Article
Emergency Medicine
Theodore W. Heyming, Chloe Knudsen-Robbins, William Feaster, Louis Ehwerhemuepha
Summary: The study developed a machine learning model to predict the disposition of pediatric emergency department patients based on triage assessment and historical information. The model demonstrated high accuracy in training and testing, providing an effective tool for healthcare providers to stratify patients effectively.
AMERICAN JOURNAL OF EMERGENCY MEDICINE
(2021)
Article
Medicine, General & Internal
Dinesh R. Pai, Balaraman Rajan, Puneet Jairath, Stephen M. Rosito
Summary: This study aims to predict the need for hospitalization of adult patients presenting to the emergency department for fall-related fractures using machine learning models. The results show that neural network models perform the best in predicting hospital admissions.
INTERNAL AND EMERGENCY MEDICINE
(2023)
Article
Computer Science, Information Systems
Vlada Rozova, Katrina Witt, Jo Robinson, Yan Li, Karin Verspoor
Summary: Accurate identification of self-harm presentations to Emergency Departments through a machine learning-based NLP model can provide timely mental health support and aid in understanding suicidal intent burden. The best-performing model achieved 90% Precision and 90% Recall on blind test data, demonstrating the practical value of NLP in identifying patients for mental health follow-up and supporting suicide prevention efforts.
JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION
(2022)
Article
Computer Science, Information Systems
Abdulaziz Ahmed, Mohammed Al-Maamari, Mohammad Firouz, Dursun Delen
Summary: In this paper, metaheuristic optimization algorithms based on simulated annealing and adaptive simulated annealing are proposed to optimize the parameters of extreme gradient boosting and categorical boosting algorithms. The optimized model is used to develop an e-triage tool for predicting the emergency severity index of emergency department patients.
INFORMATION SYSTEMS FRONTIERS
(2023)
Article
Chemistry, Multidisciplinary
Jordi Cusido, Joan Comalrena, Hamidreza Alavi, Laia Llunas
Summary: This study aims to develop an accurate prediction model for emergency department admissions to improve patient care quality and reduce emergency room overcrowding. Using data analysis techniques to predict whether patients need to be admitted, providing decision support for emergency departments.
APPLIED SCIENCES-BASEL
(2022)
Article
Biochemistry & Molecular Biology
Hsiao-Yun Chao, Chin-Chieh Wu, Avichandra Singh, Andrew Shedd, Jon Wolfshohl, Eric H. Chou, Yhu-Chering Huang, Kuan-Fu Chen
Summary: This study developed and validated a prediction model for 28-day mortality in patients with infection using machine learning algorithms. The results showed that the random forest model outperformed other models and logistic regression in predicting mortality.
Article
Endocrinology & Metabolism
Jing Qi, Jingchao Lei, Nanyi Li, Dan Huang, Huaizheng Liu, Kefu Zhou, Zheren Dai, Chuanzheng Sun
Summary: A study developed predictive models using various machine learning algorithms and rigorous data processing methods for predicting clinical outcomes of sepsis patients. The models were validated on internal and external validation sets, showing good performance. Key features for predicting outcomes in sepsis patients include age, albumin, and lactate.
FRONTIERS IN ENDOCRINOLOGY
(2022)
Article
Multidisciplinary Sciences
Andreea Vantu, Anca Vasilescu, Alexandra Baicoianu
Summary: Artificial intelligence has shown its ability to overcome challenges in daily life. The development of AI has led to more studies on machine learning solutions, including healthcare. The availability of medical records provides opportunities to explore machine learning models and their ability to process large amounts of data to solve medical problems. This study focuses on the correlation between medical records and diagnosis, particularly in the emergency department triage process.
Article
Computer Science, Interdisciplinary Applications
Zhengyu Jiang, Lulong Bo, Zhenhua Xu, Yubing Song, Jiafeng Wang, Pingshan Wen, Xiaojian Wan, Tao Yang, Xiaoming Deng, Jinjun Bian
Summary: The study analyzed specific risk factors associated with in-hospital mortality of sepsis survivors during ICU readmission using machine learning algorithm, highlighting the importance of advanced ML techniques in developing predictive models for critical care patients.
