A5012198181- Machine Learning in Healthcare
- Artificial Intelligence in Healthcare
- Chronic Disease Management Strategies
- Explainable Artificial Intelligence (XAI)
- Heart Failure Treatment and Management
- Renal Transplantation Outcomes and Treatments
- Pharmacology and Obesity Treatment
- Artificial Intelligence in Healthcare and Education
- Neonatal Respiratory Health Research
- Infant Development and Preterm Care
- Neonatal and fetal brain pathology
- Topic Modeling
- Bioinformatics and Genomic Networks
- Complementary and Alternative Medicine Studies
- Birth, Development, and Health
- Gene expression and cancer classification
- Single-cell and spatial transcriptomics
- Transplantation: Methods and Outcomes
- Dietetics, Nutrition, and Education
- Pregnancy and preeclampsia studies
- Obesity and Health Practices
- Reproductive System and Pregnancy
- Consumer Attitudes and Food Labeling
- Preterm Birth and Chorioamnionitis
- Adversarial Robustness in Machine Learning
Stanford University
2022-2024
Palo Alto University
2024
Stanford Medicine
2022
Florida State University
2018-2021
Abstract Objective To conduct a systematic scoping review of explainable artificial intelligence (XAI) models that use real-world electronic health record data, categorize these techniques according to different biomedical applications, identify gaps current studies, and suggest future research directions. Materials Methods We searched MEDLINE, IEEE Xplore, the Association for Computing Machinery (ACM) Digital Library relevant papers published between January 1, 2009 May 2019. summarized...
Although prematurity is the single largest cause of death in children under 5 years age, current definition prematurity, based on gestational lacks precision needed for guiding care decisions. Here, we propose a longitudinal risk assessment adverse neonatal outcomes newborns deep learning model that uses electronic health records (EHRs) to predict wide range over period starting shortly before conception and ending months after birth. By linking EHRs Lucile Packard Children’s Hospital...
Abstract Advanced measurement and data storage technologies have enabled high-dimensional profiling of complex biological systems. For this, modern multiomics studies regularly produce datasets with hundreds thousands measurements per sample, enabling a new era precision medicine. Correlation analysis is an important first step to gain deeper insights into the coordination underlying processes such However, construction large correlation networks in remains major computational challenge...
Abstract Objectives Prediction of post-transplant health outcomes and identification key factors remain important issues for pediatric transplant teams researchers. Outcomes research has generally relied on general linear modeling or similar techniques offering limited predictive validity. Thus far, data-driven machine learning (ML) approaches have had application success in research. The purpose the current study was to examine ML models predicting hospitalization a sample kidney, liver,...
Heart disease remains the leading cause of death in United States. Compared with risk assessment guidelines that require manual calculation scores, machine learning-based prediction for outcomes such as mortality can be utilized to save time and improve accuracy. This study built evaluated various learning models predict one-year patients diagnosed acute myocardial infarction or post syndrome MIMIC-III database. The results best performing shallow were compared a deep feedforward neural...
Abstract Acute myocardial infarction poses significant health risks and financial burden on healthcare families. Prediction of mortality risk among AMI patients using rich electronic record (EHR) data can potentially save lives costs. Nevertheless, EHR-based prediction models usually use a missing imputation method without considering its impact the performance interpretability model, hampering real-world applicability in setting. This study examines different methods for imputing values EHR...
Predicting the risk of mortality for patients with acute myocardial infarction (AMI) using electronic health records (EHRs) data can help identify risky who might need more tailored care. In our previous work, we built computational models to predict one-year admitted an intensive care unit (ICU) AMI or post syndrome. Our prior work only used structured clinical from MIMIC-III, a publicly available ICU database. this study, enhanced by adding word embedding features free-text discharge...
Deep neural networks have achieved remarkable success in various challenging tasks. However, the black-box nature of such is not acceptable to critical applications, as healthcare. In particular, existence adversarial examples and their overgeneralization irrelevant, out-ofdistribution inputs with high confidence makes it difficult, if impossible, explain decisions by networks. this paper, we analyze underlying mechanism generalization deep propose an (n, k) consensus algorithm which...
Predicting the risk of mortality for patients with acute myocardial infarction (AMI) using electronic health records (EHRs) data can help identify risky who might need more tailored care. In our previous work, we built computational models to predict one-year admitted an intensive care unit (ICU) AMI or post syndrome. Our prior work only used structured clinical from MIMIC-III, a publicly available ICU database. this study, enhanced by adding word embedding features free-text discharge...
Psychosocial and stress-related factors (PSFs), defined as internal or external stimuli that induce biological changes, are potentially modifiable accessible targets for interventions associated with adverse pregnancy outcomes (APOs). Although individual APOs have been shown to be connected PSFs, they biologically interconnected, relatively infrequent, therefore challenging model. In this context, multi-task machine learning (MML) is an ideal tool exploring the interconnectedness of on one...
Deep neural networks have achieved remarkable success in various challenging tasks. However, the black-box nature of such is not acceptable to critical applications, as healthcare. In particular, existence adversarial examples and their overgeneralization irrelevant, out-of-distribution inputs with high confidence makes it difficult, if impossible, explain decisions by networks. this paper, we analyze underlying mechanism generalization deep propose an ($n$, $k$) consensus algorithm which...
Abstract While prematurity is the single largest cause of death in children under 5 years age, current definition prematurity, based on gestational lacks precision needed for guiding care decisions. Here we propose a longitudinal risk assessment adverse neonatal outcomes newborns multi-task deep learning model that uses electronic health records (EHRs) to predict wide range over period starting shortly after time conception and ending months birth. By linking EHRs Lucile Packard Children’s...