Current applications of artificial intelligence for intraoperative decision support in surgery
Allison J. Navarrete-Welton1, Daniel A. Hashimoto1,2()
1. Surgical Artificial Intelligence and Innovation Laboratory, Massachusetts General Hospital, Boston, MA 02114, USA 2. Harvard Medical School, Boston, MA 02114, USA
Research into medical artificial intelligence (AI) has made significant advances in recent years, including surgical applications. This scoping review investigated AI-based decision support systems targeted at the intraoperative phase of surgery and found a wide range of technological approaches applied across several surgical specialties. Within the twenty-one (n=21) included papers, three main categories of motivations were identified for developing such technologies: (1) augmenting the information available to surgeons, (2) accelerating intraoperative pathology, and (3) recommending surgical steps. While many of the proposals hold promise for improving patient outcomes, important methodological shortcomings were observed in most of the reviewed papers that made it difficult to assess the clinical significance of the reported performance statistics. Despite limitations, the current state of this field suggests that a number of opportunities exist for future researchers and clinicians to work on AI for surgical decision support with exciting implications for improving surgical care.
. [J]. Frontiers of Medicine, 2020, 14(4): 369-381.
Allison J. Navarrete-Welton, Daniel A. Hashimoto. Current applications of artificial intelligence for intraoperative decision support in surgery. Front. Med., 2020, 14(4): 369-381.
Artificial intelligence applied to decision support for the surgeon during the intraoperative phase of surgery
Decision support during the preoperative and postoperative surgical phases Anesthesia, surgical training, and surgeon skill evaluation when unrelated to clinical decision support
Studies with at least 10 human patients
Studies with fewer than 10 patients
Published prior to May 25, 2019
Animal studies
Peer-reviewed published literature and conference proceedings papers
Narrative review papers, editorials, letters to the editor, abstracts
All geographical areas but written in English
Languages other than English
Tab.1
Fig.1
Fig.2
Paper
Data set
Data set processing
Data set distribution
Validation
Performancea
André et al. 2009 [17]
54 patients; colonic polyps
1036 images; videos discarded if histology and pCLE diagnoses did not match
Roughly equal (2 classes)
Leave-n-out cross validation (patient held-out)
80.1% weighted k-NN classification accuracy (benign vs. pathological)
André et al. 2010 [15]
68 patients; colonic polyps
121 single-polyp videos
Roughly equal (2 classes)
Leave-one-patient-out cross-validation
94.2% weighted k-NN classification accuracy (benign vs. pathological)
André et al. 2012 [16]
66 patients; colonic polyps
118 single-polyp videos
Unspecified (8 binary classes)
30 × 3-fold cross-validation (patient segregated)
96.7% AUC for the highest-performing semantic concept (“elongated crypt”) 49.4% Kendall τ rank correlation coefficient (Likert similarity scale)
Map locations of glioblastoma tissue, normal tissue, hypervascularized tissue, and background material
Support vector machine, convolutional neural network/deep neural network joint architecture
Tumor classification (n = 3)
Wan et al. (2015) [25]
pCLE
Distinguish between glioblastoma and meningioma
Feature descriptors, bag-of-visual-words dimensionality reduction, support vector machine
Kamen et al. (2016) [26]
pCLE
Distinguish between glioblastoma and meningioma
Feature descriptors, sparse coding with locality constraint to reduce dimensionality, support vector machine
Li et al. (2018) [27]
pCLE
Distinguish between glioblastoma and meningioma
Convolutional neural network, long short-term memory neural network
Tissue identification (n = 4)
Couceiro et al. (2012) [28]
pCLE
Classify low or high probability of inflammatory bowel disease, based on intestinal crypts
Feature descriptors, support vector machine
Halicek et al. (2017) [29]
HSI
Classify normal or cancerous thyroid and aerodigestive tract tissues
Convolutional neural network
Halicek et al. (2018) [30]
HSI
Distinguish thyroid carcinoma from normal tissue
Convolutional neural network
Hou et al. (2019) [32]
OCT
Distinguish metastatic lymph nodes from normal lymph nodes
Artificial neural network
Tab.3
1
F Spencer. Teaching and measuring surgical techniques: the technical evaluation of competence. Bull Am Coll Surg 1978; 63: 9–12
