Background/Purpose
Current conventional algorithms used for 3‐dimensional simulation in virtual hepatectomy still have difficulties distinguishing the portal vein (PV) and hepatic vein (HV). The accuracy of these algorithms was compared with a new deep‐learning based algorithm (DLA) using artificial intelligence.
Methods
A total of 110 living liver donor candidates until 2017, and 46 donor candidates until 2019 were allocated to the training group and validation groups for the DLA, respectively. All PV or HV branches were labeled based on Couinaud's segment classification and the Brisbane 2000 Terminology by hepato‐biliary surgeons. Misclassified and missing branches were compared between a conventional tracking‐based algorithm (TA) and DLA in the validation group.
Results
The sensitivity, specificity, and Dice coefficient for the PV were 0.58, 0.98, and 0.69 using the TA; and 0.84, 0.97, and 0.90 using the DLA (P < .001, excluding specificity); and for the HV, 0.81, 087, and 0.83 using the TA; and 0.93, 0.94 and 0.94 using the DLA (P < .001 to P = .001). The DLA exhibited greater accuracy than the TA.
Conclusion
Compared with the TA, artificial intelligence enhanced the accuracy of extraction of the PV and HVs in computed tomography.
Purpose Among the variations of the right hepatic artery (RHA), the identification of an aberrant RHA arising from the gastroduodenal artery (GDA) is vital for avoiding damage to the RHA during surgery, since ligation of the GDA is necessary during pancreaticoduodenectomy (PD). However, this variation is not frequently reported. The purpose of this study was to focus on an aberrant RHA arising from the GDA, which was not noted in the classifications reported by Michels and Hiatt. Methods A total of 574 patients undergoing a PD between Jan 2001 and Dec 2015 at a tertiary care hospital in Switzerland (n = 366) and between Jan 2009 and May 2015 at a hospital in Japan (n = 208) were included in the analysis. Of these, preoperative CT angiography or/and MRI angiography findings were available for 532 patients. We retrospectively analyzed the hepatic artery variations, patient demographics, and surgical outcomes. Results Among the 532 patients who received a PD, an RHA originating from the GDA was observed in 19 cases (3.5%). Eleven patients (2.1%) had both an aberrant RHA and an aberrant left hepatic artery (LHA) (Hiatt Type 4). Six patients (1.2%) had a replaced CHA arising from the SMA (Hiatt Type 5). We could, therefore, correctly identify the aberration in all cases. Conclusions We observed rarely reported but important aberrant RHA variations arising from the GDA. To prevent injury during PD in patients with this type of aberrant RHA, intensive preparations using CT and/or MRI imaging before surgery and intraoperative liver Doppler ultrasonography are considered to be essential.
Objective: Noninvasive intraductal papillary mucinous neoplasms (IPMNs) theoretically do not metastasize. The purpose of this study is to preoperatively distinguish invasive carcinomas associated with IPMN from noninvasive IPMN.Methods: A total of 131 patients who underwent surgical resection for IPMN were retrospectively analyzed to identify the predictors of invasive carcinoma, based on the International Association of Pancreatology Consensus Guidelines.Results: Of the 131 patients, 29 (22%) had invasive carcinomas and 102 (78%) had noninvasive IPMN. An enhancing mural nodule (MN) greater than or equal to 5 mm, obstructive jaundice, an abrupt change in the caliber of the pancreatic duct (PD) with distal pancreatic atrophy, and lymphadenopathy were the significant predictors of invasive carcinoma in univariate analysis. The optimal cutoff value for the size of the enhancing MN to differentiate invasive carcinoma was 13 mm. In multivariate analysis, enhancing MN greater than or equal to 13 mm, obstructive jaundice, and an abrupt change in the PD caliber were the independent predictors. When all these factors were absent, only 17% were invasive carcinomas.
Conclusions:Enhancing MN greater than or equal to 13 mm, obstructive jaundice, and an abrupt change in the PD caliber were predictive factors for invasive carcinoma. Systematic lymph node dissection may be omitted when a high-risk patient has none of these factors.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.