|
1
|
Shi J, Cui R, Wang Z, Yan Q, Ping L, Zhou H, Gao J, Fang C, Han X, Hua S, Wu W. Deep learning HRNet FCN for blood vessel identification in laparoscopic pancreatic surgery. NPJ Digit Med 2025; 8:235. [PMID: 40312536 PMCID: PMC12046043 DOI: 10.1038/s41746-025-01663-6] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/18/2024] [Accepted: 04/22/2025] [Indexed: 05/03/2025] Open
Abstract
Laparoscopic pancreatic surgery remains highly challenging due to the complexity of the pancreas and surrounding vascular structures, with risk of injuring critical blood vessels such as the Superior Mesenteric Vein (SMV)-Portal Vein (PV) axis and splenic vein. Here, we evaluated the High Resolution Network (HRNet)-Full Convolutional Network (FCN) model for its ability to accurately identify vascular contours and improve surgical safety. Using 12,694 images from 126 laparoscopic distal pancreatectomy (LDP) videos and 35,986 images from 138 Whipple procedure videos, the model demonstrated robust performance, achieving a mean Dice coefficient of 0.754, a recall of 85.00%, and a precision of 91.10%. By combining datasets from LDP and Whipple procedures, the model showed strong generalization across different surgical contexts and achieved real-time processing speeds of 11 frames per second during surgery process. These findings highlight HRNet-FCN's potential to recognize anatomical landmarks, enhance surgical precision, reduce complications, and improve laparoscopic pancreatic outcomes.
Collapse
Affiliation(s)
- Jile Shi
- Peking Union Medical College, Chinese Academy of Medical Sciences, 100730, Beijing, China
| | - Ruohan Cui
- Peking Union Medical College, Chinese Academy of Medical Sciences, 100730, Beijing, China
| | - Zhihong Wang
- Peking Union Medical College, Chinese Academy of Medical Sciences, 100730, Beijing, China
| | - Qi Yan
- School of Life Sciences, Tsinghua University, 100084, Beijing, China
| | - Lu Ping
- Peking Union Medical College, Chinese Academy of Medical Sciences, 100730, Beijing, China
| | - Hu Zhou
- Peking Union Medical College, Chinese Academy of Medical Sciences, 100730, Beijing, China
| | - Junyi Gao
- Peking Union Medical College, Chinese Academy of Medical Sciences, 100730, Beijing, China
| | - Chihua Fang
- Department of Hepatobiliary Surgery I, Zhujiang Hospital Southern Medical University, Guangzhou, 510280, China
| | - Xianlin Han
- Peking Union Medical College, Chinese Academy of Medical Sciences, 100730, Beijing, China
| | - Surong Hua
- Peking Union Medical College, Chinese Academy of Medical Sciences, 100730, Beijing, China.
| | - Wenming Wu
- Peking Union Medical College, Chinese Academy of Medical Sciences, 100730, Beijing, China.
| |
Collapse
|
|
2
|
Kjeseth T, Hagen RE, Edwin B, Lai X, Røsok BI, Tholfsen T, Sahakyan MA, Kleive D. Impact of pancreas transection site on incidence of pancreatic fistula after distal pancreatectomy: a propensity score matched study. HPB (Oxford) 2024; 26:1164-1171. [PMID: 38839509 DOI: 10.1016/j.hpb.2024.05.012] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 12/23/2023] [Revised: 04/24/2024] [Accepted: 05/20/2024] [Indexed: 06/07/2024]
Abstract
BACKGROUND Distal pancreatectomy (DP) is performed for lesions in the body and tail of the pancreas. The morbidity profile is considerable, mainly due to clinically relevant postoperative pancreatic fistula (CR-POPF). This study aims to investigate potential differences in CR-POPF related to transection site. METHODS An observational cohort study from a prospectively maintained database was performed. Subtotal distal pancreatectomy (SDP) was defined as transection over the superior mesenteric vein, and DP was defined as transection lateral to this point. Propensity score matching (PSM) in 1:1 fashion was applied based on demographical and perioperative variables. RESULTS Six hundred and six patients were included in the analysis (1997-2020). Four hundred twenty (69.3%) underwent DP, while 186 (30.7%) underwent SDP. The rate of CR-POPF was 19.3% after DP and 20.4% after SDP (p = 0.74). SDP was associated with older age (63.1 vs 60.1 years, p = 0.016), higher occurrence of ductal adenocarcinoma (37.1 vs 17.6%, p = 0.001) and more frequent use of neoadjuvant chemotherapy (3.8 vs 0.7%, p = 0.012). After PSM, 155 patients were left in each group. The difference in CR-POPF between DP and SDP remained statistically non-significant (20.6 vs 18.7%, p = 0.67). CONCLUSION This study found no difference in CR-POPF related to transection site during distal pancreatectomy.
