ABSTRACT
BACKGROUND: Hyperspectral imaging (HSI), combined with machine learning, can help to identify characteristic tissue signatures enabling automatic tissue recognition during surgery. This study aims to develop the first HSI-based automatic abdominal tissue recognition with human data in a prospective bi-center setting. METHODS: Data were collected from patients undergoing elective open abdominal surgery at two international tertiary referral hospitals from September 2020 to June 2021. HS images were captured at various time points throughout the surgical procedure. Resulting RGB images were annotated with 13 distinct organ labels. Convolutional Neural Networks (CNNs) were employed for the analysis, with both external and internal validation settings utilized. RESULTS: A total of 169 patients were included, 73 (43.2%) from Strasbourg and 96 (56.8%) from Verona. The internal validation within centers combined patients from both centers into a single cohort, randomly allocated to the training (127 patients, 75.1%, 585 images) and test sets (42 patients, 24.9%, 181 images). This validation setting showed the best performance. The highest true positive rate was achieved for the skin (100%) and the liver (97%). Misclassifications included tissues with a similar embryological origin (omentum and mesentery: 32%) or with overlaying boundaries (liver and hepatic ligament: 22%). The median DICE score for ten tissue classes exceeded 80%. CONCLUSION: To improve automatic surgical scene segmentation and to drive clinical translation, multicenter accurate HSI datasets are essential, but further work is needed to quantify the clinical value of HSI. HSI might be included in a new omics science, namely surgical optomics, which uses light to extract quantifiable tissue features during surgery.
Subject(s)
Deep Learning , Hyperspectral Imaging , Humans , Prospective Studies , Hyperspectral Imaging/methods , Male , Female , Middle Aged , Aged , Abdomen/surgery , Abdomen/diagnostic imaging , Surgery, Computer-Assisted/methodsABSTRACT
Duodenal stenosis is a condition that can be related to several diseases, being either intrinsic, such as neoplasm and inflammatory stenosis, or extrinsic, such as pancreatic pseudocyst, superior mesenteric artery syndrome, and foreign bodies. Current treatments range from endoscopic approaches, such as endoscopic resection and stent placement, to surgical approaches, including duodenal resection, pancreaticoduodenectomy, and gastrointestinal bypass. Minimally invasive robot-assisted surgery is gaining importance due to its potential to decrease surgical stress, intraoperative blood loss, and postoperative pain, while its instruments and 3D-vision facilitate fine dissection and intra-abdominal suturing, all leading to a reduced time to functional recovery and shorter hospital stay. We present a case of a 75-year-old female who underwent robotic D3 partial duodenal resection with primary side-to-side duodeno-jejunal anastomosis for a 5 cm adenoma with focal high-grade dysplasia.
Subject(s)
Laparoscopy , Robotic Surgical Procedures , Robotics , Female , Humans , Aged , Duodenum/surgery , Anastomosis, Surgical , PancreaticoduodenectomyABSTRACT
BACKGROUND: The cost-effectiveness of minimally invasive distal pancreatectomy (MIDP) is still a matter of debate. This study compares the cost-effectiveness of open (ODP), laparoscopic (LDP) and robotic distal pancreatectomy (RDP). METHODS: Pubmed, Web of Science and Cochrane Library databases were searched. Studies comparing cost-effectiveness of ODP and MIDP were included. RESULTS: A total of 1052 titles were screened and 16 articles were included in the study, 2431 patients in total. LDP resulted the most cost-efficient procedure, with a mean total cost of 14,682 ± 5665 and the lowest readmission rates. ODP had lower surgical procedure costs, 3867 ± 768 . RDP was the safest approach regarding hospital stay costs (5239 ± 1741 ), length of hospital stay, morbidity, clinically relevant pancreatic fistula and reoperations. CONCLUSION: In this meta-analysis MIDP resulted as the most cost-effective approach. LDP seems to be protective against high costs, but RDP seems to be safer.
