What question did this study set out to answer?

This study aims to improve patient cohort selection from electronic health records for oncology research and measure potential bias.

January 13, 2020Open Access

Model-assisted cohort selection with bias analysis for generating large-scale cohorts from the EHR for oncology research

View Full Paper

Ask AI

Bookmark

This study aims to improve patient cohort selection from electronic health records for oncology research and measure potential bias.
Trained a model on 17,263 patients using term-frequency inverse-document-frequency (TF-IDF) and logistic regression.
Tested model performance on a separate set of 17,292 patients and performed bias analysis.
Compared MACS-generated cohorts to traditional methods by analyzing clinical and demographic variable distributions.
Algorithm achieved an area under the curve (AUC) of 0.976 and sensitivity of 96.0%.
Abstraction efficiency gain of 77.9% was noted.
No significant differences in baseline characteristics or analysis results were observed during bias analysis.

No takes yet. Share an insight, caveat, or question.

Randomized trial reveals efficient cohort selection in oncology patients, indicating enhanced reliability of outcomes research.

Birnbaum et al. (2020) studied this question.

View Full Paper

Ask AI

Bookmark