Mary Saltz

Mary Saltz

Auto-generated profile, based on publicly available information

Mary Saltz

Department of Biomedical Informatics, Stony Brook University, New York, NY, USA

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
17 December 2020 in JMIR Medical Informatics
Verified
Journal Article
Open access
SRSina Rashidian
Sina Rashidian
KAKayley Abell-Hart
Kayley Abell-Hart
JHJanos Hajagos
Janos Hajagos
RMRichard Moffitt
Richard Moffitt
VLVeena Lingam
Veena Lingam
MSMary Saltz
Mary Saltz
Background: Diabetes affects more than 30 million patients across the United States. With such a large disease burden, even a small error in classification can be significant. Currently billing codes, assigned at the time of a medical encounter, are the “gold standard” reflecting the actual diseases present in an individual, and thus in aggregate reflect disease prevalence in the population. These codes are generated by highly trained coders and by health care providers but are not always accurate. Objective: This work provides a scalable deep learning methodology to more accurately classify individuals with diabetes across multiple health care systems. Methods: We leveraged a long short-term memory-dense neural network (LSTM-DNN) model to identify patients with or without diabetes using data from 5 acute care facilities with 187,187 patients and 275,407 encounters, incorporating data elements including laboratory test results, diagnostic/procedure codes, medications, demographic data, and admission information. Furthermore, a blinded physician panel reviewed discordant cases, providing an estimate of the total impact on the population. Results: When predicting the documented diagnosis of diabetes, our model achieved an 84% F1 score, 96% area under the curve–receiver operating characteristic curve, and 91% average precision on a heterogeneous data set from 5 distinct health facilities. However, in 81% of cases where the model disagreed with the documented phenotype, a blinded physician panel agreed with the model. Taken together, this suggests that 4.3% of our studied population have either missing or improper diabetes diagnosis. Conclusions: This study demonstrates that deep learning methods can improve clinical phenotyping even when patient data are noisy, sparse, and heterogeneous.

14 November 2023 in Journal of Clinical and Translational Science
Verified
Journal Article
Open access
PDF file
CSChristine Suver
Christine Suver
JHJeremy Harper
Jeremy Harper
JLJohanna Loomba
Johanna Loomba
MSMary Saltz
Mary Saltz
JSJulian Solway
Julian Solway
MHMelissa Haendel
Melissa Haendel
The National COVID Cohort Collaborative (N3C) is a public–private–government partnership established during the Coronavirus pandemic to create a centralized data resource called the “N3C data enclave.” This resource contains individual-level health data from participating healthcare sites nationwide to support rapid collaborative analytics. N3C has enabled analytics within a cloud-based enclave of data from electronic health records from over 17 million people (with and without COVID-19) in the USA. To achieve this goal of a shared data resource, N3C implemented a shared governance strategy involving stakeholders in decision-making. The approach leveraged best practices in data stewardship and team science to rapidly enable COVID-19-related research at scale while respecting the privacy of data subjects and participating institutions. N3C balanced equitable access to data, team-based scientific productivity, and individual professional recognition – a key incentive for academic researchers. This governance approach makes N3C research sustainable and effective beyond the initial days of the pandemic. N3C demonstrated that shared governance can overcome traditional barriers to data sharing without compromising data security and trust. The governance innovations described herein are a helpful framework for other privacy-preserving data infrastructure programs and provide a working model for effective team science beyond COVID-19.

19 July 2020 in JMIR Preprints
Verified
Preprint
Open access
SRSina Rashidian
Sina Rashidian
KAKayley Abell-Hart
Kayley Abell-Hart
JHJanos Hajagos
Janos Hajagos
RMRichard Moffitt
Richard Moffitt
VLVeena Lingam
Veena Lingam
MSMary Saltz
Mary Saltz
Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet #Preprint #PeerReviewMe: Warning: This is a unreviewed preprint. Readers are warned that the document has not been peer-reviewed by expert/patient reviewers or an academic editor, may contain misleading claims, and is likely to undergo changes before final publication, if accepted, or may have been rejected/withdrawn. Readers with interest and expertise are encouraged to sign up as peer-reviewer, if the paper is within an open peer-review period. Please cite this preprint only for review purposes or for grant applications and CVs (if you are the author). Background: Diabetes affects more than 30 million patients across the United States. With such a large disease burden, even a small error in classification can be significant. Currently billing codes, assigned at the time of a medical encounter, are the “gold standard” reflecting the actual diseases present in an individual, and thus in aggregate reflect disease prevalence in the population. These codes are generated by highly trained coders and by health care providers but are not always accurate. Objective: This work provides a scalable deep learning methodology to more accurately classify individuals with diabetes across multiple health care systems. Methods: We leveraged a long short-term memory-dense neural network (LSTM-DNN) model to identify patients with or without diabetes using data from 5 acute care facilities with 187,187 patients and 275,407 encounters, incorporating data elements including laboratory test results, diagnostic/procedure codes, medications, demographic data, and admission information. Furthermore, a blinded physician panel reviewed discordant cases, providing an estimate of the total impact on the population. Results: When predicting the documented diagnosis of diabetes, our model achieved an 84% F1 score, 96% area under the curve–receiver operating characteristic curve, and 91% average precision on a heterogeneous data set from 5 distinct health facilities. However, in 81% of cases where the model disagreed with the documented phenotype, a blinded physician panel agreed with the model. Taken together, this suggests that 4.3% of our studied population have either missing or improper diabetes diagnosis. Conclusions: This study demonstrates that deep learning methods can improve clinical phenotyping even when patient data are noisy, sparse, and heterogeneous.