Collapse
Affiliation(s)
- Trond Kjeseth
- Department of HPB Surgery, Oslo University Hospital, Rikshospitalet, Oslo, Norway.
| | - Rolf E Hagen
- Department of HPB Surgery, Oslo University Hospital, Rikshospitalet, Oslo, Norway
| | - Bjørn Edwin
- Department of HPB Surgery, Oslo University Hospital, Rikshospitalet, Oslo, Norway; The Intervention Centre, Oslo University Hospital, Oslo, Norway; Institute of Clinical Medicine, Medical Faculty, University of Oslo, Oslo, Norway
| | - Xiaoran Lai
- Oslo Centre for Biostatistics and Epidemiology, University of Oslo, Oslo, Norway
| | - Bård I Røsok
- Department of HPB Surgery, Oslo University Hospital, Rikshospitalet, Oslo, Norway
| | - Tore Tholfsen
- Department of HPB Surgery, Oslo University Hospital, Rikshospitalet, Oslo, Norway
| | - Mushegh A Sahakyan
- Department of HPB Surgery, Oslo University Hospital, Rikshospitalet, Oslo, Norway; The Intervention Centre, Oslo University Hospital, Oslo, Norway; Department of Surgery N1, Yerevan State Medical University after M. Heratsi, Yerevan, Armenia
| | - Dyre Kleive
- Department of HPB Surgery, Oslo University Hospital, Rikshospitalet, Oslo, Norway
| |
Collapse
|
|
3
|
Kjeseth T, Yaqub S, Edwin B, Kleive D, Sahakyan MA. Peri-firing compression in prevention of pancreatic fistula after distal pancreatectomy: A systematic review and a cohort study. Scand J Surg 2024; 113:73-79. [PMID: 37982224 DOI: 10.1177/14574969231211084] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/21/2023]
Abstract
BACKGROUND/AIMS Clinically relevant postoperative pancreatic fistula (CR-POPF) after distal pancreatectomy (DP) occurs in 20%-40% of patients and remains a leading cause of morbidity and increased healthcare cost in this patient group. Recently, several studies suggested decreased risk of CR-POPF with the use of peri-firing compression (PFC) technique. The aim of this report was to conduct a systematic review to get an overview of the current knowledge on the use of PFC in DP. In addition, our experience with PFC was presented. METHODS The systematic literature review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Also, 19 patients undergoing DP with the use of PFC at Oslo University Hospital were studied. The primary endpoint was incidence of CR-POPF. RESULTS Seven articles reporting a total of 771 patients were ultimately included in the systematic review. Only two of these were case-control studies examining outcomes in patients with and without PFC, while the rest were case series. These were heterogeneous in terms of staplers used, cartridge selection policy, and PFC technique. Both case-control studies reported significantly reduced CR- POPF incidence with PFC. Eight (21%) of our patients developed CR-POPF after DP with PFC. Only one patient developed CR-POPF among those with pancreatic transection site thickness ⩽1.5 cm. CONCLUSION Evidence on potential benefits of PFC in DP is limited in quantity and quality. Our findings suggest that the use of PFC does not lead to reduction in the incidence of CR-POPF. Yet, there might be a benefit from PFC when dealing with a thin pancreas.
Collapse
Affiliation(s)
- Trond Kjeseth
- Department of Hepato-Pancreato-Biliary Surgery Oslo University Hospital Rikshospitalet Sognsvannsveien 20 0372 Oslo Norway
- Department of Hepato-Pancreato-Biliary Surgery, Oslo University Hospital, Oslo, Norway
| | - Sheraz Yaqub
- Department of Hepato-Pancreato-Biliary Surgery, Oslo University Hospital, Oslo, Norway
- Institute of Clinical Medicine, University of Oslo, Oslo, Norway
| | - Bjørn Edwin
- Department of Hepato-Pancreato-Biliary Surgery, Oslo University Hospital, Oslo, Norway
- Institute of Clinical Medicine, University of Oslo, Oslo, Norway
- The Intervention Centre, Oslo University Hospital-Rikshospitalet, Oslo, Norway
- Department of Research & Development, Division of Emergencies and Critical Care, Oslo University Hospital, Oslo, Norway
| | - Dyre Kleive
- Department of Hepato-Pancreato-Biliary Surgery, Oslo University Hospital, Oslo, Norway
| | - Mushegh A Sahakyan
- The Intervention Centre, Oslo University Hospital-Rikshospitalet, Oslo, Norway
- Department of Hepato-Pancreato-Biliary Surgery, Oslo University Hospital, Oslo, Norway
- Department of Surgery N1, Yerevan State Medical University after M. Heratsi, Yerevan, Armenia
| |
Collapse
|
|
4
|
