Evaluation of suturing performance in general surgery and ocular microsurgery by combining computer vision-based software and distributed fiber optic strain sensors: a proof-of-concept

• Published in: International journal for computer assisted radiology and surgery 2020-08, Vol.15 (8), p.1359-1367
• Main Authors: Handelman, Amir, Keshet, Yariv, Livny, Eitan, Barkan, Refael, Nahum, Yoav, Tepper, Ronnie
• Format: Article
• Language: English
• Subjects: Algorithms; Clinical Competence; Computer Imaging; Computer Science; Computer vision; Cornea; Fiber optics; Health Informatics; Humans; Image processing; Imaging; Medicine; Medicine & Public Health; Microsurgery; Microsurgery - methods; Ophthalmologic Surgical Procedures; Optical fibers; Original Article; Pattern Recognition and Graphics; Performance evaluation; Quality assessment; Radiology; Sensors; Skills; Software; Stitches; Stitching; Surgery; Suture Techniques; Sutures; Transplantation; Vision; Wound healing
• ISSN: 1861-6410; 1861-6429
• DOI: 10.1007/s11548-020-02187-y

Purpose Improper suturing may cause an inadequate wound healing process and wound dehiscence as well as infection and even graft rejection in case of corneal transplantation. Hence, training surgeons in correct suturing procedures and objectively assessing their surgical skills is desirable. Methods Two complementary methods for assessment of suturing skills in two medical fields (general surgery and ocular microsurgery) were demonstrated. Suturing quality is assessed by computer vision software. Evaluation of stitching flow of operation is based on measuring strain induced in an optical fiber that is placed in proximity to the wound and parallel thereto and is pressed and passed by wound stitches. Results Our software generated a score for suturing outcome in both general surgery and ocular microsurgery when the stitching was done on a patch. Every trainee received a score in the range 0–100 that describes his/her performance. Strain values were recognized when using a patch in general surgery and a rubber patch in ocular microsurgery, but were less distinct in (disqualified) human cornea. Conclusions We proved a concept of an objective scoring method (based on various image processing algorithms) for assessment of suturing performance. It was also shown that fiber optic strain sensors are sensitive to the flow of stitching operation on a patch but are less sensitive to the flow of stitching operation on a human cornea. By combining these two methods, we can comprehensively evaluate the suturing performance objectively.