A convolutional neural network algorithm for automatic segmentation of head and neck organs at risk using deep lifelong learning. Issue 5 (4th April 2019)

• Record Type: Journal Article
• Authors: Chan, Jason W.; Kearney, Vasant; Haaf, Samuel; Wu, Susan; Bogdanov, Madeleine; Reddick, Mariah; Dixit, Nayha; Sudhyadhom, Atchar; Chen, Josephine; Yom, Sue S.; Solberg, Timothy D.
• Is Part Of: Medical physics. Volume 46:Issue 5(2019)
• Page Start: 2204
• Page End: 2213
• Publication Date: 2019-04-04
• Subjects: convolutional neural network -- autosegmentation -- head and neck -- deep lifelong learning; Medical physics -- Periodicals; Medical physics; Geneeskunde; Natuurkunde; Toepassingen; Biophysics; Periodicals; Periodicals; Electronic journals; 610.153
• DOI: 10.1002/mp.13495
• Languages: English
• ISSNs: 0094-2405

Abstract

Abstract : Purpose: This study suggests a lifelong learning‐based convolutional neural network (LL‐CNN) algorithm as a superior alternative to single‐task learning approaches for automatic segmentation of head and neck (OARs) organs at risk. Methods and materials: Lifelong learning‐based convolutional neural network was trained on twelve head and neck OARs simultaneously using a multitask learning framework. Once the weights of the shared network were established, the final multitask convolutional layer was replaced by a single‐task convolutional layer. The single‐task transfer learning network was trained on each OAR separately with early stoppage. The accuracy of LL‐CNN was assessed based on Dice score and root‐mean‐square error (RMSE) compared to manually delineated contours set as the gold standard. LL‐CNN was compared with 2D‐UNet, 3D‐UNet, a single‐task CNN (ST‐CNN), and a pure multitask CNN (MT‐CNN). Training, validation, and testing followed Kaggle competition rules, where 160 patients were used for training, 20 were used for internal validation, and 20 in a separate test set were used to report final prediction accuracies. Results: On average contours generated with LL‐CNN had higher Dice coefficients and lower RMSE than 2D‐UNet, 3D‐Unet, ST‐ CNN, and MT‐CNN. LL‐CNN required ~72 hrs to train using a distributed learning framework on 2 Nvidia 1080Ti graphics processing units. LL‐CNN required 20 s to predict all 12 OARs, which was approximately as fast as the fastest alternative methods with the exception of MT‐CNN. Conclusions: This study demonstrated that for head and neck organs at risk, LL‐CNN achieves a prediction accuracy superior to all alternative algorithms.