Automated Lung Cancer Segmentation Using a PET and CT Dual-Modality Deep Learning Neural Network.

Researchers

Journal

Modalities

Models

Abstract

To develop an automated lung tumor segmentation method for radiotherapy planning based on deep learning and dual-modality PET and CT images.A 3D convolutional neural network using inputs from diagnostic PETs and simulation CTs was constructed with two parallel convolution paths for independent feature extraction at multiple resolution levels and a single deconvolution path. At each resolution level, the extracted features from the convolution arms were concatenated and fed through the skip connections into the deconvolution path that produced the tumor segmentation. Our network was trained/validated/tested by a 3:1:1 split on 290 pairs of PET and CT images from lung cancer patients treated at our clinic, with manual physician contours as the ground truth. A stratified training strategy based on the magnitude of the gross tumor volume (GTV) was investigated to improve performance, especially for small tumors. Multiple radiation oncologists assessed the clinical acceptability of the network-produced segmentations.The mean Dice similarity coefficient (DSC), Hausdorff Distance (HD), and bi-directional local distance (BLD) comparing manual versus automated contours were 0.79±0.10, 5.8±3.2 mm, and 2.8±1.5 mm for the unstratified 3D dual-modality model. Stratification delivered the best results when the model for the large GTVs (>25 ml) was trained with all-size GTVs and model for the small GTVs (<25 ml) was trained with small GTVs only. The best combined DSC, HD, and BLD from the two stratified models on their corresponding test datasets were 0.83±0.07, 5.9±2.5 mm, and 2.8±1.4 mm, respectively. In the multi-observer review, 91.25% manual vs. 88.75% automatic contours were Accepted or Accepted with Modifications.By utilizing an expansive clinical PET and CT image database and a dual-modality architecture, the proposed 3D network with a novel GTV volume-based stratification strategy generated clinically useful lung cancer contours that were highly acceptable on physician review.Copyright © 2022. Published by Elsevier Inc.