Application of vibration signals in medical drill wear monitoring

Murat, Zrinka and Brezak, Danko and Majetić, Dubravko and Udiljak, Toma (2018) Application of vibration signals in medical drill wear monitoring. = Application of vibration signals in medical drill wear monitoring. In: The 2018 IAENG International Conference on Artificial Intelligence and Applications, 14-16.03.2018., Hong Kong, NR Kina.

Majetić, Murat, Brezak, Udljak_Application of vibration signals in medical drill wear monitoringIMECS2018_pp7-10.pdf - Published Version Jezik dokumenta:English

Download (1MB) | Preview
Official URL:


Usage of worn drills in medical interventions has highly negative influence on the heat generation which can consequently result in thermal necrosis of bone tissue and prolonged postoperative healing process. Precise real time direct measurement of cutting tool wear level during machining process is not possible. Therefore, the main goal of this study was to identify wear level using indirect monitoring technique based on tool wear features extracted from vibration signals measured during drilling process. Experiment was based on fresh bovine bone samples which were drilled with standard surgical drill used for bone and joint surgery applications. Three drill wear levels in combination with the 12 different machining parameter sets were analyzed. Drill wear features were analyzed using Radial Basis Function Neural Network algorithm with Gaussian activation function. The best overall classification precision regarding all three wear levels was around 79%, while the third and the highest wear level analyzed in this study was exactly classified in 95% of all test samples.

Item Type: Conference or Workshop Item (Lecture)
Keywords (Croatian): drilling; thermal osteonecrosis; vibrations; wear modelling
Subjects: TECHNICAL SCIENCE > Mechanical Engineering
Divisions: 1200 Department of Technology > 1230 Chair of Machine Tools
900 Department of Robotics and Production System Automation > 910 Chair of Engineering Automation
Indexed in Web of Science: No
Indexed in Current Contents: No
Date Deposited: 20 Sep 2018 11:36
Last Modified: 15 Nov 2018 14:56

Actions (login required)

View Item View Item


Downloads per month over past year

Nema podataka za dohvacanje citata