Comprehensive deep learning-based assessment of living liver donor CT angiography: from vascular segmentation to volumetric analysis.

BACKGROUND: Precise preoperative assessment of liver vasculature and volume in living donor liver transplantation is essential for donor safety and recipient surgery. Traditional manual segmentation methods are being supplemented by deep learning (DL) models, which may offer more consistent and efficient volumetric evaluations. METHODS: This study analyzed living liver donors from Samsung Medical Center using preoperative CT angiography data between April 2022 and February 2023. A DL-based 3D residual U-Net model was developed and trained on segmented CT images to calculate the liver volume and segment vasculature, with its performance compared to traditional manual segmentation by surgeons and actual graft weight. RESULTS: The DL model achieved high concordance with manual methods, exhibiting Dice Similarity Coefficients of 0.94±0.01 for the right lobe and 0.91±0.02 for the left lobe. The liver volume estimates by DL model closely matched those of surgeons, with a mean discrepancy of 9.18 mL, and correlated more strongly with actual graft weights (R-squared value of 0.76 compared to 0.68 for surgeons). CONCLUSION: The DL model demonstrates potential as a reliable tool for enhancing preoperative planning in liver transplantation, offering consistency and efficiency in volumetric assessment. Further validation is required to establish its generalizability across various clinical settings and imaging protocols.

3D auto-segmentation of biliary structure of living liver donors using magnetic resonance cholangiopancreatography for enhanced preoperative planning.

BACKGROUND: This study aimed to develop an automated segmentation system for biliary structures using a deep learning model, based on data from magnetic resonance cholangiopancreatography (MRCP). MATERIALS AND METHODS: Living liver donors who underwent MRCP using the gradient and spin echo technique followed by three-dimensional modeling were eligible for this study. A three-dimensional residual U-Net model was implemented for the deep learning process. Data were divided into training and test sets at a 9:1 ratio. Performance was assessed using the dice similarity coefficient to compare the model's segmentation with the manually labeled ground truth. RESULTS: The study incorporated 250 cases. There was no difference in the baseline characteristics between the train set (n=225) and test set (n=25). The overall mean Dice Similarity Coefficient was 0.80±0.20 between the ground truth and inference result. The qualitative assessment of the model showed relatively high accuracy especially for the common bile duct (88%), common hepatic duct (92%), hilum (96%), right hepatic duct (100%), and left hepatic duct (96%), while the third-order branch of the right hepatic duct (18.2%) showed low accuracy. CONCLUSION: The developed automated segmentation model for biliary structures, utilizing MRCP data and deep learning techniques, demonstrated robust performance and holds potential for further advancements in automation.

Left lobe living donor liver transplantation using the resection and partial liver segment 2-3 transplantation with delayed total hepatectomy (RAPID) procedure in cirrhotic patients: First case report in Korea.

In liver transplantation, the primary concern is to ensure an adequate future liver remnant (FLR) volume for the donor, while selecting a graft of sufficient size for the recipient. The living donor-resection and partial liver segment 2-3 transplantation with delayed total hepatectomy (LD-RAPID) procedure offers a potential solution to expand the donor pool for living donor liver transplantation (LDLT). We report the first case involving a cirrhotic patient with autoimmune hepatitis and hepatocellular carcinoma, who underwent left lobe LDLT using the LD-RAPID procedure. The living liver donor (LLD) underwent a laparoscopic left hepatectomy, including middle hepatic vein. The resection on the recipient side was an extended left hepatectomy, including the middle hepatic vein orifice and caudate lobe. At postoperative day 7, a computed tomography scan showed hypertrophy of the left graft from 320 g to 465 mL (i.e., a 45.3% increase in graft volume body weight ratio from 0.60% to 0.77%). After a 7-day interval, the diseased right lobe was removed in the second stage surgery. The LD-RAPID procedure using left lobe graft allows for the use of a small liver graft or small FLR volume in LLD in LDLT, which expands the donor pool to minimize the risk to LLD by enabling the donation of a smaller liver portion.

Automated 3D liver segmentation from hepatobiliary phase MRI for enhanced preoperative planning.

