Clinical Text Classification of Alzheimer’s Drugs’ Mechanism of Action

Document Type

Conference Proceeding

Publication Date

1-1-2022

Publication Title

Lecture Notes in Networks and Systems

Publisher

Springer

Publisher Location

New York, NY

Volume

235

First page number:

513

Last page number:

521

Abstract

The Alzheimer’s disease Drug Development Pipeline [1, 2] delivers updates on potential AD-treatment, as well as drug development ongoing in clinical trials. To create these reports, researchers manually extract information from several resources like ClinicalTrials.gov and drug manufacturer websites; however, some of these items require expert review, such as when predicting a drug’s Mechanism of Action (MOA). In this paper, we aim to assist researchers by predicting and suggesting a drug’s MOA using Machine Learning. We test Random Forest (RF), Logistic Regression (LR), Support Vector Machine (SVM), XGBoost, and Decision Tree (DT) models. The latter showing the most promising results, with 95% accuracy, 100% recall, and a 0.92 F1-score.

Keywords

Alzheimer’s drug; Clinical text classification; Machine learning; Mechanism of action

Disciplines

Chemicals and Drugs | Nervous System Diseases

UNLV article access

Share

COinS