Meeting Banner
Abstract #1011

A comprehensive study of machine-assisted classifiers for predicting prostate cancer Gleason grade

Jing Wang1, Yang Fan2, and Yudong Zhang3

1Center for Medical Device Evaluation, CFDA, Beijing, People's Republic of China, 2MR Research China, GE Healthcare, Beijing, People's Republic of China, 3Department of Radiology, the First Affiliated Hospital with Nanjing Medical University, Nanjing, People's Republic of China

We performed comprehensive study of eight popularized classifiers for predicting prostate cancer (PCa) Gleason score (GS). The multi-parametric MRI data was obtained from 205 histopathology-confirmed PCa. The MR features were modeled using eight classifiers to predict high-GS (4+4) PCa, including Logistic Regression (LR), Artificial Neural Network (ANN), Support Vector Machine (SVM), Naive Bayes (NB), Relevance Vector Machine (RVM), Least Absolute Shrinkage and Selection operator (LASSO), Discriminant Analysis (DA) and Decision Tree (DT) analysis. Results showed that LASSO and DA had significantly higher area under curve than other classifiers, thus could be valuable for automatic prediction of PCa grade.

This abstract and the presentation materials are available to members only; a login is required.

Join Here