Recent advancements in deep learning have facilitated significant progress in medical image analysis. However, there is lack of studies specifically addressing the needs of surgeons in terms of practicality and precision for surgical planning. Accurate understanding of anatomical structures, such as the liver and its intrahepatic structures, is crucial for preoperative planning from a surgeon's standpoint. This study proposes a deep learning model for automatic segmentation of liver parenchyma, vascular and biliary structures, and tumor mass in hepatobiliary phase liver MRI to improve preoperative planning and enhance patient outcomes. A total of 120 adult patients who underwent liver resection due to hepatic mass and had preoperative gadoxetic acid-enhanced MRI were included in the study. A 3D residual U-Net model was developed for automatic segmentation of liver parenchyma, tumor mass, hepatic vein (HV), portal vein (PV), and bile duct (BD). The model's performance was assessed using Dice similarity coefficient (DSC) by comparing the results with manually delineated structures. The model achieved high accuracy in segmenting liver parenchyma (DSC 0.92 ± 0.03), tumor mass (DSC 0.77 ± 0.21), hepatic vein (DSC 0.70 ± 0.05), portal vein (DSC 0.61 ± 0.03), and bile duct (DSC 0.58 ± 0.15). The study demonstrated the potential of the 3D residual U-Net model to provide a comprehensive understanding of liver anatomy and tumors for preoperative planning, potentially leading to improved surgical outcomes and increased patient safety.

Graft-recipient-weight ratio and lowered immunosuppression is important for the success of adult liver retransplantation.

This study analyzed the risk of liver retransplantation and factors related to better outcome. Adult liver transplantations performed during 1996-2021 were included. Comparison between first transplantation and retransplantation were performed. Among retransplantation cases, comparison between whole liver and partial liver graft was performed. Multivariable Cox analyses for analyzing risk factors for primary graft and overall patient survival were performed for the entire cohort as well as the subgroup of patients with retransplantation. A total 2237 transplantations from 2135 adults were included and 103 cases were retransplantation. A total of 44 cases (42.7%) were related to acute graft dysfunction while 59 cases (57.3%) were related to subacute or chronic graft dysfunction. Retransplantation was related poor primary graft (HR 3.439, CI 2.230-5.304, P < 0.001) and overall patient survival. (HR 2.905, CI 2.089-4.040, P < 0.001) Among retransplantations, mean serum FK506 trough level ≥ 9 ng/mL was related to poor primary graft (HR 3.692, CI 1.288-10.587, P = 0.015) and overall patient survival. (HR 2.935, CI 1.195-7.211, P = 0.019) Graft-recipient-weight ratio under 1.0% was related to poor overall patient survival in retransplantations. (HR 3.668, CI 1.150-11.698, P = 0.028). Retransplantation can be complicated with poor graft and patient survival compared to first transplantation, especially when the graft size is relatively small. Lowering the FK506 trough level during the first month can be beneficial for outcome.

ChatGPT goes to the operating room: evaluating GPT-4 performance and its potential in surgical education and training in the era of large language models.

Purpose: This study aimed to assess the performance of ChatGPT, specifically the GPT-3.5 and GPT-4 models, in understanding complex surgical clinical information and its potential implications for surgical education and training. Methods: The dataset comprised 280 questions from the Korean general surgery board exams conducted between 2020 and 2022. Both GPT-3.5 and GPT-4 models were evaluated, and their performances were compared using McNemar test. Results: GPT-3.5 achieved an overall accuracy of 46.8%, while GPT-4 demonstrated a significant improvement with an overall accuracy of 76.4%, indicating a notable difference in performance between the models (P < 0.001). GPT-4 also exhibited consistent performance across all subspecialties, with accuracy rates ranging from 63.6% to 83.3%. Conclusion: ChatGPT, particularly GPT-4, demonstrates a remarkable ability to understand complex surgical clinical information, achieving an accuracy rate of 76.4% on the Korean general surgery board exam. However, it is important to recognize the limitations of large language models and ensure that they are used in conjunction with human expertise and judgment.

Improved recurrence-free survival in patients with HCC with post-transplant plasma exchange.

Total plasma exchange (TPE) can play a role in cancer treatment by eliminating immune checkpoint inhibitors. This study investigated whether TPE improved oncological outcomes in patients with HCC who underwent ABO-incompatible living donor liver transplantation (LT). The study included 152 patients who underwent ABO-incompatible living donor LT for HCC between 2010 and 2021 at Samsung Medical Center. Overall survival was analyzed using the Kaplan-Meier curve, whereas HCC-specific recurrence-free survival (RFS) was analyzed using the cumulative incidence curve after propensity score matching. Cox regression and competing risks subdistribution hazard models were used to identify the risk factors associated with overall survival and HCC-specific RFS, respectively. The propensity score matching resulted in 54 matched pairs, grouped according to whether they underwent postoperative TPE [post-transplant TPE(+)] or not [post-transplant TPE(-)]. The 5-year HCC-specific RFS cumulative incidence was superior in the post-transplant TPE (+) group [12.5% (95% CI: 3.1%-21.9%)] compared with the post-transplant TPE(-) group [38.1% (95% CI: 24.4%-51.8%), p = 0.005]. In subgroup analysis for patients with microvascular invasion and those beyond the Milan criteria, the post-transplant TPE(+) group showed significantly superior HCC-specific survival. The multivariable analysis also showed that postoperative TPE had a protective effect on HCC-specific RFS (HR = 0.26, 95% CI: 0.10-0.64, p = 0.004) and that the more post-transplant TPE was performed, the better RFS was observed (HR = 0.71, 95% CI: 0.55-0.93, p = 0.012). Post-transplant TPE was found to improve RFS after ABO-incompatible living donor LT for HCC, particularly in advanced cases with microvascular invasion and beyond Milan criteria. These findings suggest that TPE may have a potential role in improving oncological outcomes in patients with HCC undergoing LT.

