Share This Author
Democratizing EHR analyses with FIDDLE: a flexible data-driven preprocessing pipeline for structured clinical data
- Shengpu Tang, Parmida Davarmanesh, Yanmeng Song, Danai Koutra, M. Sjoding, J. Wiens
- Computer ScienceJ. Am. Medical Informatics Assoc.
- 11 October 2020
FIDDLE, an open-source preprocessing pipeline that streamlines the preprocessing of data extracted from the EHR, facilitates applying ML to structured EHR data and is generalizable across different prediction times, ML algorithms, and data sets.
Model Selection for Offline Reinforcement Learning: Practical Considerations for Healthcare Settings
This work investigates a model selection pipeline for offline RL that relies on off-policy evaluation (OPE) as a proxy for validation performance and proposes a simple two-stage approach to accelerate model selection by avoiding potentially unnecessary computation.
Evaluating a Widely Implemented Proprietary Deterioration Index Model among Hospitalized Patients with COVID-19
- Karandeep Singh, T. Valley, B. Nallamothu
- MedicineAnnals of the American Thoracic Society
- 24 December 2020
The EDI identifies small subsets of high-risk and low-risk patients with COVID-19 with good discrimination, although its clinical use as an early warning system is limited by low sensitivity.
Predicting Acute Graft-Versus-Host Disease Using Machine Learning and Longitudinal Vital Sign Data From Electronic Health Records
- Shengpu Tang, Grant Chappell, Amanda Mazzoli, M. Tewari, S. Choi, J. Wiens
- MedicineJCO Clinical Cancer Informatics
- 1 February 2020
Leveraging readily available clinical data from EHRs, a machine-learning model for aGVHD prediction in patients undergoing HCT is developed and continuous monitoring of vital signs could potentially help clinicians more accurately identify patients at high risk for aGFD.
Clinician-in-the-Loop Decision Making: Reinforcement Learning with Near-Optimal Set-Valued Policies
This work proposes a model-free algorithm based on temporal difference learning and a near-greedy heuristic for action selection that exhibits good convergence properties and discovers meaningful near-equivalent actions in healthcare settings.
Validating a Widely Implemented Deterioration Index Model Among Hospitalized COVID-19 Patients
Introduction: The coronavirus disease 2019 (COVID-19) pandemic is straining the capacity of U.S. healthcare systems. Accurately identifying subgroups of hospitalized COVID-19 patients at high- and…
Predicting postoperative opioid use with machine learning and insurance claims in opioid-naïve patients.
Early identification of patients admitted to hospital for covid-19 at risk of clinical deterioration: model development and multisite external validation study
A model to predict clinical deterioration was developed rapidly in response to the covid-19 pandemic at a single hospital, was applied externally without the sharing of data, and performed well across multiple medical centers, patient subgroups, and time periods, showing its potential as a tool for use in optimizing healthcare resources.
Respecting Autonomy And Enabling Diversity: The Effect Of Eligibility And Enrollment On Research Data Demographics.
Compared with the overall clinical population, patients who consented to enroll in the research data bank were significantly less diverse in terms of age, sex, race, ethnicity, and socioeconomic status.
Relaxed Parameter Sharing: Effectively Modeling Time-Varying Relationships in Clinical Time-Series
- Jeeheh Oh, Jiaxuan Wang, Shengpu Tang, M. Sjoding, J. Wiens
- Computer ScienceMachine Learning in Health Care
A novel RNN formulation based on a mixture model in which relaxed parameter sharing over time is proposed, which outperforms standard LSTMs and other state-of-the-art baselines across all tasks and can lead to improved patient risk stratification performance.