Purpose: This study aimed to develop and evaluate machine learning models, specifically CatBoost and extreme gradient boosting (XGBoost), for diagnosing lower urinary tract symptoms (LUTS) in male patients. The objective is to differentiate between bladder outlet obstruction (BOO) and detrusor underactivity (DUA) using a comprehensive dataset that includes patient-reported outcomes, uroflowmetry measurements, and ultrasound-derived features. Methods: The dataset used in this study was collected from male patients aged 40 and older who presented with LUTS and sought treatment at the urology department of Samsung Medical Center. We developed and trained CatBoost and XGBoost models using this dataset. These models incorporated features like prostate size, voiding parameters, and responses from questionnaires. Their performance was assessed using standard metrics such as accuracy, precision, recall, F1-score, and area under the receiver operating characteristic curve (AUROC). Results: The results indicated that the CatBoost models displayed greater sensitivity, rendering them effective for initial screenings by accurately identifying true positive cases. Conversely, the XGBoost models showed higher specificity and precision, making them more suitable for confirming diagnoses and reducing false positives. In terms of overall performance for both BOO and DUA, XGBoost surpassed CatBoost, achieving an AUROC of 0.826 and 0.819, respectively. Conclusions Integrating these machine learning models into the diagnostic workflow for LUTS can significantly enhance clinical decision-making by offering noninvasive, cost-effective, and patient-friendly diagnostic alternatives. The combined application of CatBoost and XGBoost models has the potential to improve diagnostic accuracy and provide customized treatment plans for patients, ultimately leading to better clinical outcomes.

Purpose: This study aimed to develop and evaluate machine learning models, specifically CatBoost and extreme gradient boosting (XGBoost), for diagnosing lower urinary tract symptoms (LUTS) in male patients. The objective is to differentiate between bladder outlet obstruction (BOO) and detrusor underactivity (DUA) using a comprehensive dataset that includes patient-reported outcomes, uroflowmetry measurements, and ultrasound-derived features. Methods: The dataset used in this study was collected from male patients aged 40 and older who presented with LUTS and sought treatment at the urology department of Samsung Medical Center. We developed and trained CatBoost and XGBoost models using this dataset. These models incorporated features like prostate size, voiding parameters, and responses from questionnaires. Their performance was assessed using standard metrics such as accuracy, precision, recall, F1-score, and area under the receiver operating characteristic curve (AUROC). Results: The results indicated that the CatBoost models displayed greater sensitivity, rendering them effective for initial screenings by accurately identifying true positive cases. Conversely, the XGBoost models showed higher specificity and precision, making them more suitable for confirming diagnoses and reducing false positives. In terms of overall performance for both BOO and DUA, XGBoost surpassed CatBoost, achieving an AUROC of 0.826 and 0.819, respectively. Conclusions Integrating these machine learning models into the diagnostic workflow for LUTS can significantly enhance clinical decision-making by offering noninvasive, cost-effective, and patient-friendly diagnostic alternatives. The combined application of CatBoost and XGBoost models has the potential to improve diagnostic accuracy and provide customized treatment plans for patients, ultimately leading to better clinical outcomes.

Purpose: This study aimed to develop and evaluate machine learning models, specifically CatBoost and extreme gradient boosting (XGBoost), for diagnosing lower urinary tract symptoms (LUTS) in male patients. The objective is to differentiate between bladder outlet obstruction (BOO) and detrusor underactivity (DUA) using a comprehensive dataset that includes patient-reported outcomes, uroflowmetry measurements, and ultrasound-derived features. Methods: The dataset used in this study was collected from male patients aged 40 and older who presented with LUTS and sought treatment at the urology department of Samsung Medical Center. We developed and trained CatBoost and XGBoost models using this dataset. These models incorporated features like prostate size, voiding parameters, and responses from questionnaires. Their performance was assessed using standard metrics such as accuracy, precision, recall, F1-score, and area under the receiver operating characteristic curve (AUROC). Results: The results indicated that the CatBoost models displayed greater sensitivity, rendering them effective for initial screenings by accurately identifying true positive cases. Conversely, the XGBoost models showed higher specificity and precision, making them more suitable for confirming diagnoses and reducing false positives. In terms of overall performance for both BOO and DUA, XGBoost surpassed CatBoost, achieving an AUROC of 0.826 and 0.819, respectively. Conclusions Integrating these machine learning models into the diagnostic workflow for LUTS can significantly enhance clinical decision-making by offering noninvasive, cost-effective, and patient-friendly diagnostic alternatives. The combined application of CatBoost and XGBoost models has the potential to improve diagnostic accuracy and provide customized treatment plans for patients, ultimately leading to better clinical outcomes.