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2021)
Article
Cardiac & Cardiovascular Systems
Shinichi Goto, Divyarajsinhji Solanki, Jenine E. John, Ryuichiro Yagi, Max Homilius, Genki Ichihara, Yoshinori Katsumata, Hanna K. Gaggin, Yuji Itabashi, Calum A. MacRae, Rahul C. Deo
Summary: This study developed machine learning models using ECGs and echocardiograms to detect and differentiate hypertrophic cardiomyopathy (HCM) from other cardiac conditions. The models showed good generalizability across multiple institutions when trained in a federated manner, improving the accuracy of HCM detection.
Article
Emergency Medicine
Dong Hyun Choi, Ki Jeong Hong, Jeong Ho Park, Sang Do Shin, Young Sun Ro, Kyoung Jun Song, Ki Hong Kim, Sungwan Kim
Summary: This study developed and validated machine learning models to predict bacteremia in the emergency department during triage and disposition stages. The Triage XGB model could be used to identify patients with a low risk of bacteremia immediately after initial ED triage, while the Disposition XGB model showed excellent discriminative performance.
AMERICAN JOURNAL OF EMERGENCY MEDICINE
(2022)
Article
Emergency Medicine
Ian Ward A. Maia, Lucas Oliveira J. E. Silva, Henrique Herpich, Luciano Diogo, Joao Carlos Batista Santana, Daniel Pedrollo, Mario Castro Alvarez Perez, Rafael Nicolaidis
Summary: The qSOFA score showed low sensitivity in predicting in-hospital mortality among patients with suspected infection, indicating that it should not be used as a screening tool for ruling out infection in the initial evaluation by emergency physicians or ED triage nurses in low-to-middle income countries.
AMERICAN JOURNAL OF EMERGENCY MEDICINE
(2021)
Article
Medicine, General & Internal
Dhavalkumar Patel, Satya Narayan Cheetirala, Ganesh Raut, Jules Tamegue, Arash Kia, Benjamin Glicksberg, Robert Freeman, Matthew A. Levin, Prem Timsina, Eyal Klang
Summary: In this study, an inclusive gradient boosting model was used to predict hospital admission from the emergency department at different time points. The results showed that the performance of the full model was comparable to the single models at all time points. Therefore, an ML-based prediction model can be used for identifying hospital admission.
JOURNAL OF CLINICAL MEDICINE
(2022)
Article
Medicine, Research & Experimental
Behrad Barghi, Nasibeh Azadeh-Fard
Summary: Sepsis is a systemic inflammatory response to infection and remains a critical problem. This study aimed to predict the risk of sepsis using machine learning methods, and found that the Bootstrap Forest performed the best in predicting sepsis risk.
EUROPEAN JOURNAL OF MEDICAL RESEARCH
(2022)
Article
Computer Science, Information Systems
Huilin Jiang, Haifeng Mao, Huimin Lu, Peiyi Lin, Wei Garry, Huijing Lu, Guangqian Yang, Timothy H. Rainer, Xiaohui Chen
Summary: This study compared the performance of four common machine learning models in assisting decision making of triage levels for patients with suspected cardiovascular disease at the emergency department. XGBoost demonstrated a slight advantage over other models, with blood pressure, pulse rate, oxygen saturation, and age being identified as the most significant variables for triage decisions. The models could be used for differential triage of low-risk and high-risk patients to improve efficiency and resource allocation.
INTERNATIONAL JOURNAL OF MEDICAL INFORMATICS
(2021)
Review
Endocrinology & Metabolism
Kuo Ren Tan, Jun Jie Benjamin Seng, Yu Heng Kwan, Ying Jie Chen, Sueziani Binte Zainudin, Dionne Hui Fang Loh, Nan Liu, Lian Leng Low
Summary: This study evaluated the quality and performance of machine learning models in predicting microvascular and macrovascular diabetes complications in a Type 2 diabetes population. The results showed that neural networks were the most frequently used models, and age, duration of diabetes, and body mass index were common predictors. Most models demonstrated good discrimination performance, with random forest showing the best overall performance. However, further validation studies are needed before the clinical implementation of these models.
