Related Experiment Video
Updated: Nov 24, 2025
07:15
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
7.2K
Machine learning monitoring for laser osteotomy
Sergey Shevchik1, Hervé Nguendon Kenhagho2, Tri Le-Quang1
1Laboratory for Advanced Materials Processing, Empa-Swiss Federal Laboratories for Materials Science and Technology, Thun, Switzerland.
Journal of Biophotonics
|December 28, 2020
Summary
This study introduces a new method using sound analysis and machine learning to identify tissue types during laser osteotomy. This technology can help surgeons precisely remove tissue and protect healthy areas.
Area of Science:
- Biomedical Engineering
- Surgical Technology
- Acoustic Signal Processing
Background:
- Laser osteotomy requires precise tissue ablation.
- Current methods lack real-time feedback on tissue type and laser energy dosage.
- Minimizing damage to surrounding healthy tissues is crucial.
Purpose of the Study:
- To develop an online monitoring system for laser osteotomy.
- To differentiate between various ablated tissue types and applied laser energy doses.
- To enhance surgical precision and safety during bone cutting procedures.
Main Methods:
- Utilizing airborne microphone sensors to detect laser-induced acoustic emissions.
- Analyzing acoustic signals with a supervised machine learning algorithm.
- Experimentally evaluating the system on skin, fat, muscle, and bone tissues.
Main Results:
- Achieved classification accuracy ranging from 84% to 99% across different tissue types.
- Demonstrated the feasibility of collecting training datasets under real-life surgical conditions.
- Compared the performance of several cutting-edge machine learning frameworks.
Conclusions:
- The proposed acoustic analysis method effectively monitors laser osteotomy in real-time.
- This technology has the potential to assist surgeons by improving tissue differentiation and energy control.
- Future applications include minimizing collateral damage and enabling cleaner pathological tissue removal.
