eLife (Jul 2020)
Augmented curation of clinical notes from a massive EHR system reveals symptoms of impending COVID-19 diagnosis
- Tyler Wagner,
- FNU Shweta,
- Karthik Murugadoss,
- Samir Awasthi,
- AJ Venkatakrishnan,
- Sairam Bade,
- Arjun Puranik,
- Martin Kang,
- Brian W Pickering,
- John C O'Horo,
- Philippe R Bauer,
- Raymund R Razonable,
- Paschalis Vergidis,
- Zelalem Temesgen,
- Stacey Rizza,
- Maryam Mahmood,
- Walter R Wilson,
- Douglas Challener,
- Praveen Anand,
- Matt Liebers,
- Zainab Doctor,
- Eli Silvert,
- Hugo Solomon,
- Akash Anand,
- Rakesh Barve,
- Gregory Gores,
- Amy W Williams,
- William G Morice II,
- John Halamka,
- Andrew Badley,
- Venky Soundararajan
Affiliations
- Tyler Wagner
- nference, Cambridge, United States
- FNU Shweta
- ORCiD
- Mayo Clinic, Rochester, United States
- Karthik Murugadoss
- nference, Cambridge, United States
- Samir Awasthi
- nference, Cambridge, United States
- AJ Venkatakrishnan
- ORCiD
- nference, Cambridge, United States
- Sairam Bade
- nference Labs, Bangalore, India
- Arjun Puranik
- nference, Cambridge, United States
- Martin Kang
- nference, Cambridge, United States
- Brian W Pickering
- Mayo Clinic, Rochester, United States
- John C O'Horo
- Mayo Clinic, Rochester, United States
- Philippe R Bauer
- Mayo Clinic, Rochester, United States
- Raymund R Razonable
- Mayo Clinic, Rochester, United States
- Paschalis Vergidis
- Mayo Clinic, Rochester, United States
- Zelalem Temesgen
- Mayo Clinic, Rochester, United States
- Stacey Rizza
- Mayo Clinic, Rochester, United States
- Maryam Mahmood
- Mayo Clinic, Rochester, United States
- Walter R Wilson
- Mayo Clinic, Rochester, United States
- Douglas Challener
- ORCiD
- Mayo Clinic, Rochester, United States
- Praveen Anand
- ORCiD
- nference Labs, Bangalore, India
- Matt Liebers
- nference, Cambridge, United States
- Zainab Doctor
- nference, Cambridge, United States
- Eli Silvert
- nference, Cambridge, United States
- Hugo Solomon
- nference, Cambridge, United States
- Akash Anand
- nference Labs, Bangalore, India
- Rakesh Barve
- nference Labs, Bangalore, India
- Gregory Gores
- Mayo Clinic, Rochester, United States
- Amy W Williams
- Mayo Clinic, Rochester, United States
- William G Morice II
- Mayo Clinic, Rochester, United States; Mayo Clinic Laboratories, Rochester, United States
- John Halamka
- Mayo Clinic, Rochester, United States
- Andrew Badley
- Mayo Clinic, Rochester, United States
- Venky Soundararajan
- ORCiD
- nference, Cambridge, United States
- DOI
- https://doi.org/10.7554/eLife.58227
- Journal volume & issue
-
Vol. 9
Abstract
Understanding temporal dynamics of COVID-19 symptoms could provide fine-grained resolution to guide clinical decision-making. Here, we use deep neural networks over an institution-wide platform for the augmented curation of clinical notes from 77,167 patients subjected to COVID-19 PCR testing. By contrasting Electronic Health Record (EHR)-derived symptoms of COVID-19-positive (COVIDpos; n = 2,317) versus COVID-19-negative (COVIDneg; n = 74,850) patients for the week preceding the PCR testing date, we identify anosmia/dysgeusia (27.1-fold), fever/chills (2.6-fold), respiratory difficulty (2.2-fold), cough (2.2-fold), myalgia/arthralgia (2-fold), and diarrhea (1.4-fold) as significantly amplified in COVIDpos over COVIDneg patients. The combination of cough and fever/chills has 4.2-fold amplification in COVIDpos patients during the week prior to PCR testing, in addition to anosmia/dysgeusia, constitutes the earliest EHR-derived signature of COVID-19. This study introduces an Augmented Intelligence platform for the real-time synthesis of institutional biomedical knowledge. The platform holds tremendous potential for scaling up curation throughput, thus enabling EHR-powered early disease diagnosis.
Keywords
- electronic health record
- neural networks
- machine learning
- artificial intelligence
- COVID-19
- SARS-CoV-2