JOURNAL OF DIABETES SCIENCE AND TECHNOLOGY
(2023)
Article
Emergency Medicine
Andrew Fu Wah Ho, Priscilla Zi Yin Ting, Jamie Sin Ying Ho, Stephanie Fook-Chong, Nur Shahidah, Pin Pin Pek, Nan Liu, Seth Teoh, Ching-Hui Sia, Daniel Yan Zheng Lim, Shir Lynn Lim, Ting Hway Wong, Marcus Eng Hock Ong
Summary: By studying the relationship between socioeconomic status and bystander cardiopulmonary resuscitation (CPR) in cases of out-of-hospital cardiac arrest (OHCA), it was found that lower building-level socioeconomic status is associated with a lower rate of bystander CPR. Females are more susceptible to the effect of low socioeconomic status on bystander CPR rates.
PREHOSPITAL EMERGENCY CARE
(2023)
Article
Mathematics, Interdisciplinary Applications
Madhurima Panja, Tanujit Chakraborty, Sk Shahid Nadim, Indrajit Ghosh, Uttam Kumar, Nan Liu
Summary: Dengue fever is a widespread virulent disease that affects millions of people globally and puts a strain on healthcare systems. Due to the lack of specific drugs and vaccines, policymakers rely on early warning systems for intervention decisions. However, existing forecasting models often provide unstable and unreliable forecasts. This study proposes a new model called XEWNet that incorporates wavelet transformation into an ensemble neural network framework to improve the accuracy and reliability of dengue outbreak predictions.
CHAOS SOLITONS & FRACTALS
(2023)
Article
Critical Care Medicine
Ingvild B. M. Tjelmeland, Jan Wnent, Siobhan Masterson, Jo Kramer-Johansen, Marcus Eng Hock Ong, Karen Smith, Eirik Skogvol, Rolf Lefering, Shir Lynn Lim, Nan Liu, Bridget Dicker, Andrew Swain, Stephen Ball, Jan-Thorsten Grasner
Summary: This study investigated the impact of the Covid-19 lockdown on bystander cardiopulmonary resuscitation (CPR) provision using population-based registries. The results showed a steady increase in bystander CPR from 2017 to 2020, and the lockdown did not affect this trend. There were variations in the incidence of bystander CPR among different registries.
Article
Medicine, General & Internal
Zhaoran Wang, Gilbert Lim, Wei Yan Ng, Tien-En Tan, Jane Lim, Sing Hui Lim, Valencia Foo, Joshua Lim, Laura Gutierrez Sinisterra, Feihui Zheng, Nan Liu, Gavin Siew Wei Tan, Ching-Yu Cheng, Gemmy Chui Ming Cheung, Tien Yin Wong, Daniel Shu Wei Ting
Summary: Age-related macular degeneration (AMD) is a leading cause of vision impairment worldwide and early detection is crucial. Deep learning systems have shown potential for detecting eye diseases, but require large datasets. This study aims to develop synthetic fundus photos of AMD lesions using GANs and assess their realness using an objective scale.
FRONTIERS IN MEDICINE
(2023)
Article
Health Care Sciences & Services
Jae Yong Yu, Sejin Heo, Feng Xie, Nan Liu, Sun Yung Yoon, Han Sol Chang, Taerim Kim, Se Uk Lee, Marcus Eng Hock Ong, Yih Yng Ng, Sang Do Shin, Kentaro Kajino, Won Chul Cha
Summary: This study aimed to develop and validate an interpretable field triage scoring system based on a multinational trauma registry in Asia. Age and vital sign were found to be significant variables for predicting mortality. External validation showed that the model had an accuracy of 0.756-0.850.
LANCET REGIONAL HEALTH-WESTERN PACIFIC
(2023)
Article
Biochemical Research Methods
Feng Xie, Yilin Ning, Mingxuan Liu, Siqi Li, Seyed Ehsan Saffari, Han Yuan, Victor Volovici, Daniel Shu Wei Ting, Benjamin Alan Goldstein, Marcus Eng Hock Ong, Roger Vaughan, Bibhas Chakraborty, Nan Liu
Summary: The AutoScore framework automates the generation of data-driven clinical scores for various clinical applications. This article presents a protocol using the open-source AutoScore package to develop clinical scoring systems for binary, survival, and ordinal outcomes. The protocol includes steps for package installation, data processing and checking, variable ranking, variable selection, score generation, fine-tuning, and evaluation.
Review
Emergency Medicine
Sze Ling Chan, Jin Wee Lee, Marcus Eng Hock Ong, Fahad Javaid Siddiqui, Nicholas Graves, Andrew Fu Wah Ho, Nan Liu
Summary: This scoping review examines the implementation of prediction models in the emergency department (ED) and provides insights on contributing factors and outcomes from an implementation science perspective. The most common prediction models implemented in the ED are early warning scores, with implementation strategies involving training stakeholders, infrastructural changes, and evaluative or iterative strategies. Key determinants of successful implementation include stakeholder engagement, codevelopment of workflows and implementation strategies, education, and usability.
