Writings submitted and/or under review that benefit from published addenda are listed in our branch "under-review" for a modicum of impartiality/reviewer blindness. Click through the link to see these listings.
Performance Trade-Offs for the Safe Sharing of a Fully De-Identified SVM Model for Child Abuse and Neglect
Paul M. Heider, PhD, Aviv Y. Landau, PhD, MSW, Shahad Althobaiti, MS, Leslie A. Lenert, MD, Rochelle F. Hanson, PhD
An Extensible Evaluation Framework Applied to Clinical Text Deidentification Natural Language Processing Tools: Multisystem and Multicorpus Study
Paul M. Heider, Stephane M. Meystre
J Med Internet Res 2024;26:e55676. doi: 10.2196/55676. PMID: 38805692
Algorithmic Bias in De-Identification Tools
Paul M. Heider
Significance Testing and Reporting are Critically Underutilized Tools for Clinical NLP
Paul M. Heider, PhD, Glenn T. Gobbel, DVM, PhD, MS, Jihad S. Obeid, MD, Alexander V. Alekseyenko, PhD, FAMIA
Implicit Provider Bias as Assessed through Explicit Mentions of Pejorative and Laudative Terms in MIMIC-III
Paul M. Heider, PhD, Jihad S. Obeid, MD, Leslie A. Lenert, MD
Post-Hoc Ensemble Generation for Clinical NLP: A Study of Concept Recognition, Normalization, and Context Attributes
Paul M. Heider, PhD, Kexin Chen, MPH, Ronak Pipaliya, BS, Stephane M. Meystre, MD, PhD
Natural Language Processing Enabling COVID-19 Predictive Analytics to Support Data-Driven Patient Advising and Pooled Testing
Stéphane M Meystre, MD, PhD, Paul M Heider, PhD, Youngjun Kim, PhD, Matthew Davis, Jihad Obeid, MD, James Madory, DO, Alexander V Alekseyenko, PhD
- doi: 10.1093/jamia/ocab186
- PMID: 34415311.
@article{10.1093/jamia/ocab186,
author = {Meystre, Stéphane M and Heider, Paul M and Kim, Youngjun and Davis, Matthew and Obeid, Jihad and Madory, James and Alekseyenko, Alexander V},
title = "{Natural Language Processing Enabling COVID-19 Predictive Analytics to Support Data-Driven Patient Advising and Pooled Testing}",
journal = {Journal of the American Medical Informatics Association},
year = {2021},
month = {08},
abstract = "{The COVID-19 pandemic response at MUSC included virtual care visits for patients with suspected SARS-CoV-2 infection. The telehealth system used for these visits only exports a text note to integrate with the EHR, but structured and coded information about COVID-19 (e.g., exposure, risk factors, symptoms) was needed to support clinical care and early research as well as predictive analytics for data-driven patient advising and pooled testing.To capture COVID-19 information from multiple sources, a new data mart and a new Natural Language Processing (NLP) application prototype were developed. The NLP application combined reused components with dictionaries and rules crafted by domain experts. It was deployed as a web service for hourly processing of new data from patients assessed or treated for COVID-19. The extracted information was then used to develop algorithms predicting SARS-CoV-2 diagnostic test results based on symptoms and exposure information.The dedicated data mart and NLP application were developed and deployed in a mere 10-day sprint in March 2020. The NLP application was evaluated with good accuracy (85.8\\% recall and 81.5\\% precision). The SARS-CoV-2 testing predictive analytics algorithms were configured to provide patients with data-driven COVID-19 testing advices with a sensitivity of 81-92\\% and to enable pooled testing with a negative predictive value of 90-91\\% reducing the required tests to about 63\\%.SARS-CoV-2 testing predictive analytics and NLP successfully enabled data-driven patient advising and pooled testing.}",
issn = {1527-974X},
doi = {10.1093/jamia/ocab186},
url = {https://doi.org/10.1093/jamia/ocab186},
note = {ocab186},
eprint = {https://academic.oup.com/jamia/advance-article-pdf/doi/10.1093/jamia/ocab186/39829927/ocab186.pdf},
}