A comparison of methods for fully automatic segmentation of tumors and involved nodes in PET/CT of head and neck cancers

Abstract

Target volume delineation is a vital but time-consuming and challenging part of radiotherapy, where the goal is to deliver sufficient dose to the target while reducing risks of side effects. For head and neck cancer (HNC) this is complicated by the complex anatomy of the head and neck region and the proximity of target volumes to organs at risk. The purpose of this study was to compare and evaluate conventional PET thresholding methods, six classical machine learning algorithms and a 2D U-Net convolutional neural network (CNN) for automatic gross tumor volume (GTV) segmentation of HNC in PET/CT images. For the latter two approaches the impact of single versus multimodality input on segmentation quality was also assessed. 197 patients were included in the study. The cohort was split into training and test sets (157 and 40 patients, respectively). Five-fold cross-validation was used on the training set for model comparison and selection. Manual GTV delineations represented the ground truth. Tresholding, classical machine learning and CNN segmentation models were ranked separately according to the cross-validation Sørensen–Dice similarity coefficient (Dice). PET thresholding gave a maximum mean Dice of 0.62, whereas classical machine learning resulted in maximum mean Dice scores of 0.24 (CT) and 0.66 (PET; PET/CT). CNN models obtained maximum mean Dice scores of 0.66 (CT), 0.68 (PET) and 0.74 (PET/CT). The difference in cross-validation Dice between multimodality PET/CT and single modality CNN models was significant (p ≤ 0.0001). The top-ranked PET/CT-based CNN model outperformed the best-performing thresholding and classical machine learning models, giving significantly better segmentations in terms of cross-validation and test set Dice, true positive rate, positive predictive value and surface distance-based metrics (p ≤ 0.0001). Thus, deep learning based on multimodality PET/CT input resulted in superior target coverage and less inclusion of surrounding normal tissue.

Fully Automatic Quantitative Measurement of 18F-FDG PET/CT in Thymic Epithelial Tumors Using a Convolutional Neural Network

“…Nevertheless, with this relatively small number of PET/CT images with heterogeneous tumor FDG uptake, we still achieved higher segmentation performance than in previous studies. [12][13][14][15] Second, our PET/CT images were from a specific series of PET/CT scanners in a single institution, which could hamper the generalizability of our results. The semiquantitative metabolic parameters are affected by scanner model and reconstruction protocol.…”

Automated Liver Lesion Detection in 68Ga DOTATATE PET / CT Using a Deep Fully Convolutional Neural Network

“…To our knowledge, this is the rst study to report automated hepatic lesion detection using deep learning in 68 Ga-DOTATATE PET. Hepatic lesion detectability is an especially di cult task in 68 Ga-DOTATATE imaging because of higher normal background liver activity [5,25], and high variability of uptake compared to normal hepatic FDG uptake [25][26][27][28]. Prior studies of 68 Ga-DOTATATE PET have con rmed that detectability is not only dependent on tumor uptake, but it is also highly dependent upon the normal background, and the background image noise [29,30].…”

Semi-automatic tumor segmentation of rectal cancer based on functional magnetic resonance imaging

“…The union of both manual contours was used as ground truth. The features were arranged in a data matrix as described in [29] and [30] and illustrated in Supplementary Fig. S1 .…”