ANNALS OF EMERGENCY MEDICINE
(2023)
Review
Computer Science, Information Systems
Pinyan Liu, Ziwen Wang, Nan Liu, Marco Aurelio Peres
Summary: Data-driven population segmentation is widely used in clinical settings to divide heterogeneous populations into relatively homogeneous groups. Machine learning algorithms have shown potential in improving algorithm development across various phenotypes and healthcare situations. This study evaluates the application of machine learning-based segmentation in different populations, segmentation details, and outcome evaluations.
JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION
(2023)
Article
Computer Science, Artificial Intelligence
Madhurima Panja, Tanujit Chakraborty, Uttam Kumar, Nan Liu
Summary: Infectious diseases continue to be a major cause of illness and death worldwide, with epidemic waves of infection. The lack of specific drugs and vaccines exacerbates the situation, leading to a reliance on accurate and reliable epidemic forecasters. The proposed Ensemble Wavelet Neural Network (EWNet) model effectively characterizes the non-stationary behavior and seasonal dependencies of epidemic time series, improving the accuracy of epidemic forecasting compared to other methods.
Article
Multidisciplinary Sciences
Shu-Ling Chong, Chenglin Niu, Gene Yong-Kwang Ong, Rupini Piragasam, Zi Xean Khoo, Zhi Xiong Koh, Dagang Guo, Jan Hau Lee, Marcus Eng Hock Ong, Nan Liu
Summary: The study aimed to develop a risk scoring system (FIRST and FIRST +) to quantify the risk of serious bacterial infections. The two models derived from machine learning and logistic regression can assist in triaging and prioritizing febrile infants.
SCIENTIFIC REPORTS
(2023)
Article
Multidisciplinary Sciences
Yan Gao, Nicholas Yock Teck Soh, Nan Liu, Gilbert Lim, Daniel Ting, Lionel Tim-Ee Cheng, Kang Min, Charlene Liew, Hong Choon Oh, Jin Rong Tan, Narayan Venkataraman, Siang Hiong Goh, Yet Yen Yan
Summary: This paper describes the development of a deep learning model for predicting hip fractures on pelvic radiographs. The model achieved high sensitivity and specificity when tested on emergency department radiographs, including those with suboptimal image quality, non-hip fractures, and metallic implants. The study also investigated the impact of ethnicity on model performance and the accuracy of a visualization algorithm for fracture localization.
Article
Public, Environmental & Occupational Health
Jamie S. Y. Ho, Andrew F. W. Ho, Eric Jou, Nan Liu, Huili Zheng, Joel Aik
Summary: The study found that the implementation of smoke-free laws in outdoor and residential areas in Singapore was associated with a decrease in AMIs. People above the age of 65 and men appeared to benefit the most.
Article
Health Care Sciences & Services
Mingxuan Liu, Yilin Ning, Salinelat Teixayavong, Mayli Mertens, Jie Xu, Daniel Shu Wei Ting, Lionel Tim-Ee Cheng, Jasmine Chiat Ling Ong, Zhen Ling Teo, Ting Fang Tan, Narrendar Ravichandran, Fei Wang, Leo Anthony Celi, Marcus Eng Hock Ong, Nan Liu
Summary: Artificial intelligence has shown its ability to extract insights from data, but ensuring fairness in high-stakes fields such as healthcare remains a concern. The notion of fairness in clinical contexts requires careful examination and alignment with ethical considerations. A multidisciplinary approach involving AI researchers, clinicians, and ethicists is necessary to bridge the gap between technical developments and clinical needs.
NPJ DIGITAL MEDICINE
(2023)
Article
Critical Care Medicine
Yohei Okada, Mayli Mertens, Nan Liu, Sean Shao Wei Lam, Marcus Eng Hock Ong
Summary: Artificial intelligence (AI) and machine learning (ML) have gained attention in the medical field, but keeping up with the latest research is challenging for clinicians. This article aims to translate research concepts and concerns to healthcare professionals interested in applying AI and ML to resuscitation research.
RESUSCITATION PLUS
(2023)