Clinically Acceptable Segmentation of Organs at Risk in Cervical Cancer Radiation Treatment from Clinically Available Annotations
This addresses the challenge of accurate organ segmentation for cervical cancer patients using clinically available but imperfect data, representing an incremental improvement in medical imaging.
The paper tackles the problem of automatic segmentation of Organs at Risk in cervical cancer radiation treatment from a large, noisy CT dataset with missing annotations, using data cleaning and a semi-supervised learning approach, resulting in segmentation masks that are equally clinically acceptable as manual contours.
Deep learning models benefit from training with a large dataset (labeled or unlabeled). Following this motivation, we present an approach to learn a deep learning model for the automatic segmentation of Organs at Risk (OARs) in cervical cancer radiation treatment from a large clinically available dataset of Computed Tomography (CT) scans containing data inhomogeneity, label noise, and missing annotations. We employ simple heuristics for automatic data cleaning to minimize data inhomogeneity and label noise. Further, we develop a semi-supervised learning approach utilizing a teacher-student setup, annotation imputation, and uncertainty-guided training to learn in presence of missing annotations. Our experimental results show that learning from a large dataset with our approach yields a significant improvement in the test performance despite missing annotations in the data. Further, the contours generated from the segmentation masks predicted by our model are found to be equally clinically acceptable as manually generated contours.