Background/Aims: Papillary adenocarcinoma is classified to differentiated-type gastric cancer and is indicated for endoscopic submucosal dissection. However, due to its rare nature, there are limited studies on it. The purpose of this study was to determine the outcome of endoscopic submucosal dissection in patients with papillary-type early gastric cancer and to find the risk factors of lymph node metastasis. Methods: Patients diagnosed with papillary-type early gastric cancer at eight medical centers, who underwent endoscopic submucosal dissection or surgical treatment, were retrospectively reviewed. The clinical results and long-term outcomes of post-endoscopic submucosal dissection were evaluated, and the risk factors of lymph node metastasis in the surgery group were analyzed. Results: One-hundred and seventy-six patients with papillary-type early gastric cancer were enrolled: 44.9% (n=79) in the surgery group and 55.1% (n=97) in the endoscopic submucosal dissection group. As a result of endoscopic submucosal dissection, the en bloc resection and curative resection rates were 91.8% and 86.6%, respectively. The procedure-related complication rate was 4.1%, and local recurrence occurred in 3.1% of patients. Submucosal invasion (odds ratio, 3.735; 95% confidence interval, 1.026 to 12.177; p=0.047) and lymphovascular invasion (odds ratio, 7.636; 95% confidence interval, 1.730 to 22.857; p=0.004) were the risk factors of lymph node metastasis in papillary-type early gastric cancer patients. Conclusions The clinical results of endoscopic submucosal dissection in papillary-type early gastric cancer were relatively favorable, and endoscopic submucosal dissection is considered safe if appropriate indications are confirmed by considering the risk of lymph node metastasis.

Purpose: High-dose chemotherapy followed by autologous stem cell transplantation (ASCT) is the standard management for relapsed or high-risk non-Hodgkin’s lymphoma (NHL). We reported the busulfan, melphalan, and etoposide (BuME) conditioning regimen was effective in patients with relapsed or high-risk NHL. Moreover, the busulfan, cyclophosphamide, and etoposide (BuCE) conditioning regimen has been used widely in ASCT for NHL. Therefore, based on these encouraging results, this randomized phase II multicenter trial compared the outcomes of BuME and BuCE as conditioning therapies for ASCT in patients with NHL. Materials and Methods: Patients were randomly assigned to receive either BuME (n=36) or BuCE (n=39). The BuME regimen was comprised of busulfan (3.2 mg/kg/day, intravenously) administered on days –7, –6, and –5, etoposide (400 mg/m2 intravenously) on days –5 and –4, and melphalan (50 mg/m2/day intravenously) on days –3 and –2. The BuCE regimen was comprised of busulfan (3.2 mg/kg/day intravenously) on days –7, –6, and –5, etoposide (400 mg/m2/day intravenously) on days –5 and –4, and cyclophosphamide (50 mg/kg/day intravenously) on days –3 and –2. The primary endpoint was 2-year progression-free survival (PFS). Results: Seventy-five patients were enrolled. Eleven patients (30.5%) in the BuME group and 13 patients (33.3%) in the BuCE group had disease progression or died. The 2-year PFS rate was 65.4% in the BuME group and 60.6% in the BuCE group (p=0.746). There were no non-relapse mortalities within 100 days after transplantation. Conclusion There were no significant differences in PFS between the two groups. Therefore, busulfan-based conditioning regimens, BuME and BuCE, may be important treatment substitutes for the BCNU-containing regimens.

