Using uncertainty estimation to reduce false positives in liver lesion detection

TL;DR

This paper presents a method combining neural network uncertainty estimation with an SVM classifier to significantly reduce false positives in liver lesion detection from medical images, improving diagnostic accuracy.

Contribution

The study introduces a novel approach that leverages uncertainty maps and an SVM classifier to effectively filter out false positives in deep learning-based medical image analysis.

Findings

Dropout rate of 0.5 minimizes false positives.

The classifier filters out approximately 90% of false positives.

Method improves accuracy of liver lesion detection.

Abstract

Despite the successes of deep learning techniques at detecting objects in medical images, false positive detections occur which may hinder an accurate diagnosis. We propose a technique to reduce false positive detections made by a neural network using an SVM classifier trained with features derived from the uncertainty map of the neural network prediction. We demonstrate the effectiveness of this method for the detection of liver lesions on a dataset of abdominal MR images. We find that the use of a dropout rate of 0.5 produces the least number of false positives in the neural network predictions and the trained classifier filters out approximately 90% of these false positives detections in the test-set.