of-concept for risk classification of MC based on real-world routine clinical data using ML with explainable features and variance control at the point of care. Future research on ML-based prediction of MC should include multi-center, multinational data collection, more in-depth data per patient, more patients, and an attention-based ML model to include free-text.</p>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"194 ","pages":"105679"},"PeriodicalIF":3.7,"publicationDate":"2024-11-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142683779","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Electronic health records and data exchange in the WHO European region: A subregional analysis of achievements, challenges, and prospects","authors":"Roberto Tornero Costa ,&nbsp;Keyrellous Adib ,&nbsp;Nagui Salama ,&nbsp;Stefania Davia ,&nbsp;Antonio Martínez Millana ,&nbsp;Vicente Traver ,&nbsp;Karapet Davtyan","doi":"10.1016/j.ijmedinf.2024.105687","DOIUrl":"10.1016/j.ijmedinf.2024.105687","url":null,"abstract":"&lt;div&gt;&lt;div&gt;Electronic health record (EHR) systems are powerful tools that enhance healthcare quality. They improve efficiency, enable data exchange, and ensure authorized access to patient information. In 2022, the World Health Organization Regional Office for Europe (WHO EURO) conducted a survey to assess the digital health capabilities of the 53 Member States. This article provides a sub-regional analysis of the status of EHR systems and major barriers to their implementation, their readiness for information sharing, and the access and re-use of EHR data.&lt;/div&gt;&lt;div&gt;Generally, EHR implementation and national data exchange are at an advanced stage in the region, though achievements and challenges vary across subregions. While more Member States in the Eastern, Western, and Southern Europe subregions reported having centralized national EHR systems, the situation is more diverse in Northern Europe and the Asian subregions, where both centralized and decentralized EHR systems are in use. Significant barriers to EHR implementation, including funding, technical capacity, competing priorities, and lack of interoperability standards are frequently cited, while others like demand, knowledge or acceptance challenges are not reported as significant. Significant barriers were reported the most by the Central Asia subregion, while barriers had least significance in Western Europe. Five out of the six subregions reported a wide adoption of national strategies and have dedicated agencies to ensure interoperability and secure data exchange. However, only 29 Member States have established legal requirements for healthcare providers to adopt EHR systems that conform to national standards for both clinical terminology and electronic messaging, with this being most notable in Western, Eastern, and Northern Europe, and the lowest percentage of Member States in Central Asia. All Member States of the six sub regions have passed privacy and data protection legislation. The use of EHR data is widely regulated, with only five remaining Member States of WHO Europe to develop EHR legislation distributed across subregions (Southern Europe, Western and Central Asia).&lt;/div&gt;&lt;div&gt;Looking ahead, Member States are encouraged to define national legislation governing EHR systems and their use, while ensuring the interconnectivity of the local and regional EHR systems. Sustainable funding should be allocated to the development and maintenance of these systems. Efforts should also focus on creating comprehensive roadmaps for the full implementation of health data standards, addressing interoperability at local and regional levels, and developing quality management systems for testing and certification. Additionally, monitoring and evaluation should be conducted to assess whether EHRs are contributing to national health objectives. Finally, engaging patients and intersectoral partners will be key to developing a more patient-centered approach, ensuring that EHR systems meet patient ne","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"194 ","pages":"Article 105687"},"PeriodicalIF":3.7,"publicationDate":"2024-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142655800","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Balancing accuracy and Interpretability: An R package assessing complex relationships beyond the Cox model and applications to clinical prediction","authors":"Diana Shamsutdinova ,&nbsp;Daniel Stamate ,&nbsp;Daniel Stahl","doi":"10.1016/j.ijmedinf.2024.105700","DOIUrl":"10.1016/j.ijmedinf.2024.105700","url":null,"abstract":"<div><h3>Background</h3><div>Accurate and interpretable models are essential for clinical decision-making, where predictions can directly impact patient care. Machine learning (ML) survival methods can handle complex multidimensional data and achieve high accuracy but require post-hoc explanations. Traditional models such as the Cox Proportional Hazards Model (Cox-PH) are less flexible, but fast, stable, and intrinsically transparent. Moreover, ML does not always outperform Cox-PH in clinical settings, warranting a diligent model validation. We aimed to develop a set of R functions to help explore the limits of Cox-PH compared to the tree-based and deep learning survival models for clinical prediction modelling, employing ensemble learning and nested cross-validation.