2
JW Suliburk, QM Buck, CJ Pirko, NN Massarweh, NR Barshes, H Singh, TK Rosengart. Analysis of human performance deficiencies associated with surgical adverse events. JAMA Netw Open 2019; 2(7): e198067 https://doi.org/10.1001/jamanetworkopen.2019.8067
pmid: 31365107
3
CM Pugh, S Santacaterina, DA DaRosa, RE Clark. Intra-operative decision making: more than meets the eye. J Biomed Inform 2011; 44(3): 486–496 https://doi.org/10.1016/j.jbi.2010.01.001
pmid: 20096376
4
DA Hashimoto, CG Axelsson, CB Jones, R Phitayakorn, E Petrusa, SK McKinley, D Gee, C Pugh. Surgical procedural map scoring for decision-making in laparoscopic cholecystectomy. Am J Surg 2019; 217(2): 356–361 https://doi.org/10.1016/j.amjsurg.2018.11.011
pmid: 30470551
5
CM Pugh, DA DaRosa. Use of cognitive task analysis to guide the development of performance-based assessments for intraoperative decision making. Mil Med 2013; 178(10 Suppl): 22–27 https://doi.org/10.7205/MILMED-D-13-00207
pmid: 24084302
DA Hashimoto, G Rosman, ER Witkowski, C Stafford, AJ Navarette-Welton, DW Rattner, KD Lillemoe, DL Rus, OR Meireles. Computer vision analysis of intraoperative video: automated recognition of operative steps in laparoscopic sleeve gastrectomy. Ann Surg 2019; 270(3): 414–421 https://doi.org/10.1097/SLA.0000000000003460
pmid: 31274652
K Bera, KA Schalper, DL Rimm, V Velcheti, A Madabhushi. Artificial intelligence in digital pathology—new tools for diagnosis and precision oncology. Nat Rev Clin Oncol 2019; 16(11): 703–715 https://doi.org/10.1038/s41571-019-0252-y
pmid: 31399699
11
DT Hogarty, JC Su, K Phan, M Attia, M Hossny, S Nahavandi, P Lenane, FJ Moloney, A Yazdabadi. Artificial intelligence in dermatology—where we are and the way to the future: a review. Am J Clin Dermatol 2020; 21(1): 41–47
pmid: 31278649
12
L Maier-Hein, SS Vedula, S Speidel, N Navab, R Kikinis, A Park, M Eisenmann, H Feussner, G Forestier, S Giannarou, M Hashizume, D Katic, H Kenngott, M Kranzfelder, A Malpani, K März, T Neumuth, N Padoy, C Pugh, N Schoch, D Stoyanov, R Taylor, M Wagner, GD Hager, P Jannin. Surgical data science for next-generation interventions. Nat Biomed Eng 2017; 1(9): 691–696 https://doi.org/10.1038/s41551-017-0132-7
pmid: 31015666
B Harangi, A Hajdu, R Lampe, P Torok. Recognizing ureter and uterine artery in endoscopic images using a convolutional neural network. In: 2017 IEEE 30th International Symposium on Computer-Based Medical Systems (CBMS). 2017. 726–727. doi: 10.1109/CBMS.2017.137
15
B André, T Vercauteren, AM Buchner, MB Wallace, N Ayache. Endomicroscopic video retrieval using mosaicing and visualwords. In: 2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro. 2010. doi: 10.1109/isbi.2010.5490265
16
B André, T Vercauteren, AM Buchner, MB Wallace, N Ayache. Learning semantic and visual similarity for endomicroscopy video retrieval. IEEE Trans Med Imaging 2012; 31(6): 1276–1288 https://doi.org/10.1109/TMI.2012.2188301
pmid: 22353403
17
B André, T Vercauteren, A Perchant, A Buchner, M Wallace, N Ayache. Endomicroscopic image retrieval and classification using invariant visual features. In: 2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro. 2009. doi: 10.1109/isbi.2009.5193055
18
M Kohandani Tafresh, N Linard, B André, N Ayache, T Vercauteren. Semi-automated query construction for content-based endomicroscopy video retrieval. In: Medical Image Computing and Computer-Assisted Intervention — MICCAI 2014. Springer International Publishing, 2014. 89–96. doi: 10.1007/978-3-319-10404-1_12
19
Y Gu, J Yang, GZ Yang. Multi-view multi-modal feature embedding for endomicroscopy mosaic classification. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. 2016. 11–19
20
Y Gu, K Vyas, J Yang, GZ Yang. Unsupervised feature learning for endomicroscopy image retrieval. In: Medical Image Computing and Computer Assisted Intervention — MICCAI 2017. Springer International Publishing, 2017. 64–71 doi: 10.1007/978-3-319-66179-7_8
21
G Quellec, M Lamard, G Cazuguel, Z Droueche, C Roux, B Cochener. Real-time retrieval of similar videos with application to computer-aided retinal surgery. Conf Proc IEEE Eng Med Biol Soc 2011; 2011: 4465–4468 https://doi.org/10.1109/IEMBS.2011.6091107
pmid: 22255330
22
K Ritschel, I Pechlivanis, S Winter. Brain tumor classification on intraoperative contrast-enhanced ultrasound. Int J CARS 2015; 10(5): 531–540 https://doi.org/10.1007/s11548-014-1089-6
pmid: 24956998
23
