Amin Madani1, Babak Namazi2, Maria S Altieri3, Daniel A Hashimoto4, Angela Maria Rivera1, Philip H Pucher5, Allison Navarrete-Welton4, Ganesh Sankaranarayanan2, L Michael Brunt6, Allan Okrainec1, Adnan Alseidi7. 1. Department of Surgery, University Health Network, Toronto, ON, Canada. 2. Center for Evidence-Based Simulation, Baylor University Medical Center, Dallas, TX. 3. Department of Surgery, East Carolina University Brody School of Medicine, Greenville, NC. 4. Department of Surgery, Surgical AI & Innovation Laboratory, Massachusetts General Hospital, Boston, MA. 5. Department of Surgery, Imperial College London, London, UK. 6. Department of Surgery, Washington University School of Medicine, St. Louis, MO. 7. Department of Surgery, University of California - San Francisco, San Francisco, CA.
Abstract
OBJECTIVE: The aim of this study was to develop and evaluate the performance of artificial intelligence (AI) models that can identify safe and dangerous zones of dissection, and anatomical landmarks during laparoscopic cholecystectomy (LC). SUMMARY BACKGROUND DATA: Many adverse events during surgery occur due to errors in visual perception and judgment leading to misinterpretation of anatomy. Deep learning, a subfield of AI, can potentially be used to provide real-time guidance intraoperatively. METHODS: Deep learning models were developed and trained to identify safe (Go) and dangerous (No-Go) zones of dissection, liver, gallbladder, and hepatocystic triangle during LC. Annotations were performed by 4 high-volume surgeons. AI predictions were evaluated using 10-fold cross-validation against annotations by expert surgeons. Primary outcomes were intersection- over-union (IOU) and F1 score (validated spatial correlation indices), and secondary outcomes were pixel-wise accuracy, sensitivity, specificity, ± standard deviation. RESULTS: AI models were trained on 2627 random frames from 290 LC videos, procured from 37 countries, 136 institutions, and 153 surgeons. Mean IOU, F1 score, accuracy, sensitivity, and specificity for the AI to identify Go zones were 0.53 (±0.24), 0.70 (±0.28), 0.94 (±0.05), 0.69 (±0.20). and 0.94 (±0.03), respectively. For No-Go zones, these metrics were 0.71 (±0.29), 0.83 (±0.31), 0.95 (±0.06), 0.80 (±0.21), and 0.98 (±0.05), respectively. Mean IOU for identification of the liver, gallbladder, and hepatocystic triangle were: 0.86 (±0.12), 0.72 (±0.19), and 0.65 (±0.22), respectively. CONCLUSIONS: AI can be used to identify anatomy within the surgical field. This technology may eventually be used to provide real-time guidance and minimize the risk of adverse events.
OBJECTIVE: The aim of this study was to develop and evaluate the performance of artificial intelligence (AI) models that can identify safe and dangerous zones of dissection, and anatomical landmarks during laparoscopic cholecystectomy (LC). SUMMARY BACKGROUND DATA: Many adverse events during surgery occur due to errors in visual perception and judgment leading to misinterpretation of anatomy. Deep learning, a subfield of AI, can potentially be used to provide real-time guidance intraoperatively. METHODS: Deep learning models were developed and trained to identify safe (Go) and dangerous (No-Go) zones of dissection, liver, gallbladder, and hepatocystic triangle during LC. Annotations were performed by 4 high-volume surgeons. AI predictions were evaluated using 10-fold cross-validation against annotations by expert surgeons. Primary outcomes were intersection- over-union (IOU) and F1 score (validated spatial correlation indices), and secondary outcomes were pixel-wise accuracy, sensitivity, specificity, ± standard deviation. RESULTS: AI models were trained on 2627 random frames from 290 LC videos, procured from 37 countries, 136 institutions, and 153 surgeons. Mean IOU, F1 score, accuracy, sensitivity, and specificity for the AI to identify Go zones were 0.53 (±0.24), 0.70 (±0.28), 0.94 (±0.05), 0.69 (±0.20). and 0.94 (±0.03), respectively. For No-Go zones, these metrics were 0.71 (±0.29), 0.83 (±0.31), 0.95 (±0.06), 0.80 (±0.21), and 0.98 (±0.05), respectively. Mean IOU for identification of the liver, gallbladder, and hepatocystic triangle were: 0.86 (±0.12), 0.72 (±0.19), and 0.65 (±0.22), respectively. CONCLUSIONS: AI can be used to identify anatomy within the surgical field. This technology may eventually be used to provide real-time guidance and minimize the risk of adverse events.
Authors: Lawrence W Way; Lygia Stewart; Walter Gantert; Kingsway Liu; Crystine M Lee; Karen Whang; John G Hunter Journal: Ann Surg Date: 2003-04 Impact factor: 12.969
Authors: Amin Madani; Jordan Gornitsky; Yusuke Watanabe; Cassandre Benay; Maria S Altieri; Philip H Pucher; Roger Tabah; Elliot J Mitmaker Journal: World J Surg Date: 2018-02 Impact factor: 3.352
Authors: Anand Malpani; S Swaroop Vedula; Henry C Lin; Gregory D Hager; Russell H Taylor Journal: Int J Comput Assist Radiol Surg Date: 2020-05-08 Impact factor: 2.924
Authors: Selwyn O Rogers; Atul A Gawande; Mary Kwaan; Ann Louise Puopolo; Catherine Yoon; Troyen A Brennan; David M Studdert Journal: Surgery Date: 2006-07 Impact factor: 3.982
Authors: Andre Esteva; Brett Kuprel; Roberto A Novoa; Justin Ko; Susan M Swetter; Helen M Blau; Sebastian Thrun Journal: Nature Date: 2017-01-25 Impact factor: 49.962
Authors: Simon Laplante; Babak Namazi; Parmiss Kiani; Daniel A Hashimoto; Adnan Alseidi; Mauricio Pasten; L Michael Brunt; Sujata Gill; Brian Davis; Matthew Bloom; Luise Pernar; Allan Okrainec; Amin Madani Journal: Surg Endosc Date: 2022-08-02 Impact factor: 3.453
Authors: R B den Boer; C de Jongh; W T E Huijbers; T J M Jaspers; J P W Pluim; R van Hillegersberg; M Van Eijnatten; J P Ruurda Journal: Surg Endosc Date: 2022-08-04 Impact factor: 3.453
Authors: Belinda De Simone; Fikri M Abu-Zidan; Andrew A Gumbs; Elie Chouillard; Salomone Di Saverio; Massimo Sartelli; Federico Coccolini; Luca Ansaloni; Toby Collins; Yoram Kluger; Ernest E Moore; Andrej Litvin; Ari Leppaniemi; Pietro Mascagni; Luca Milone; Micaela Piccoli; Mohamed Abu-Hilal; Michael Sugrue; Walter L Biffl; Fausto Catena Journal: World J Emerg Surg Date: 2022-02-10 Impact factor: 5.469
Authors: Julia Gong; F Christopher Holsinger; Julia E Noel; Sohei Mitani; Jeff Jopling; Nikita Bedi; Yoon Woo Koh; Lisa A Orloff; Claudio R Cernea; Serena Yeung Journal: Sci Rep Date: 2021-07-12 Impact factor: 4.379