Deep learning classification of nasal anatomical variants on computed tomography for endoscopic sinus surgery.
Authors
Affiliations (6)
Affiliations (6)
- Division of Cancer, Department of Surgery and Cancer, Faculty of Medicine, Imperial College London, London, W12 0NN, UK.
- Department of Rhinology, Faculty of Medicine, Imperial College London, London, SW7 2AZ, UK.
- Department of Computing, Faculty of Engineering, Imperial College London, London, SW7 2AZ, UK.
- Department of Rhinology, Faculty of Medicine, Imperial College London, London, SW7 2AZ, UK. [email protected].
- Division of Cancer, Department of Surgery and Cancer, Faculty of Medicine, Imperial College London, London, W12 0NN, UK. [email protected].
- Department of Environmental Health Sciences, Yale University, New Haven, CT, 06520, USA. [email protected].
Abstract
Endoscopic sinus surgery (ESS) carries risks such as vision loss and intracranial injury due to the proximity of critical structures and unrecognised anatomical variants. We developed convolutional neural networks (CNNs) to detect free-floating anterior ethmoidal arteries (FFAEA), Onodi cells, and Haller cells on coronal sinus CT, and evaluated BioMedCLIP, a biomedical vision-language model (VLM), in a few-shot setting. CT scans from 122 ESS patients were anonymised, standardised coronal CT images were captured, and variant presence was annotated. Five ImageNet-pretrained CNN backbones were assessed using repeated patient-wise five-fold cross-validation across 40 configurations. The best CNNs achieved balanced accuracies of 77.8 ± 2.3% (FFAEA), 74.6 ± 2.5% (Onodi), and 63.7 ± 6.2% (Haller). BioMedCLIP achieved 65.5 ± 3.2%, 63.8 ± 1.8%, and 73.5 ± 3.6%, respectively, outperforming CNNs for Haller cell detection while providing competitive performance for the other variants. These models demonstrate proof-of-concept performance for automated identification of selected sinonasal variants on standardised coronal CT images under internal patient-wise cross-validation.