Although machine learning has become a powerful tool to augment doctors in clinical analysis, the immense amount of labeled data that is necessary to train supervised learning approaches burdens each development task as time and resource intensive. The vast majority of dense clinical information is stored in written reports, detailing pertinent patient information. The challenge with utilizing natural language data for standard model development is due to the complex nature of the modality. In this research, a model pipeline was developed to utilize an unsupervised approach to train an encoder-language model, a recurrent network, to generate document encodings; which then can be used as features passed into a decoder-classifier model that requires magnitudes less labeled data than previous approaches to differentiate between fine-grained disease classes accurately. The language model was trained on unlabeled radiology reports from the Massachusetts General Hospital Radiology Department (n=218,159) and terminated with a loss of 1.62. The classification models were trained on three labeled datasets of head CT studies of reported patients, presenting large vessel occlusion (n=1403), acute ischemic strokes (n=331), and intracranial hemorrhage (n=4350), to identify a variety of different findings directly from the radiology report data; resulting in AUCs of 0.98, 0.95, and 0.99, respectively, for the large vessel occlusion, acute ischemic stroke, and intracranial hemorrhage datasets. The output encodings are able to be used in conjunction with imaging data, to create models that can process a multitude of different modalities. The ability to automatically extract relevant features from textual data allows for faster model development and integration of textual modality, overall, allowing clinical reports to become a more viable input for more encompassing and accurate deep learning models.
[1]
Luca Antiga,et al.
Automatic differentiation in PyTorch
,
2017
.
[2]
E. Hing,et al.
Use and characteristics of electronic health record systems among office-based physician practices: United States, 2001-2012.
,
2012,
NCHS data brief.
[3]
E. Hing,et al.
Use and characteristics of electronic health record systems among office-based physician practices: United States, 2001-2013.
,
2014,
NCHS data brief.
[4]
Francesca N. Delling,et al.
Heart Disease and Stroke Statistics—2019 Update: A Report From the American Heart Association
,
2019,
Circulation.
[5]
George Hripcsak,et al.
Research Paper: The Role of Domain Knowledge in Automating Medical Text Report Classification
,
2003,
J. Am. Medical Informatics Assoc..
[6]
Bing Shi,et al.
Multi-Step-Ahead Prediction with Long Short Term Memory Networks and Support Vector Regression
,
2018,
2018 37th Chinese Control Conference (CCC).
[7]
Jennifer G. Dy,et al.
Medical coding classification by leveraging inter-code relationships
,
2010,
KDD.
[8]
J. Gornbein,et al.
Ischemic Strokes Due to Large-Vessel Occlusions Contribute Disproportionately to Stroke-Related Dependence and Death: A Review
,
2017,
Front. Neurol..
[9]
S. Brunak,et al.
Mining electronic health records: towards better research applications and clinical care
,
2012,
Nature Reviews Genetics.
[10]
A. Rabinstein,et al.
Absolute and Relative Contraindications to IV rt-PA for Acute Ischemic Stroke
,
2015,
The Neurohospitalist.
[11]
Lukasz Kaiser,et al.
Attention is All you Need
,
2017,
NIPS.