Accepted for/Published in: JMIR Medical Informatics
Date Submitted: Jan 15, 2020
Open Peer Review Period: Jan 14, 2020 - Feb 17, 2020
Date Accepted: Mar 25, 2020
(closed for review but you can still tweet)
Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Use of Clinical Notes and Machine Learning to Predict Onset of Dementia
ABSTRACT
Background:
Clinical trials need efficient tools to assist in recruiting patients at risk of Alzheimer’s Disease and related dementia (ADRD). Early detection can also assist patients with financial planning for long-term care. Clinical notes are an important under-utilized source of information in machine learning models due to the cost of collection and complexity of analysis.
Objective:
This study investigates using de-identified clinical notes from multiple hospital systems collected over 10 years to augment retrospective machine learning models of ADRD risk.
Methods:
The models use two years of data to predict a future outcome of ADRD onset. Notes data are provided in a de-identified format with specific terms and sentiments. Terms in notes are embedded into a 100-dimensional vector space to identify clusters of related terms and abbreviations that differ across hospital systems and individual clinicians.
Results:
When using notes, AUC improved from 85% to 94% and positive predictive value (PPV) increased from 45% to 68% in the model at disease onset. Models with notes improved in both AUC and PPV in years 3-6 when notes volume was largest, results are mixed in years 7 and 8 with smallest cohorts.
Conclusions:
While notes helped in the short term, the presence of ADRD symptomatic terms years earlier than onset adds evidence to other studies that clinicians under-code ADRD diagnoses. De-identified clinical notes increase the accuracy of risk models. Clinical notes collected across multiple hospital systems via natural language processing can be merged using post-processing techniques to aid model accuracy.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.