Purpose: Non-operative management (NOM) has emerged as a promising organ-preserving strategy for patients with rectal cancer who achieve a clinical complete response (cCR) after neoadjuvant chemoradiotherapy (CRT). However, no standardized treatment protocol has been established for watch-and-wait strategies. Materials and Methods: This prospective study evaluated oncological outcomes of NOM combined with 4 months of adjuvant capecitabine. Patients with resectable rectal cancer (≤ 8 cm from the anal verge, cT2-4 or N+) underwent CRT (50-54 Gy in 25-27 fractions with capecitabine). Eight weeks post-CRT, a multidisciplinary team assessed cCR. Patients achieving cCR received six cycles of capecitabine (2 weeks on/1 week off) and were actively monitored. Results: Among 89 patients receiving CRT (2018-2023), 17 (19.1%) achieved cCR and were included. The median age was 65 years, and 64.7% were male. Eleven (64.7%) completed all six cycles of adjuvant therapy. After a median follow-up of 31.4 months, 11 patients (64.7%) remained disease-free. Local regrowth occurred in six patients (35.3%) with 2- and 4-year rates of 34.5% and 47.6%, respectively. Five underwent radical surgery, and one received transanal excision with systemic chemotherapy. At the time of assessment, 15 patients (88.2%) showed no evidence of disease, while two (11.8%) received palliative chemotherapy. All patients were alive. Conclusion NOM with adjuvant capecitabine showed promising oncological outcomes, offering an alternative to passive watch-and-wait approaches. Further refinement through multidisciplinary strategies is warranted.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Background/Aims: We investigated the clinical practice patterns of Korean endoscopists for the endoscopic resection of colorectal polyps. Methods: From September to November 2021, an online survey was conducted regarding the preferred resection methods for colorectal polyps, and responses were compared with the international guidelines. Results: Among 246 respondents, those with <4 years, 4–9 years, and ≥10 years of experiencein colonoscopy practices accounted for 25.6%, 34.1%, and 40.2% of endoscopists, respectively. The most preferred resection methods for non-pedunculated lesions were cold forceps polypectomy for ≤3 mm lesions (81.7%), cold snare polypectomy for 4–5 mm (61.0%) and 6–9 mm (43.5%) lesions, hot endoscopic mucosal resection (EMR) for 10–19 mm lesions (72.0%), precut EMR for 20–25 mm lesions (22.0%), and endoscopic submucosal dissection (ESD) for ≥26 mm lesions (29.3%). Hot EMR was favored for pedunculated lesions with a head size <20 mm and stalk size <10 mm (75.6%) and for those with a head size ≥20 mm or stalk size ≥10 mm (58.5%). For suspected superficial and deep submucosal lesions measuring 10–19 mm and ≥20 mm, ESD (26.0% and 38.6%) and surgery (36.6% and 46.3%) were preferred, respectively. The adherence rate to the guidelines ranged from 11.2% to 96.9%, depending on the size, shape, and histology of the lesions. Conclusions Adherence to the guidelines for endoscopic resection techniques varied depend-ing on the characteristics of colorectal polyps. Thus, an individualized approach is required to increase adherence to the guidelines.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Background/Aims: We investigated the clinical practice patterns of Korean endoscopists for the endoscopic resection of colorectal polyps. Methods: From September to November 2021, an online survey was conducted regarding the preferred resection methods for colorectal polyps, and responses were compared with the international guidelines. Results: Among 246 respondents, those with <4 years, 4–9 years, and ≥10 years of experiencein colonoscopy practices accounted for 25.6%, 34.1%, and 40.2% of endoscopists, respectively. The most preferred resection methods for non-pedunculated lesions were cold forceps polypectomy for ≤3 mm lesions (81.7%), cold snare polypectomy for 4–5 mm (61.0%) and 6–9 mm (43.5%) lesions, hot endoscopic mucosal resection (EMR) for 10–19 mm lesions (72.0%), precut EMR for 20–25 mm lesions (22.0%), and endoscopic submucosal dissection (ESD) for ≥26 mm lesions (29.3%). Hot EMR was favored for pedunculated lesions with a head size <20 mm and stalk size <10 mm (75.6%) and for those with a head size ≥20 mm or stalk size ≥10 mm (58.5%). For suspected superficial and deep submucosal lesions measuring 10–19 mm and ≥20 mm, ESD (26.0% and 38.6%) and surgery (36.6% and 46.3%) were preferred, respectively. The adherence rate to the guidelines ranged from 11.2% to 96.9%, depending on the size, shape, and histology of the lesions. Conclusions Adherence to the guidelines for endoscopic resection techniques varied depend-ing on the characteristics of colorectal polyps. Thus, an individualized approach is required to increase adherence to the guidelines.