Share Email Print

Proceedings Paper

Monte Carlo investigation of deep learning tissue classification performance in OCT-based smart laser bone surgery (Conference Presentation)
Author(s): Yakub Aqib Bayhaqi; Arsham Hamidi; Alexander Navarini; Philippe C. Cattin; Azhar Zam

Paper Abstract

Automatic tissue classification using optical coherence tomography (OCT) explores the possibility to control laser ablation in prevention for collateral damage of critical tissues. During ablation, tissue experience thermal dissipation which induces mechanical expansion and optical properties alteration. We reconstructed OCT images of bone, fat, and muscle tissues for pre and post ablation temperatures condition using Monte Carlo simulation. We trained a deep neural network to recognize tissue type based on reconstructed OCT images with pre-ablation temperature condition and tested it on post-ablation temprature condition. The reconstructed images show small changes in the tissue structure but do not significantly affect the performance of the classifier.

Paper Details

Date Published: 9 March 2020
Proc. SPIE 11229, Advanced Biomedical and Clinical Diagnostic and Surgical Guidance Systems XVIII, 112290H (9 March 2020); doi: 10.1117/12.2541604
Show Author Affiliations
Yakub Aqib Bayhaqi, Univ. Basel (Switzerland)
Arsham Hamidi, Univ. Basel (Switzerland)
Alexander Navarini, Univ. Basel (Switzerland)
Philippe C. Cattin, Univ. Basel (Switzerland)
Azhar Zam, Univ. Basel (Switzerland)

Published in SPIE Proceedings Vol. 11229:
Advanced Biomedical and Clinical Diagnostic and Surgical Guidance Systems XVIII
Anita Mahadevan-Jansen, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?