</div></div><div><h3>Methods</h3><div>We developed a set of R functions, publicly available as the package “survcompare”. It supports Cox-PH and Cox-Lasso, and Survival Random Forest (SRF) and DeepHit are the ML alternatives, along with the ensemble methods integrating Cox-PH with SRF or DeepHit designed to isolate the marginal value of ML. The package performs a repeated nested cross-validation and tests for statistical significance of the ML’s superiority using the survival-specific performance metrics, the concordance index, time-dependent AUC-ROC and calibration slope.</div><div>To get practical insights, we applied this methodology to clinical and simulated datasets with varying complexities and sizes.</div></div><div><h3>Results</h3><div>In simulated data with non-linearities or interactions, ML models outperformed Cox-PH at sample sizes ≥ 500. ML superiority was also observed in imaging and high-dimensional clinical data. However, for tabular clinical data, the performance gains of ML were minimal; in some cases, regularised Cox-Lasso recovered much of the ML’s performance advantage with significantly faster computations. Ensemble methods combining Cox-PH and ML predictions were instrumental in quantifying Cox-PH’s limits and improving ML calibration. Traditional models like Cox-PH or Cox-Lasso should not be overlooked while developing clinical predictive models from tabular data or data of limited size.</div></div><div><h3>Conclusion</h3><div>Our package offers researchers a framework and practical tool for evaluating the accuracy-interpretability trade-off, helping make informed decisions about model selection.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"194 ","pages":"Article 105700"},"PeriodicalIF":3.7,"publicationDate":"2024-11-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142640312","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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 adjusted sequential CUSUM-analyses are superior to cross-sectional analysis of excess mortality after surgery","authors":"Florian Bösch ,&nbsp;Stina Schild-Suhren ,&nbsp;Elif Yilmaz ,&nbsp;Michael Ghadimi ,&nbsp;Athanasios Karampalis ,&nbsp;Nikolaus Börner ,&nbsp;Markus Bo Schoenberg","doi":"10.1016/j.ijmedinf.2024.105684","DOIUrl":"10.1016/j.ijmedinf.2024.105684","url":null,"abstract":"<div><h3>Background</h3><div>The assessment of clinical outcome quality, particularly in surgery, is crucial for healthcare improvement. Traditional cross-sectional analyses often fall short in timely and systematic identification of clinical quality issues. This study explores the efficacy of machine learning adjusted sequential CUSUM (Cumulative Sum) analyses in monitoring post-surgical mortality.</div></div><div><h3>Material and methods</h3><div>Utilizing the Global Open Source Severity of Illness Score (GOSSIS) dataset involving 91,714 patient records from 147 hospitals, this study involved the development of a machine learning model for mortality using a modified LightGBM algorithm. With this, sequential and cross sectional quality monitoring was simulated and compared.</div></div><div><h3>Results</h3><div>The modified LightGBM model demonstrated superior predictive accuracy (ROC AUC of 0.88). Simulations revealed that the AI risk-adjusted CUSUM required fewer patient outcome alterations to detect atypical trends compared to standard methods.</div></div><div><h3>Conclusion</h3><div>The AI risk-adjusted CUSUM analysis represents a significant advancement in monitoring clinical outcome quality in healthcare, especially in surgery. Its ability to detect minor discrepancies in mortality rates with greater sensitivity and specificity positions it as a valuable tool for healthcare providers. This approach could lead to earlier interventions and improved patient care.</div></div>","PeriodicalId":54950,"journal":{"name":"International Journal of Medical Informatics","volume":"193 ","pages":"Article 105684"},"PeriodicalIF":3.7,"publicationDate":"2024-11-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142640309","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}