Purpose: Breast cancer brain metastases (BCBM) may involve subtypes that differ from the primary breast cancer lesion. This study aimed to develop a radiomics-based model that utilizes preoperative brain MRI for multiclass classification of BCBM subtypes and to investigate whether the model offers better prediction accuracy than the assumption that primary lesions and their BCBMs would be of the same subtype (non-conversion model) in an external validation set. Materials and Methods: The training and external validation sets each comprised 51 cases (102 cases total). Four machine learning classifiers combined with three feature selection methods were trained on radiomic features and primary lesion subtypes for prediction of the following four subtypes: 1) hormone receptor (HR)+/human epidermal growth factor receptor 2 (HER2)-, 2) HR+/HER2+, 3) HR-/HER2+, and 4) triple-negative. After training, the performance of the radiomics-based model was compared to that of the non-conversion model in an external validation set using accuracy and F1-macro scores. Results: The rate of discrepant subtypes between primary lesions and their respective BCBMs were 25.5% (n=13 of 51) in the training set and 23.5% (n=12 of 51) in the external validation set. In the external validation set, the accuracy and F1-macro score of the radiomics-based model were significantly higher than those of the non-conversion model (0.902 vs. 0.765, p=0.004; 0.861 vs. 0.699, p=0.002). Conclusion Our radiomics-based model represents an incremental advance in the classification of BCBM subtypes, thereby facilitating a more appropriate personalized therapy.

Background/Aims: Appropriate tissue tension and clear visibility of the dissection area using traction are essential for effective and safe endoscopic submucosal dissection (ESD). In this study, we developed a retractable robot-assisted traction device and evaluated its performance in colorectal ESD. Methods: An experienced endoscopist performed ESD 18 times on an ex vivo porcine colon using the robot and 18 times using the conventional method. The outcome measures were procedure time, dissection speed, procedure-related adverse events, and blind dissection rate. Results: Thirty-six colonic lesions were resected from ex vivo porcine colon samples. The total procedure time was significantly shorter in robot-assisted ESD (RESD) than in conventional ESD (CESD) (20.1±4.1 minutes vs 34.3±8.3 minutes, p<0.05). The submucosal dissection speed was significantly faster in the RESD group than in the CESD group (36.8±9.2 mm 2 /min vs 18.1±4.7 mm 2 /min, p<0.05). The blind dissection rate was also significantly lower in the RESD group (12.8%±3.4% vs 35.1%±3.9%, p<0.05). In an in vivo porcine feasibility study, the robotic device was attached to a colonoscope and successfully inserted into the proximal colon without damaging the colonic wall, and ESD was successfully performed. Conclusions The dissection speed and safety profile improved significantly with the retractable RESD. Thus, our robotic device has the potential to provide simple, effective, and safe multidirectional traction during colonic ESD.