Bencini L, Minuzzo A. Distal pancreatectomy with or without radical approach, vascular resections and splenectomy: Easier does not always mean easy. World J Gastrointest Surg 2023; 15:1020-1032. [PMID: 37405088 PMCID: PMC10315131 DOI: 10.4240/wjgs.v15.i6.1020] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 01/05/2023] [Revised: 01/24/2023] [Accepted: 04/17/2023] [Indexed: 06/15/2023] Open
Abstract
Because distal pancreatectomy (DP) has no reconstructive steps and less frequent vascular involvement, it is thought to be the easier counterpart of pancreaticoduodenectomy. This procedure has a high surgical risk and the overall incidences of perioperative morbidity (mainly pancreatic fistula), and mortality are still high, in addition to the challenges that accompany delayed access to adjuvant therapies (if any) and prolonged impairment of daily activities. Moreover, surgery to remove malignancy of the body or tail of the pancreas is associated with poor long-term oncological outcomes. From this perspective, new surgical approaches, and aggressive techniques, such as radical antegrade modular pancreato-splenectomy and DP with celiac axis resection, could lead to improved survival in those affected by more locally advanced tumors. Conversely, minimally invasive approaches such as laparoscopic and robotic surgeries and the avoidance of routine concomitant splenectomy have been developed to reduce the burden of surgical stress. The purpose of ongoing surgical research has been to achieve significant reductions in perioperative complications, length of hospital stays and the time between surgery and the beginning of adjuvant chemotherapy. Because a dedicated multidisciplinary team is crucial to pancreatic surgery, hospital and surgeon volumes have been confirmed to be associated with better outcomes in patients affected by benign, borderline, and malignant diseases of the pancreas. The purpose of this review is to examine the state of the art in distal pancreatectomies, with a special focus on minimally invasive approaches and oncological-directed techniques. The widespread reproducibility, cost-effectiveness and long-term results of each oncological procedure are also taken into deep consideration.
Collapse
Affiliation(s)
- Lapo Bencini
- Oncology and Robotic Surgery, Careggi Main Regional and University Hospital, Florence 50131, Italy
| | - Alessio Minuzzo
- Oncology and Robotic Surgery, Careggi Main Regional and University Hospital, Florence 50131, Italy
| |
Collapse
|
|
5
|
van Ramshorst TM, Edwin B, Han HS, Nakamura M, Yoon YS, Ohtsuka T, Tholfsen T, Besselink MG, Abu Hilal M. Learning curves in laparoscopic distal pancreatectomy: a different experience for each generation. Int J Surg 2023; 109:1648-1655. [PMID: 37144678 PMCID: PMC10389345 DOI: 10.1097/js9.0000000000000408] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/31/2022] [Accepted: 04/06/2023] [Indexed: 05/06/2023]
Abstract
BACKGROUND Learning curves of laparoscopic distal pancreatectomy (LDP) are mostly based on 'self-taught' surgeons who acquired sufficient proficiency largely through self-teaching. No learning curves have been investigated for 'trained' surgeons who received training and built on the experience of the 'self-taught' surgeons. This study compared the learning curves and outcome of LDP between 'self-taught' and 'trained' surgeons in terms of feasibility and proficiency using short-term outcomes. MATERIALS AND METHODS Data of consecutive patients with benign or malignant disease of the left pancreas who underwent LDP by four 'self-taught' and four 'trained' surgeons between 1997 and 2019 were collected, starting from the first patient operated by a contributing surgeon. Risk-adjusted cumulative sum (RA-CUSUM) analyses were performed to determine phase-1 feasibility (operative time) and phase-2 proficiency (major complications) learning curves. Outcomes were compared based on the inflection points of the learning curves. RESULTS The inflection points for the feasibility and proficiency learning curves were 24 and 36 procedures for 'trained' surgeons compared to 64 and 85 procedures for 'self-taught' surgeons, respectively. In 'trained' surgeons, operative time was reduced after completion of the learning curves (230.5-203 min, P= 0.028). In 'self-taught' surgeons, operative time (240-195 min, P ≤0.001), major complications (20.6-7.8%, P= 0.008), and length of hospital stay (9-5 days, P ≤0.001) reduced after completion of the learning curves. CONCLUSION This retrospective international cohort study showed that the feasibility and proficiency learning curves for LDP of 'trained' surgeons were at least halved as compared to 'self-taught' surgeons.
