Relational reasoning network for anatomical landmarking

• Published in: Journal of medical imaging (Bellingham, Wash.) Vol. 10; no. 2; p. 024002
• Main Authors: Torosdagli, Neslisah, Anwar, Syed, Verma, Payal, Liberton, Denise K., Lee, Janice S., Han, Wade W., Bagci, Ulas
• Format: Journal Article
• Language: English
• Published: United States Society of Photo-Optical Instrumentation Engineers 01.03.2023; SPIE
• Subjects: Ablation (Surgery); Birth defects; CT imaging; Image Processing; Orthodontics; anatomical landmarking; deep relational learning; craniomaxillofacial bones; surgical modeling; relational reasoning
• ISSN: 2329-4302; 2329-4310; 2329-4310
• DOI: 10.1117/1.JMI.10.2.024002

We perform anatomical landmarking for craniomaxillofacial (CMF) bones without explicitly segmenting them. Toward this, we propose a simple, yet efficient, deep network architecture, called relational reasoning network (RRN), to accurately learn the local and the global relations among the landmarks in CMF bones; specifically, mandible, maxilla, and nasal bones. The proposed RRN works in an end-to-end manner, utilizing learned relations of the landmarks based on dense-block units. For a given few landmarks as input, RRN treats the landmarking process similar to a data imputation problem where predicted landmarks are considered missing. We applied RRN to cone-beam computed tomography scans obtained from 250 patients. With a fourfold cross-validation technique, we obtained an average root mean squared error of per landmark. Our proposed RRN has revealed unique relationships among the landmarks that help us in inferring informativeness of the landmark points. The proposed system identifies the missing landmark locations accurately even when severe pathology or deformations are present in the bones. Accurately identifying anatomical landmarks is a crucial step in deformation analysis and surgical planning for CMF surgeries. Achieving this goal without the need for explicit bone segmentation addresses a major limitation of segmentation-based approaches, where segmentation failure (as often is the case in bones with severe pathology or deformation) could easily lead to incorrect landmarking. To the best of our knowledge, this is the first-of-its-kind algorithm finding anatomical relations of the objects using deep learning.