First and second kidney transplantations from living donors offer comparable outcomes for patients.

Background: Increasingly many patients are being relisted for repeat kidney transplantation due to longer survival times after transplantation. This study compared the outcomes of second living donor kidney transplantations (LDKTs) with those of first LDKTs. Methods: Data were collected retrospectively for 1,429 LDKTs performed from 1995 to 2020 at Samsung Medical Center. The demographics of recipients and donors, immunologic factors, and outcomes of second LDKTs were compared to those of first LDKTs. Results: Among 1,429 cases of LDKT, 1,355 were first LDKT cases and 74 were second LDKT cases. Basic demographic data were comparable for the two groups of patients. The 5- and 10-year graft survival rates were 94% and 84% for first LDKTs and 96% and 86% for second LDKTs, respectively, with neither difference statistically significant (P=0.399). The 5- and 10-year patient survival rates were 98% and 94% for the first and 96% and 93% for the second LDKTs, respectively; neither difference was statistically significant (P=0.766). Multivariate analysis confirmed that a history of previous transplantation was not a statistically significant risk factor for graft loss (hazard ratio [HR], 0.83; P=0.677) or patient death (HR, 1.68; P=0.396). Conclusions: These results indicate that repeat kidney transplantation from living donors is a reasonable choice for patients who have experienced graft loss.

Microsatellite Instability and Effectiveness of Adjuvant Treatment in pT1N1 Gastric Cancer: A Multicohort Study.

BACKGROUND: Microsatellite status is a prognostic biomarker in advanced gastric cancer. This retrospective study aimed to investigate the usefulness of microsatellite status in predicting prognosis and response to adjuvant treatment in pT1N1 gastric cancer. PATIENTS AND METHODS: Among 875 patients who underwent radical gastrectomy for pT1N1 gastric cancer at two tertiary hospitals, 838 with available microsatellite instability (MSI) data were included and classified into two groups according to microsatellite status: microsatellite stable (MSS) and MSI-high (MSI-H). Recurrence-free survival rate and risk factors for tumor recurrence were analyzed. RESULTS: Of 838 gastric cancer patients, 100 (11.9%) were MSI-H and 307 (36.6%) received adjuvant treatment. During median follow-up of 70 months, 42 (5.0%) patients experienced gastric cancer recurrence; hematogenous recurrences were the most common (45.2%). Recurrence-free survival was similar in the MSS and MSI-H groups (p = 0.27), and adjuvant treatment did not show an oncological benefit over surgery alone for pT1N1 gastric cancer (p = 0.53). On univariate analysis, age, operation period, and Lauren classification were significantly associated with tumor recurrence, while adjuvant treatment and MSI status were not associated with tumor recurrence. On multivariate analysis, MSI status was not associated with tumor recurrence, and adjuvant treatment worsened the tumor recurrence risk [hazard ratio (HR) 2.373, 95% confidence interval (CI) 1.125-5.006, p = 0.023). CONCLUSION: MSI status may not be a prognostic factor for tumor recurrence or a predictor of response to adjuvant treatment in pT1N1 gastric cancer patients. Considering that the effect of adjuvant treatment to decrease the risk of tumor recurrence is not clear, it may not be indicated in pT1N1 patients.

Characteristics and Distribution of Surgical Diseases in North Korean Research Papers Published between 2006 and 2017.

BACKGROUND: Little is known about the surgical discipline in North Korea from the perspective of the outside world. This study aimed to examine the disease entities covered by articles published in the major medical journal in North Korea, "Surgery." METHODS: Content and frequency analyses of 2,132 articles published in "Surgery" from 2006 to 2017 were conducted. Two medical doctors who majored in surgery and anesthesiology perused the articles and compiled the diseases being elucidated in each article. The diseases described in each article were stratified into 13 surgical subspecialties. RESULTS: Articles from "Surgery," similar to articles from the Western surgical community, also covered a wide variety of surgical diseases from different subspecialties, and the number of publications continued to grow consistently. Moreover, a number of studies focused on the fields of orthopedics and general surgery dealing with benign diseases. Some articles focused on minimally invasive surgeries using laparoscopy. CONCLUSION: The studies published in the North Korean journal "Surgery" encompass various clinical areas, but their quality is unclear.