Collapse
Affiliation(s)
- Tess M.E. van Ramshorst
- Department of General Surgery, Istituto Ospedaliero Fondazione Poliambulanza, Brescia, Italy
- Amsterdam UMC, University of Amsterdam, Department of Surgery
- Cancer Center Amsterdam, Amsterdam, The Netherlands
| | - Bjørn Edwin
- The Intervention Centre and Department of HPB Surgery, Oslo University Hospital, also Institute of Medicine, University of Oslo
| | - Ho-Seong Han
- Department of Surgery, Seoul National University Bundang Hospital, Seoul National University College of Medicine, Seoul, Republic of Korea
| | - Masafumi Nakamura
- Department of Surgery and Oncology, Graduate School of Medical Sciences, Kyushu University, Fukuoka
| | - Yoo-Seok Yoon
- Department of Surgery, Seoul National University Bundang Hospital, Seoul National University College of Medicine, Seoul, Republic of Korea
| | - Takao Ohtsuka
- First Department of Surgery, Kagoshima University School of Medicine, Kagoshima, Japan
| | - Tore Tholfsen
- Department of Hepato-Pancreato-Biliary Surgery, Rikshospitalet, Oslo University Hospital, Oslo, Norway
| | - Marc G. Besselink
- Amsterdam UMC, University of Amsterdam, Department of Surgery
- Cancer Center Amsterdam, Amsterdam, The Netherlands
| | - Mohammad Abu Hilal
- Department of General Surgery, Istituto Ospedaliero Fondazione Poliambulanza, Brescia, Italy
- Department of Surgery, University Hospital Southampton National Health Service, Southampton, Hampshire, United Kingdom
| |
Collapse
|
|
6
|
Waage A, Vinge-Holmquist O, Labori KJ, Paulsen V, Aabakken L, Lenz H, Felix Magnus HC, Tholfsen T, Hauge T. Tailored surgery in chronic pancreatitis after implementation of a multidisciplinary team assessment; a prospective observational study. HPB (Oxford) 2022; 24:2157-2166. [PMID: 36272955 DOI: 10.1016/j.hpb.2022.09.007] [Citation(s) in RCA: 4] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 07/07/2022] [Revised: 09/15/2022] [Accepted: 09/28/2022] [Indexed: 01/24/2023]
Abstract
INTRODUCTION Optimal management of chronic pancreatitis involves several specialties. Selection of patients for surgery may benefit from evaluation by a multidisciplinary team (MDT), similar to cancer care. The aim of this study was to evaluate outcomes in patients selected for surgery after MDT decision. METHODS A prospective, observational study of consecutive patients operated for pain due to chronic pancreatitis after implementation of a MDT. The main outcome was Quality of life (QoL) assessed by EORTC-QLQ C30 and pain relief in patients followed >3 months. Complications were registered and predictive factors for pain relief analyzed. RESULTS Of 269 patients evaluated by the MDT, 60 (22%) underwent surgery. Postoperative surgical complications occurred in five patients (8.3%) and reoperation within 30 days in two. There was no 90-days mortality. Complete or partial pain relief was achieved in 44 of 50 patients followed >3 months (88%). Preoperative duration of pain predicted lower probability of success. Postoperative improvement in QoL was most prominent for pain, appetite and nausea. CONCLUSIONS After MDT evaluation, one in five patients was selected for surgery. Pain relief was obtained in a majority of patients with improved QoL. A tailored approach through a MDT seems warranted and efficient.
Collapse
Affiliation(s)
- Anne Waage
- Department of Hepato-Pancreato-Biliary Surgery, Oslo University Hospital, Oslo, Norway.
| | - Olof Vinge-Holmquist
- Department of Digestive Surgery, St Olav's University Hospital, Trondheim, Norway; Institute of Clinical Medicine, University of Oslo, Oslo, Norway
| | - Knut J Labori
- Department of Hepato-Pancreato-Biliary Surgery, Oslo University Hospital, Oslo, Norway; Institute of Clinical Medicine, University of Oslo, Oslo, Norway
| | - Vemund Paulsen
- Department of Transplantation Medicine, Oslo University Hospital, Oslo, Norway
| | - Lars Aabakken
- Institute of Clinical Medicine, University of Oslo, Oslo, Norway; Department of Transplantation Medicine, Oslo University Hospital, Oslo, Norway
| | - Harald Lenz
- Department of Anesthesiology, Division of Emergencies and Critical Care, Oslo University Hospital, Oslo, Norway
| | - Henrik C Felix Magnus
- Department of Hepato-Pancreato-Biliary Surgery, Oslo University Hospital, Oslo, Norway
| | - Tore Tholfsen
- Department of Hepato-Pancreato-Biliary Surgery, Oslo University Hospital, Oslo, Norway
| | - Truls Hauge
- Institute of Clinical Medicine, University of Oslo, Oslo, Norway; Department of Gastroenterology, Oslo University Hospital, Oslo, Norway.
| |
Collapse
|