Remote AI algorithm detects Critical View of Safety in pediatric surgery across hospitals

Remote AI algorithm reliably detects safety criteria in pediatric surgery

Chinese Academy of SciencesJun 1 2026 Laparoscopic cholecystectomy is increasingly performed in children, yet bile duct injury remains a serious and potentially devastating complication. The CVS requiring complete dissection of the hepatocystic triangle, separation of the gallbladder from the liver, and clear visualization of the two structures entering the gallbladder -- was designed to reduce this risk. However, correct identification of CVS is highly subjective and varies widely among surgeons. Standardized scoring systems exist but are inconsistently applied, particularly in pediatric settings where procedures are less frequent than in adults. While AI has shown promise for CVS detection in adult surgery, its use in children and remote environments has remained completely unexplored. Due to these challenges, a dedicated evaluation of real-time, AI-assisted CVS detection in pediatric laparoscopic cholecystectomy is urgently needed. Researchers in Argentina have now tested exactly that. In a study (DOI: 10.1136/wjps-2025-001125) published on July 1, 2026, in the World Journal of Pediatric Surgery , a team from the Hospital del Niño Prof. Dr. Ramón Exeni and the Hospital Dr. Cosme Argerich demonstrated that a remotely deployed AI algorithm can reliably detect CVS during live pediatric laparoscopic cholecystectomy. The system processed surgical video transmitted via a standard Zoom teleconferencing platform, identified anatomical structures in real time, and alerted the surgeon when all safety criteria were met -- all without requiring any on-site AI hardware at the operating location. The team first trained their AI algorithm on more than 1,000 images extracted from 346 adult laparoscopic cholecystectomy videos, teaching it to recognize the cystic artery, cystic duct, hepatocystic window, and cystic plate. They then tested the system live in 50 pediatric patients aged 6 to 18 years. The surgical video was streamed 21 kilometers to a second hospital, where the algorithm analyzed each frame, drawing colored boxes around target structures -- blue for the cystic artery, green for the cystic duct, light gray for the hepatocystic window, and dark gray for the cystic plate. Only when all four appeared together did the system trigger an audible "detected" alarm, signaling that the three Strasberg criteria for CVS were fully met. Two expert surgeons, completely blinded to the AI's output, independently assessed CVS presence. The agreement was perfect, with a Cohen's kappa value of 1.0. In 38 of 50 cases, a complete CVS was detected. In the remaining 12 cases, both the algorithm and the surgeons agreed that one or more elements were missing -- most often the cystic artery or the cystic plate. No postoperative complications occurred, and the remote setup added no technical failures beyond three excluded transmission issues. The system was designed to prioritize high specificity, avoiding false alarms even if that meant occasional false negatives. The authors said the system was not meant to replace surgical judgment but to act as an extra set of eyes during a critical step in the operation. "We designed the algorithm to be very specific -- it only alarms when every safety element is visible," they explained. "That means it may miss some incomplete dissections on purpose, but it never cries wolf. For pediatric surgeons, who perform far fewer gallbladder operations than their adult colleagues, this kind of remote assistance could be a real safety net during training or in low-resource settings. We see it as a scalable tool, not a substitute for experience." This approach could help reduce the subjectivity of CVS identification, particularly in hospitals without specialized AI infrastructure or where pediatric caseloads are low. Because the algorithm runs on a remote server and communicates via standard teleconferencing, the operating room only requires an internet connection and a video feed. This design makes advanced intraoperative guidance potentially accessible to community hospitals and surgical training centers. The authors note that larger, multicenter prospective studies are still needed, especially to test the system in younger children under 5 years and in patients with severe inflammation or rare anatomical variations such as aberrant ducts or supernumerary arteries. If validated further, remote AI-assisted CVS detection could become a practical, scalable tool for improving surgical safety in pediatric laparoscopic cholecystectomy worldwide. Source: Chinese Academy of Sciences Journal reference: Olivieri, S. E., et al. (2026). Remote detection of the critical view of safety in pediatric laparoscopic cholecystectomy using artifitial intelligence. World Journal of Pediatric Surgery. DOI: 10.1136/wjps-2025-001125. https://wjps.bmj.com/content/9/1/e001125

Virtual Eyes in the Operating Room: AI Detects Surgical Safety Landmark Across Hospitals | Newswise

