Assessing Endoscopic Response in Locally Advanced Rectal Cancer Treated with Total Neoadjuvant Therapy: Development and Validation of a Highly Accurate Convolutional Neural Network.

BACKGROUND: Rectal tumors display varying degrees of response to total neoadjuvant therapy (TNT). We evaluated the performance of a convolutional neural network (CNN) in interpreting endoscopic images of either a non-complete response to TNT or local regrowth during watch-and-wait surveillance. METHODS: Endoscopic images from stage II/III rectal cancers treated with TNT from 2012 to 2020 at a single institution were retrospectively reviewed. Images were labelled as Tumor or No Tumor based on endoscopy timing (before, during, or after treatment) and the tumor's endoluminal response. A CNN was trained using ResNet-50 architecture. The area under the curve (AUC) was analyzed during training and for two test sets. The main test set included images of tumors treated with TNT. The other contained images of local regrowth. The model's performance was compared to sixteen surgeons and surgical trainees who evaluated 119 images for evidence of tumor. Fleiss' kappa was calculated by respondent experience level. RESULTS: A total of 2717 images from 288 patients were included; 1407 (51.8%) contained tumor. The AUC was 0.99, 0.98, and 0.92 for training, main test, and local regrowth test sets. The model performed on par with surgeons of all experience levels for the main test set. Interobserver agreement was good ( k = 0.71-0.81). All groups outperformed the model in identifying tumor from images of local regrowth. Interobserver agreement was fair to moderate ( k = 0.24-0.52). CONCLUSIONS: A highly accurate CNN matched the performance of colorectal surgeons in identifying a noncomplete response to TNT. However, the model demonstrated suboptimal accuracy when analyzing images of local regrowth.

Automatic segmentation for magnetic resonance imaging guided individual elective lymph node irradiation in head and neck cancer patients.

Background and purpose: In head and neck squamous cell carcinoma (HNSCC) patients, the radiation dose to nearby organs at risk can be reduced by restricting elective neck irradiation from lymph node levels to individual lymph nodes. However, manual delineation of every individual lymph node is time-consuming and error prone. Therefore, automatic magnetic resonance imaging (MRI) segmentation of individual lymph nodes was developed and tested using a convolutional neural network (CNN). Materials and methods: In 50 HNSCC patients (UMC-Utrecht), individual lymph nodes located in lymph node levels Ib-II-III-IV-V were manually segmented on MRI by consensus of two experts, obtaining ground truth segmentations. A 3D CNN (nnU-Net) was trained on 40 patients and tested on 10. Evaluation metrics were Dice Similarity Coefficient (DSC), recall, precision, and F1-score. The segmentations of the CNN was compared to segmentations of two observers. Transfer learning was used with 20 additional patients to re-train and test the CNN in another medical center. Results: nnU-Net produced automatic segmentations of elective lymph nodes with median DSC: 0.72, recall: 0.76, precision: 0.78, and F1-score: 0.78. The CNN had higher recall compared to both observers (p = 0.002). No difference in evaluation scores of the networks in both medical centers was found after re-training with 5 or 10 patients. Conclusion: nnU-Net was able to automatically segment individual lymph nodes on MRI. The detection rate of lymph nodes using nnU-Net was higher than manual segmentations. Re-training nnU-Net was required to successfully transfer the network to the other medical center.