A5020949607- Electronic Health Records Systems
- Machine Learning in Healthcare
- ECG Monitoring and Analysis
- Artificial Intelligence in Healthcare and Education
- Cardiovascular Function and Risk Factors
- Radiomics and Machine Learning in Medical Imaging
- Heart Failure Treatment and Management
- Cardiac Imaging and Diagnostics
- Prostate Cancer Diagnosis and Treatment
- Medical Imaging Techniques and Applications
- Bacterial biofilms and quorum sensing
- Artificial Intelligence in Healthcare
- Prostate Cancer Treatment and Research
- Cystic Fibrosis Research Advances
- Global Cancer Incidence and Screening
- Nosocomial Infections in ICU
- Cardiac electrophysiology and arrhythmias
- Bayesian Modeling and Causal Inference
- Quality and Safety in Healthcare
- Cardiovascular Disease and Adiposity
- COVID-19 diagnosis using AI
- Topic Modeling
- Healthcare Technology and Patient Monitoring
- Traditional Chinese Medicine Studies
- Acute Myocardial Infarction Research
Stanford University
2023-2025
Stanford Medicine
2024
Cardiovascular Institute of the South
2023
Massachusetts Institute of Technology
2020
Vassar College
2020
Fred Hutch Cancer Center
2018
Abstract The electrocardiogram (ECG) is the most frequently performed cardiovascular diagnostic test, but it unclear how much information resting ECGs contain about long term risk. Here we report that a deep convolutional neural network can accurately predict long-term risk of mortality and disease based on ECG alone. Using large dataset 12-lead collected at Stanford University Medical Center, developed SEER, Estimator Electrocardiogram Risk. SEER predicts 5-year with an area under receiver...
This study seeks to develop a fully automated method of generating synthetic data from real dataset that could be employed by medical organizations distribute health researchers, reducing the need for access data. We hypothesize application Bayesian networks will improve upon predominant existing method, medBGAN, in handling complexity and dimensionality healthcare data.We learn probabilistic graphical structures simulated patient records learned structure. used University California Irvine...
Weak acids such as acetic acid and N-acetyl cysteine (NAC) at pH less than their pKa can effectively eradicate biofilms due to ability penetrate the biofilm matrix cell membrane. However, optimum conditions for activity against drug resistant strains, safety, need be understood application treat infections or inactivate on hard surfaces. Here, we investigate efficacy which weak biofilms. We compared of various mono triprotic (NAC), acid, formic citric in eradicating found that monoprotic...
0. Abstract Background The integration of large language models (LLMs) in healthcare offers immense opportunity to streamline tasks, but also carries risks such as response accuracy and bias perpetration. To address this, we conducted a red-teaming exercise assess LLMs developed dataset clinically relevant scenarios for future teams use. Methods We convened 80 multi-disciplinary experts evaluate the performance popular across multiple medical scenarios. Teams composed clinicians, engineering...
Deep learning models may combat widening racial disparities in heart failure outcomes through early identification of individuals at high risk. However, demographic biases the performance these have not been well-studied.
Red teaming, the practice of adversarially exposing unexpected or undesired model behaviors, is critical towards improving equity and accuracy large language models, but non-model creator-affiliated red teaming scant in healthcare. We convened teams clinicians, medical engineering students, technical professionals (80 participants total) to stress-test models with real-world clinical cases categorize inappropriate responses along axes safety, privacy, hallucinations/accuracy, bias. Six...
Clinicians spend large amounts of time on clinical documentation, and inefficiencies impact quality care increase clinician burnout. Despite the promise electronic medical records (EMR), transition from paper-based has been negatively associated with wellness, in part due to poor user experience, increased burden alert fatigue. In this study, we present Almanac Copilot, an autonomous agent capable assisting clinicians EMR-specific tasks such as information retrieval order placement. On...
BACKGROUND Racial disparities in prostate cancer survival (PCS) narrowed during the prostate‐specific antigen (PSA) era, suggesting that screening may induce more equitable outcomes. However, effects of lead time and overdiagnosis can inflate even without real benefit. METHODS A simulation model PCS early PSA era (1991‐2000) was created. The modeled started with baseline pre‐PSA (1975‐1990) added times using estimates from published studies. authors quantified 1) discrepancies between...
ABSTRACT Background Deep learning models may combat widening racial disparities in heart failure outcomes through early identification of individuals at high risk. However, demographic biases the performance these have not been well studied. Methods This retrospective analysis used 12-lead ECGs taken between 2008 - 2018 from 290,252 patients referred for standard clinical indications to Stanford Hospital. The primary model was a convolutional neural network trained predict incident within 5...
Clinicians spend large amounts of time on clinical documentation, and inefficiencies impact quality care increase clinician burnout. Despite the promise electronic medical records (EMR), transition from paper-based has been negatively associated with wellness, in part due to poor user experience, increased burden alert fatigue. In this study, we present Almanac Copilot, an autonomous agent capable assisting clinicians EMR-specific tasks such as information retrieval order placement. On...
Introduction: The prevalence of hypertrophic cardiomyopathy (HCM) in the UK Biobank based on ICD-10 codes (.07%) is lower than global estimates disease (0.2 - 0.5%). Prior studies using this data have remarked limitations findings given likely underdiagnosis. availability cardiac MRI scans a fraction participants offers an opportunity to identify missed diagnoses. Aims: This study seeks utilize generalizable deep learning model detect cases undiagnosed from MRIs Biobank. Methods:...
Cardiac MRI allows for a comprehensive assessment of myocardial structure, function, and tissue characteristics. Here we describe foundational vision system cardiac MRI, capable representing the breadth human cardiovascular disease health. Our deep learning model is trained via self-supervised contrastive learning, by which visual concepts in cine-sequence scans are learned from raw text accompanying radiology reports. We train evaluate our on data four large academic clinical institutions...
Introduction: Our goal was to create a generalizable Cardiac MRI (CMR) deep learning system capable of understanding the breadth human disease and health. With traditional supervised approaches, parameters learned for one clinical problem rarely generalize others. Models must be re-trained from scratch each new task interest, requiring thousands training examples every time. Unlike clinicians, models lack baseline fund knowledge over which specific tasks can accelerated. Methods: We use...