Classification of Motor Control Features for Surgical Skills Assessment during Robotic Assisted Surgery
Gorantla, Kuber Reddy
MetadataShow full item record
With the advent of new technologies in Surgery, Robotic Assisted Surgery (RAS) has been the point of research. Surgical Skill Assessment in RAS has increased interest through which the training and feedback to surgeons can be given based on the task performance. In this research, we use the experimental data of Urethro-vesicle anastomosis (UVA) on an inanimate model done by two groups of participants, namely experts(E), Novice(N)using the da Vinci® Surgical System (Sunnyvale, CA). Motor Control features, which are a part of psychomotor learning, are developed based on the camera plane coordinates by performing the feature engineering and the classiﬁcation is being done using the pattern recognition principles. Our classiﬁcation method when compared to the manual encoding of subject expertise based on the Dreyfus model resulted in an accuracy>98% for the intra-subject classiﬁcation and >76% average accuracy for Inter-Subject Classiﬁcation. Additionally, this study could form a basis for closed loop surgical training, speciﬁcally for the novitiate surgeons.