Clinically Applicable Segmentation of Head and Neck Anatomy for Radiotherapy: Deep Learning Algorithm Development and Validation Study

• Published in: Journal of medical Internet research 2021-07, Vol.23 (7), p.e26151
• Main Authors: Nikolov, Stanislav, Blackwell, Sam, Zverovitch, Alexei, Mendes, Ruheena, Livne, Michelle, De Fauw, Jeffrey, Patel, Yojan, Meyer, Clemens, Askham, Harry, Romera-Paredes, Bernadino, Kelly, Christopher, Karthikesalingam, Alan, Chu, Carlton, Carnell, Dawn, Boon, Cheng, D'Souza, Derek, Moinuddin, Syed Ali, Garie, Bethany, McQuinlan, Yasmin, Ireland, Sarah, Hampton, Kiarna, Fuller, Krystle, Montgomery, Hugh, Rees, Geraint, Suleyman, Mustafa, Back, Trevor, Hughes, Cían Owen, Ledsam, Joseph R, Ronneberger, Olaf
• Format: Article
• Language: English
• Subjects: Algorithms; Anatomy; Artificial intelligence; Automation; Biological organs; Business metrics; Clinical medicine; Cognitive style; Contours; Datasets; Deep learning; Delineation; Generalizability; Head & neck cancer; Human performance; Medical imaging; Original Paper; Patients; Planning; Radiation; Radiation therapy; Radiographers; Radiotherapy; Relevance; Segmentation; Tomography; Tumors; Validation studies
• ISSN: 1438-8871; 1439-4456; 1438-8871
• DOI: 10.2196/26151

Background: Over half a million individuals are diagnosed with head and neck cancer each year globally. Radiotherapy is an important curative treatment for this disease, but it requires manual time to delineate radiosensitive organs at risk. This planning process can delay treatment while also introducing interoperator variability, resulting in downstream radiation dose differences. Although auto-segmentation algorithms offer a potentially time-saving solution, the challenges in defining, quantifying, and achieving expert performance remain. Objective: Adopting a deep learning approach, we aim to demonstrate a 3D U-Net architecture that achieves expert-level performance in delineating 21 distinct head and neck organs at risk commonly segmented in clinical practice. Methods: The model was trained on a data set of 663 deidentified computed tomography scans acquired in routine clinical practice and with both segmentations taken from clinical practice and segmentations created by experienced radiographers as part of this research, all in accordance with consensus organ at risk definitions. Results: We demonstrated the model’s clinical applicability by assessing its performance on a test set of 21 computed tomography scans from clinical practice, each with 21 organs at risk segmented by 2 independent experts. We also introduced surface Dice similarity coefficient, a new metric for the comparison of organ delineation, to quantify the deviation between organ at risk surface contours rather than volumes, better reflecting the clinical task of correcting errors in automated organ segmentations. The model’s generalizability was then demonstrated on 2 distinct open-source data sets, reflecting different centers and countries to model training. Conclusions: Deep learning is an effective and clinically applicable technique for the segmentation of the head and neck anatomy for radiotherapy. With appropriate validation studies and regulatory approvals, this system could improve the efficiency, consistency, and safety of radiotherapy pathways.