Purpose: The programmed death protein 1 (PD-1) pathway is the critical mechanism in development of hepatocellular carcinoma (HCC). The present study analyzed the prognostic impact of pretransplant serum soluble PD-1 (sPD-1) concentration and α-FP–des-γ- –tumor volume (ADV) score in patients with previously untreated HCC undergone liver transplantation (LT). Methods: This retrospective single-center study enrolled 100 patients with HCC who underwent living donor LT from 2010 to 2016. Concentrations of sPD-1 were measured in stored serum samples. Results: Receiver operating characteristic curve analysis of 2-year tumor recurrence resulted in an sPD-1 cutoff of 177.1 µg/mL, which was associated with higher rates of tumor recurrence (P = 0.022), but not with overall patient survival (P = 0.460). The derived cutoff for pretransplant ADV score was 5.4log, which was associated with higher tumor recurrence rate (P < 0.001) and lower overall patient survival rate (P < 0.001). Both sPD-1 of >177.1 µg/mL (hazard ratio [HR], 2.26; P = 0.020) and pretransplant ADV score of >5.4log (HR, 3.56; P < 0.001) were independent risk factors for posttransplant HCC recurrence. The combination of these 2 factors enabled the stratification of patients into 3 groups, with groups having 0, 1, and 2 risk factors differing significantly in the prognosis of tumor recurrence (P < 0.001) and overall patient survival (P = 0.006). Conclusion Both sPD-1 concentration and ADV score have prognostic impacts in patients who underwent LT for untreated HCCs. These factors, both individually and combined, can help in predicting posttransplant prognosis.

Background and Objectives: Identifying patients with high bleeding risk (HBR) is important when making decisions for antiplatelet therapy strategy. This study evaluated the impact of ticagrelor monotherapy after 3-month dual antiplatelet therapy (DAPT) according to HBR in acute coronary syndrome (ACS) patients treated with drug eluting stents (DESs). Methods: In this post-hoc analysis of the TICO trial, HBR was defined by 2 approaches: meeting Academic Research Consortium for HBR (ARC-HBR) criteria or Predicting Bleeding Complications in Patients Undergoing Stent Implantation and Subsequent DAPT (PRECISEDAPT) score ≥25. The primary outcome was a 3–12 months net adverse clinical event (composite of major bleeding and adverse cardiac and cerebrovascular events). Results: Of the 2,980 patients without adverse events during the first 3 months after DES implantation, 453 (15.2%) were HBR by ARC-HBR criteria and 504 (16.9%) were HBR by PRECISE-DAPT score. The primary outcome rate was higher in HBR versus non-HBR patients (by ARC-HBR criteria: hazard ratio [HR], 2.87; 95% confidence interval [CI], 1.76– 4.69; p<0.001; by PRECISE-DAPT score: HR, 3.09; 95% CI, 1.92–4.98; p<0.001). Ticagrelor monotherapy after 3-month DAPT was associated with lower primary outcome rate than ticagrelor-based 12-month DAPT regardless of HBR by ARC-HBR criteria, with similar magnitudes of therapy effect for HBR and non-HBR patients (p-interaction=0.400). Results were consistent by PRECISE-DAPT score (p-interaction=0.178). Conclusions In ACS patients treated with DESs, ticagrelor monotherapy after 3-month DAPT was associated with lower rate of adverse clinical outcomes regardless of HBR, with similar magnitudes of therapy effect between HBR and non-HBR.Trial Registration: ClinicalTrials.gov Identifier: NCT02494895

Purpose: When splitting a liver for adult and pediatric graft recipients, the retained left medial section (S4) will undergo ischemic necrosis and the right trisection graft becomes an extended right liver (ERL) graft. We investigated the fates of the retained S4 and its prognostic impact in adult split liver transplantation (SLT) using an ERL graft. Methods: This was a retrospective analysis of 25 adult SLT recipients who received split ERL grafts. Results: The mean model for end-stage liver disease (MELD) score was 27.3 ± 10.9 and graft-recipient weight ratio (GRWR) was 1.98 ± 0.44. The mean donor age was 26.5 ± 7.7 years. The split ERL graft weight was 1,181.5 ± 252.8 g, which resulted in a mean GRWR of 1.98 ± 0.44. Computed tomography of the retained S4 parenchyma revealed small ischemic necrosis in 16 patients (64.0%) and large ischemic necrosis in the remaining 9 patients (36.0%). No S4-associated biliary complications were developed. The mean GRWR was 1.87 ± 0.43 in the 9 patients with large ischemic necrosis and 2.10 ± 0.44 in the 15 cases with small ischemic necrosis (P = 0.283). The retained S4 parenchyma showed gradual atrophy on follow-up imaging studies. The amount of S4 ischemic necrosis was not associated with graft (P = 0.592) or patient (P = 0.243) survival. A MELD score of >30 and pretransplant ventilator support were associated with inferior outcomes. Conclusion The amount of S4 ischemic necrosis is not a prognostic factor in adult SLT recipients, probably due to a sufficiently large GRWR.