E Ilunga-Mbuyamba, D Lindner, J Avina-Cervantes, F Arlt, H Rostro-Gonzalez, I Cruz-Aceves, C Chalopin. Fusion of intraoperative 3D B-mode and contrast-enhanced ultrasound data for automatic identification of residual brain tumors. Appl Sci (Basel) 2017; 7(4): 415 https://doi.org/10.3390/app7040415
24
P Dollar, Z Tu, P Perona, S Belongie. Integral channel features. In: Procedings of the British Machine Vision Conference. 2009. doi: 10.5244/c.23.91
25
S Wan, S Sun, S Bhattacharya, S Kluckner, A Gigler, E Simon, M Fleischer, P Charalampaki, T Chen, A Kamen. Towards an efficient computational framework for guiding surgical resection through intra-operative endo-microscopic pathology. In: Medical Image Computing and Computer-Assisted Intervention — MICCAI 2015. Springer International Publishing, 2015. 421–429. doi: 10.1007/978-3-319-24553-9_52
26
A Kamen, S Sun, S Wan, S Kluckner, T Chen, AM Gigler, E Simon, M Fleischer, M Javed, S Daali, A Igressa, P Charalampaki. Automatic tissue differentiation based on confocal endomicroscopic images for intraoperative guidance in neurosurgery. BioMed Res Int 2016; 2016: 6183218 https://doi.org/10.1155/2016/6183218
pmid: 27127791
27
Y Li, P Charalampaki, Y Liu, GZ Yang, S Giannarou. Context aware decision support in neurosurgical oncology based on an efficient classification of endomicroscopic data. Int J CARS 2018; 13(8): 1187–1199 https://doi.org/10.1007/s11548-018-1806-7
pmid: 29948845
28
S Couceiro, JP Barreto, P Freire, P Figueiredo. Description and classification of confocal endomicroscopic images for the automatic diagnosis of inflammatory bowel disease. In: Machine Learning in Medical Imaging. Springer Berlin Heidelberg, 2012. 144–151. doi: 10.1007/978-3-642-35428-1_18
29
M Halicek, G Lu, JV Little, X Wang, M Patel, CC Griffith, MW El-Deiry, AY Chen, B Fei. Deep convolutional neural networks for classifying head and neck cancer using hyperspectral imaging. J Biomed Opt 2017; 22(6): 60503 https://doi.org/10.1117/1.JBO.22.6.060503
pmid: 28655055
30
M Halicek, JV Little, X Wang, M Patel, CC Griffith, MW El-Deiry, AY Chen, B Fei. Optical biopsy of head and neck cancer using hyperspectral imaging and convolutional neural networks. Proc SPIE Int Soc Opt Eng 2018; 104690X doi: 10.1117/12.2289023 https://doi.org/10.1117/12.2289023
pmid: 30197462
31
H Fabelo, M Halicek, S Ortega, M Shahedi, A Szolna, JF Piñeiro, C Sosa, AJ O’Shanahan, S Bisshopp, C Espino, M Márquez, M Hernández, D Carrera, J Morera, GM Callico, R Sarmiento, B Fei. Deep learning-based framework for in vivo identification of glioblastoma tumor using hyperspectral images of human brain. Sensors (Basel) 2019; 19(4): 920 https://doi.org/10.3390/s19040920
pmid: 30813245
32
F Hou, Y Liang, Z Yang, W Gu, Y Yu. Automatic identification of metastatic lymph nodes in OCT images. Proceedings Volume 10867, Optical Coherence Tomography and Coherence Domain Optical Methods in Biomedicine XXIII; 108673G. 2019. doi: 10.1117/12.2511588 https://doi.org/10.1117/12.2511588
33
S Tian, XC Yin, ZB Wang, F Zhou, HW Hao. A VidEo-Based Intelligent Recognition and Decision System for the phacoemulsification cataract surgery. Comput Math Methods Med 2015; 2015: 202934 https://doi.org/10.1155/2015/202934
pmid: 26693249
34
B Fan, HX Li, Y Hu. An intelligent decision system for intraoperative somatosensory evoked potential monitoring. IEEE Trans Neural Syst Rehabil Eng 2016; 24(2): 300–307 https://doi.org/10.1109/TNSRE.2015.2477557
pmid: 26415181
35
L Gordon, T Grantcharov, F Rudzicz. Explainable artificial intelligence for safe intraoperative decision support. JAMA Surg 2019; 154(11): 1064 https://doi.org/10.1001/jamasurg.2019.2821
pmid: 31509185
J Krause, V Gulshan, E Rahimy, P Karth, K Widner, GS Corrado, L Peng, DR Webster. Grader variability and the importance of reference standards for evaluating machine learning models for diabetic retinopathy. Ophthalmology 2018; 125(8): 1264–1272 https://doi.org/10.1016/j.ophtha.2018.01.034
pmid: 29548646
38
H Lee, S Yune, M Mansouri, M Kim, SH Tajmir, CE Guerrier, SA Ebert, SR Pomerantz, JM Romero, S Kamalian, RG Gonzalez, MH Lev, S Do. An explainable deep-learning algorithm for the detection of acute intracranial haemorrhage from small datasets. Nat Biomed Eng 2019; 3(3): 173–182 https://doi.org/10.1038/s41551-018-0324-9
pmid: 30948806
39
A Esteva, B Kuprel, RA Novoa, J Ko, SM Swetter, HM Blau, S Thrun. Dermatologist-level classification of skin cancer with deep neural networks. Nature 2017; 542(7639): 115–118 https://doi.org/10.1038/nature21056
pmid: 28117445