Real-time detection of the cystic artery (blue box), cystic duct (green box), and hepatocystic window (gray box) by the AI algorithm. CA, cystic artery; CD, cystic duct; CP, cystic plate; CVS, critical view of safety; HCW, hepatocystic window. Newswise -- Laparoscopic cholecystectomy is increasingly performed in children, yet bile duct injury remains a serious and potentially devastating complication. The CVSrequiring complete dissection of the hepatocystic triangle, separation of the gallbladder from the liver, and clear visualization of the two structures entering the gallbladder -- was designed to reduce this risk. However, correct identification of CVS is highly subjective and varies widely among surgeons. Standardized scoring systems exist but are inconsistently applied, particularly in pediatric settings where procedures are less frequent than in adults. While AI has shown promise for CVS detection in adult surgery, its use in children and remote environments has remained completely unexplored. Due to these challenges, a dedicated evaluation of real-time, AI-assisted CVS detection in pediatric laparoscopic cholecystectomy is urgently needed. Researchers in Argentina have now tested exactly that. In a study (DOI: 10.1136/wjps-2025-001125) published on July 1, 2026, in the World Journal of Pediatric Surgery , a team from the Hospital del Niño Prof. Dr. Ramón Exeni and the Hospital Dr. Cosme Argerich demonstrated that a remotely deployed AI algorithm can reliably detect CVS during live pediatric laparoscopic cholecystectomy. The system processed surgical video transmitted via a standard Zoom teleconferencing platform, identified anatomical structures in real time, and alerted the surgeon when all safety criteria were met -- all without requiring any on-site AI hardware at the operating location. The team first trained their AI algorithm on more than 1,000 images extracted from 346 adult laparoscopic cholecystectomy videos, teaching it to recognize the cystic artery, cystic duct, hepatocystic window, and cystic plate. They then tested the system live in 50 pediatric patients aged 6 to 18 years. The surgical video was streamed 21 kilometers to a second hospital, where the algorithm analyzed each frame, drawing colored boxes around target structures -- blue for the cystic artery, green for the cystic duct, light gray for the hepatocystic window, and dark gray for the cystic plate. Only when all four appeared together did the system trigger an audible "detected" alarm, signaling that the three Strasberg criteria for CVS were fully met. Two expert surgeons, completely blinded to the AI's output, independently assessed CVS presence. The agreement was perfect, with a Cohen's kappa value of 1.0. In 38 of 50 cases, a complete CVS was detected. In the remaining 12 cases, both the algorithm and the surgeons agreed that one or more elements were missing -- most often the cystic artery or the cystic plate. No postoperative complications occurred, and the remote setup added no technical failures beyond three excluded transmission issues. The system was designed to prioritize high specificity, avoiding false alarms even if that meant occasional false negatives. The authors said the system was not meant to replace surgical judgment but to act as an extra set of eyes during a critical step in the operation. "We designed the algorithm to be very specific -- it only alarms when every safety element is visible," they explained. "That means it may miss some incomplete dissections on purpose, but it never cries wolf. For pediatric surgeons, who perform far fewer gallbladder operations than their adult colleagues, this kind of remote assistance could be a real safety net during training or in low-resource settings. We see it as a scalable tool, not a substitute for experience." This approach could help reduce the subjectivity of CVS identification, particularly in hospitals without specialized AI infrastructure or where pediatric caseloads are low. Because the algorithm runs on a remote server and communicates via standard teleconferencing, the operating room only requires an internet connection and a video feed. This design makes advanced intraoperative guidance potentially accessible to community hospitals and surgical training centers. The authors note that larger, multicenter prospective studies are still needed, especially to test the system in younger children under 5 years and in patients with severe inflammation or rare anatomical variations such as aberrant ducts or supernumerary arteries. If validated further, remote AI-assisted CVS detection could become a practical, scalable tool for improving surgical safety in pediatric laparoscopic cholecystectomy worldwide. ### References DOI 10.1136/wjps-2025-001125 Original Source URL https://doi.org/10.1136/wjps-2025-001125 About World Journal of Pediatric Surgery World Journal of Pediatric Surgery (WJPS), founded in 2018, is the open-access, peer-reviewed journal in pediatric surgery area. Sponsored by Zhejiang University and Children's Hospital Zhejiang University School of Medicine, and published by BMJ Group. WJPS aims to be a leading international platform for advances in pediatric surgical research and practice. Indexed in PubMed, ESCI, Scopus, CAS, DOAJ, and CSCD, WJPS achieved the latest Impact Factor (IF) of 1.3/Q3 and an estimate 2025 IF of 2.0.