Title: Advancing surgical instrument recognition through shape recognition techniques in the medical industry

Authors: Bijay Kumar Paikaray; Sonia Rathee; Shalu; Amita Yadav; Tiruveedula GopiKrishna; Bijay Kishor Shishir Sekhar Pattanaik

Addresses: Centre for Data Science, Department of Computer Science and Engineering, Siksha 'O' Anusandhan (Deemed to be) University, Odisha, 751003, India ' Department of Computer Science and Engineering, Maharaja Surajmal Institute of Technology, New Delhi, 110058, India ' Department of Computer Science and Engineering, Maharaja Surajmal Institute of Technology, New Delhi, 110058, India ' Department of Computer Science and Engineering, Maharaja Surajmal Institute of Technology, New Delhi, 110058, India ' Department of Computer Science and Engineering, Adama Science and Technology University, Adama, 1888, Ethiopia ' Department of Computer Science and Engineering, Gandhi Institute of Technological Advancement (GITA) Autonomous College, Bhubaneswar, Odisha, 752054, India

Abstract: Computer assisted intervention (CAI) system is anticipating surgical workflow. Its goal is to use of instruments, supporting the intraoperative clinical decision support. The shape recognition of surgical instruments enables the identification of the different surgical instruments which are almost similar in shape and size and also helps to understand the category of each instrument. The proposed method is where one can extract the feature set of an object and compare it with the feature set of the universal collection of objects. The patch-based segmentation algorithm that is being suggested can get an F-score of 0.90. With a variety of instrument layouts, on average, the recommended force based grasping protocol achieves a 92% picking success rate, and the recommended attention based instrument recognition module achieves a 95.6% recognition accuracy. The end result consists of the name of the object, along with the percent classification and percent recognition with the universal object collection.

Keywords: universal object collection; Euclidean distance; percent classification; percent recognition; object pixels.

DOI: 10.1504/IJCBDD.2025.151207

International Journal of Computational Biology and Drug Design, 2025 Vol.16 No.4, pp.301 - 321

Received: 20 Apr 2024
Accepted: 01 Jan 2025
Published online: 19 Jan 2026 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article