Automatic Detection of Pulmonary Embolism using Computational Intelligence

TL;DR

This paper presents an automated system for detecting pulmonary embolism in lung scans using image segmentation, PCA for feature extraction, and a neural network ensemble trained with Hybrid Monte Carlo, achieving promising results.

Contribution

It introduces a novel combination of segmentation, PCA, and ensemble neural networks trained with Hybrid Monte Carlo for pulmonary embolism detection.

Findings

Effective lung scan segmentation and feature extraction.

High accuracy achieved with neural network ensemble.

Robust detection performance demonstrated.

Abstract

This article describes the implementation of a system designed to automatically detect the presence of pulmonary embolism in lung scans. These images are firstly segmented, before alignment and feature extraction using PCA. The neural network was trained using the Hybrid Monte Carlo method, resulting in a committee of 250 neural networks and good results are obtained.