Deep-Learning-Based Fast Optical Coherence Tomography (OCT) Image Denoising for Smart Laser Osteotomy

被引:16
作者
Bayhaqi, Yakub A. [1 ]
Hamidi, Arsham [1 ]
Canbaz, Ferda [1 ]
Navarini, Alexander A. [2 ,3 ]
Cattin, Philippe C. [4 ]
Zam, Azhar [5 ,6 ,7 ]
机构
[1] Univ Basel, Dept Biomed Engn, Biomed Laser & Opt Grp BLOG, CH-4123 Allschwil, Switzerland
[2] Univ Basel, Dept Biomed Engn, Digital Dermatol Grp, CH-4123 Allschwil, Switzerland
[3] Univ Hosp Basel, Dept Dermatol, CH-4031 Basel, Switzerland
[4] Univ Basel, Dept Biomed Engn, Ctr Med Image Anal & Nav CIAN, CH-4123 Allschwil, Switzerland
[5] Univ Basel, Dept Biomed Engn, CH-4123 Allschwil, Switzerland
[6] NYU, Dept Biomed Engn, Brooklyn, NY 11201 USA
[7] New York Univ Abu Dhabi, Div Engn, Abu Dhabi, U Arab Emirates
关键词
Laser ablation; Speckle; Laser feedback; Image denoising; Artificial neural networks; Noise reduction; Laser beam cutting; Deep learning; image denoising; image processing; optical coherence tomography; BREAKDOWN SPECTROSCOPY LIBS; TISSUE; SURGERY; NOISE; DIFFERENTIATION; ENHANCEMENT; SPECKLE;
D O I
10.1109/TMI.2022.3168793
中图分类号
TP39 [计算机的应用];
学科分类号
081203 ; 0835 ;
摘要
Laser osteotomy promises precise cutting and minor bone tissue damage. We proposed Optical Coherence Tomography (OCT) to monitor the ablation process toward our smart laser osteotomy approach. The OCT image is helpful to identify tissue type and provide feedback for the ablation laser to avoid critical tissues such as bone marrow and nerve. Furthermore, in the implementation, the tissue classifier's accuracy is dependent on the quality of the OCT image. Therefore, image denoising plays an important role in having an accurate feedback system. A common OCT image denoising technique is the frame-averaging method. Inherent to this method is the need for multiple images, i.e., the more images used, the better the resulting image quality. However, this approach comes at the price of increased acquisition time and sensitivity to motion artifacts. To overcome these limitations, we applied a deep-learning denoising method capable of imitating the frame-averaging method. The resulting image had a similar image quality to the frame-averaging and was better than the classical digital filtering methods. We also evaluated if this method affects the tissue classifier model's accuracy that will provide feedback to the ablation laser. We found that image denoising significantly increased the accuracy of the tissue classifier. Furthermore, we observed that the classifier trained using the deep learning denoised images achieved similar accuracy to the classifier trained using frame-averaged images. The results suggest the possibility of using the deep learning method as a pre-processing step for real-time tissue classification in smart laser osteotomy.
引用
收藏
页码:2615 / 2628
页数:14
相关论文
共 43 条
[1]   Laser-induced breakdown spectroscopy as a potential tool for autocarbonization detection in laserosteotomy [J].
Abbasi, Hamed ;
Rauter, Georg ;
Guzman, Raphael ;
Cattin, Philippe C. ;
Zam, Azhar .
JOURNAL OF BIOMEDICAL OPTICS, 2018, 23 (07)
[2]  
[Anonymous], 2014, COMPUT RES REPOSITOR
[3]  
Arjovsky M, 2017, PR MACH LEARN RES, V70
[4]   Clinical applicability of robot-guided contact-free laser osteotomy in cranio-maxillo-facial surgery: in-vitro simulation and in-vivo surgery in minipig mandibles [J].
Baek, K. -W. ;
Deibel, W. ;
Marinov, D. ;
Griessen, M. ;
Bruno, A. ;
Zeilhofer, H. -F. ;
Cattin, Ph ;
Juergens, Ph .
BRITISH JOURNAL OF ORAL & MAXILLOFACIAL SURGERY, 2015, 53 (10) :976-981
[5]   A comparative investigation of bone surface after cutting with mechanical tools and Er:YAG laser [J].
Baek, Kyung-won ;
Deibel, Waldemar ;
Marinov, Dilyan ;
Griessen, Mathias ;
Dard, Michel ;
Bruno, Alfredo ;
Zeilhofer, Hans-Florian ;
Cattin, Philippe ;
Juergens, Philipp .
LASERS IN SURGERY AND MEDICINE, 2015, 47 (05) :426-432
[6]   Signal averaging improves signal-to-noise in OCT images: But which approach works best, and when? [J].
Baumann, Bernhard ;
Merkle, Conrad W. ;
Leitgeb, Rainer A. ;
Augustin, Marco ;
Wartak, Andreas ;
Pircher, Michael ;
Hitzenberger, Christoph .
BIOMEDICAL OPTICS EXPRESS, 2019, 10 (11) :5755-5775
[7]   Optoacoustic monitoring of cutting efficiency and thermal damage during laser ablation [J].
Bay, Erwin ;
Douplik, Alexandre ;
Razansky, Daniel .
LASERS IN MEDICAL SCIENCE, 2014, 29 (03) :1029-1035
[8]   Optimizing controlled laser cutting of hard tissue (bone) [J].
Bernal, Lina M. Beltran ;
Schmidt, Iris T. ;
Vulin, Nikola ;
Widmer, Jonas ;
Snedeker, Jess G. ;
Cattin, Philippe C. ;
Zam, Azhar ;
Rauter, Georg .
AT-AUTOMATISIERUNGSTECHNIK, 2018, 66 (12) :1072-1082
[9]   Optical coherence tomography for optical biopsy - Properties and demonstration of vascular pathology [J].
Brezinski, ME ;
Tearney, GJ ;
Bouma, BE ;
Izatt, JA ;
Hee, MR ;
Swanson, EA ;
Southern, JF ;
Fujimoto, JG .
CIRCULATION, 1996, 93 (06) :1206-1213