Automated Pulmonary Embolism Detection from CTPA Images Using an End-to-End Convolutional Neural Network

TL;DR

This paper introduces an end-to-end CNN for pulmonary embolism detection in CTPA images, jointly optimizing candidate detection and false positive removal, leading to improved sensitivity over existing methods.

Contribution

The study presents a novel integrated CNN architecture with three subnets that jointly optimize PE candidate detection and false positive elimination, enhancing detection accuracy.

Findings

Achieved 78.9% sensitivity at 2 false positives per volume on PE challenge dataset.

Achieved 86.8% sensitivity at 5mm localization error on own dataset.

Outperformed state-of-the-art methods in PE detection accuracy.

Abstract

Automated methods for detecting pulmonary embolisms (PEs) on CT pulmonary angiography (CTPA) images are of high demand. Existing methods typically employ separate steps for PE candidate detection and false positive removal, without considering the ability of the other step. As a result, most existing methods usually suffer from a high false positive rate in order to achieve an acceptable sensitivity. This study presents an end-to-end trainable convolutional neural network (CNN) where the two steps are optimized jointly. The proposed CNN consists of three concatenated subnets: 1) a novel 3D candidate proposal network for detecting cubes containing suspected PEs, 2) a 3D spatial transformation subnet for generating fixed-sized vessel-aligned image representation for candidates, and 3) a 2D classification network which takes the three cross-sections of the transformed cubes as input and eliminates false positives. We have evaluated our approach using the 20 CTPA test dataset from the PE challenge, achieving a sensitivity of 78.9%, 80.7% and 80.7% at 2 false positives per volume at 0mm, 2mm and 5mm localization error, which is superior to the state-of-the-art methods. We have further evaluated our system on our own dataset consisting of 129 CTPA data with a total of 269 emboli. Our system achieves a sensitivity of 63.2%, 78.9% and 86.8% at 2 false positives per volume at 0mm, 2mm and 5mm localization error.