Deep semi-supervised learning for automatic segmentation of inferior alveolar nerve using a convolutional neural network

BMC Oral Health

Seung-Hyun Kim¹  In-Seok Song²  Ho-Kyung Lim³  Seok-Ki Jung⁴  Yongwon Cho⁵ 

[1] Department of Medical Humanities, Korea University College of Medicine, 46, Gaeunsa 2-gil, Seongbuk-gu, 02842, Seoul, Republic of Korea;Department of Oral and Maxillofacial Surgery, Korea University Anam Hospital, 73, Goryeodae-ro, Seongbuk-gu, 02841, Seoul, Republic of Korea;Department of Oral and Maxillofacial Surgery, Korea University Guro Hospital, 148, Gurodong-ro, Guro-gu, 08308, Seoul, Republic of Korea;Department of Orthodontics, Korea University Guro Hospital, 148, Gurodong-ro, Guro-gu, 08308, Seoul, Republic of Korea;Department of Radiology and AI Center, Korea University College of Medicine, Korea University Anam Hospital, 73, Goryeodae-ro, Seongbuk-gu, 02841, Seoul, Republic of Korea;

关键词: Deep learning;    Convolutional neural network;    Automatic segmentation;    Inferior alveolar nerve;   

DOI  :  10.1186/s12903-021-01983-5

来源: Springer

【 摘 要 】

BackgroundThe inferior alveolar nerve (IAN) innervates and regulates the sensation of the mandibular teeth and lower lip. The position of the IAN should be monitored prior to surgery. Therefore, a study using artificial intelligence (AI) was planned to image and track the position of the IAN automatically for a quicker and safer surgery.MethodsA total of 138 cone-beam computed tomography datasets (Internal: 98, External: 40) collected from multiple centers (three hospitals) were used in the study. A customized 3D nnU-Net was used for image segmentation. Active learning, which consists of three steps, was carried out in iterations for 83 datasets with cumulative additions after each step. Subsequently, the accuracy of the model for IAN segmentation was evaluated using the 50 datasets. The accuracy by deriving the dice similarity coefficient (DSC) value and the segmentation time for each learning step were compared. In addition, visual scoring was considered to comparatively evaluate the manual and automatic segmentation.ResultsAfter learning, the DSC gradually increased to 0.48 ± 0.11 to 0.50 ± 0.11, and 0.58 ± 0.08. The DSC for the external dataset was 0.49 ± 0.12. The times required for segmentation were 124.8, 143.4, and 86.4 s, showing a large decrease at the final stage. In visual scoring, the accuracy of manual segmentation was found to be higher than that of automatic segmentation.ConclusionsThe deep active learning framework can serve as a fast, accurate, and robust clinical tool for demarcating IAN location.

【 授权许可】

CC BY