Keywords: Deep learning, maxillary sinus lesion, label-free segmentation
Abstract: To alleviate the workload of data annotation, we propose a label-free approach (SinusNet) for the segmentation of maxillary sinus lesions in CBCT images via learning the anomaly features from diverse synthetic lesions within the normal maxillary sinus. Our SinusNet achieved average F1 of 80.9 ± 11.6%, precision of 82.7 ± 9.1%, and recall of 80.1 ± 15.0%, respectively, and comparable performance with those of previous supervised approaches.
Registration: I acknowledge that acceptance of this work at MIDL requires at least one of the authors to register and present the work during the conference.
Authorship: I confirm that I am the author of this work and that it has not been submitted to another publication before.
Paper Type: novel methodological ideas without extensive validation
Primary Subject Area: Segmentation
Secondary Subject Area: Learning with Noisy Labels and Limited Data
Confidentiality And Author Instructions: I read the call for papers and author instructions. I acknowledge that exceeding the page limit and/or altering the latex template can result in desk rejection.