Purpose: When splitting a liver for adult and pediatric graft recipients, the retained left medial section (S4) will undergo ischemic necrosis and the right trisection graft becomes an extended right liver (ERL) graft. We investigated the fates of the retained S4 and its prognostic impact in adult split liver transplantation (SLT) using an ERL graft. Methods: This was a retrospective analysis of 25 adult SLT recipients who received split ERL grafts. Results: The mean model for end-stage liver disease (MELD) score was 27.3 ± 10.9 and graft-recipient weight ratio (GRWR) was 1.98 ± 0.44. The mean donor age was 26.5 ± 7.7 years. The split ERL graft weight was 1,181.5 ± 252.8 g, which resulted in a mean GRWR of 1.98 ± 0.44. Computed tomography of the retained S4 parenchyma revealed small ischemic necrosis in 16 patients (64.0%) and large ischemic necrosis in the remaining 9 patients (36.0%). No S4-associated biliary complications were developed. The mean GRWR was 1.87 ± 0.43 in the 9 patients with large ischemic necrosis and 2.10 ± 0.44 in the 15 cases with small ischemic necrosis (P = 0.283). The retained S4 parenchyma showed gradual atrophy on follow-up imaging studies. The amount of S4 ischemic necrosis was not associated with graft (P = 0.592) or patient (P = 0.243) survival. A MELD score of >30 and pretransplant ventilator support were associated with inferior outcomes. Conclusion The amount of S4 ischemic necrosis is not a prognostic factor in adult SLT recipients, probably due to a sufficiently large GRWR.

BACKGROUND: Hepatocellular carcinoma (HCC) recurrence and development of de novo malignancy (DNM) after liver transplantation (LT) are the major causes of late recipient death.METHODS: We analyzed the incidence of extrahepatic DNM following living donor LT according to the status of pretransplant hepatic malignancy. We selected 2,076 adult patients who underwent primary LDLT during 7 years from January 2010 to December 2016.RESULTS: The pretransplant hepatic malignancy group (n = 1,012) showed 45 cases (4.4%) of the following extrahepatic DNMs: posttransplant lymphoproliferative disease (PTLD) in 10; lung cancer in 10; stomach cancer in 6; colorectal cancer in 5; urinary bladder cancer in 3; and other cancers in 11. The pretransplant no hepatic malignancy group (n = 1,064) showed 25 cases (2.3%) of the following extrahepatic DNMs: colorectal cancer in 3; stomach cancer in 3; leukemia in 3; lung cancer in 3; PTLD in 2; prostate cancer in 2; and other cancers in 9. Incidences of extrahepatic DNM in the pretransplant hepatic malignancy and no hepatic malignancy groups were as follows: 1.1% and 0.5% at 1 year, 3.2% and 2.0% at 3 years, 4.6% and 2.5% at 5 years, and 5.4% and 2.8% at 8 years, respectively (P = 0.006). Their overall patient survival rates were as follows: 97.3% and 97.2% at 1 year, 91.6% and 95.9% at 3 years, 89.8% and 95.4% at 5 years, and 89.2% and 95.4% at 8 years, respectively (P < 0.001). Pretransplant hepatic malignancy was the only significant risk factor for posttransplant extrahepatic DNM.CONCLUSION: Our results suggest that patients who had pretransplant hepatic malignancy be followed up more strictly because they have a potential risk of primary hepatic malignancy recurrence as well as a higher risk of extrahepatic DNM than patients without pretransplant hepatic malignancy.