[A Study on Radiation Dermatitis Grading Support System Based on Deep Learning by Hybrid Generation Method].

Purpose: Radiation dermatitis is one of the most common adverse events in patients undergoing radiotherapy. However, the objective evaluation of this condition is difficult to provide because the clinical evaluation of radiation dermatitis is made by visual assessment based on Common Terminology Criteria for Adverse Events (CTCAE). Therefore, we created a radiation dermatitis grading support system (RDGS) using a deep convolutional neural network (DCNN) and then evaluated the effectiveness of the RDGS.

A feasibility study of a hybrid breast-immobilization system for early breast cancer in proton beam therapy.

Purpose: We aimed to develop a new breast-immobilizing system for proton beam therapy (PBT) of early breast cancer (EBC) that would provide the optimum breast shape during the treatment as well as increased fixation reliability by reducing the influence of respiratory movement.

Methods: The breast-immobilizing system (HyBIS; hybrid breast-immobilizing system) consists of a whole body immobilization system (WBIS), position-converting device (to change patient position), photo-scanning system, breast cup (made using a three-dimensional printer), breast cup-fitting apparatus, breast cup-holding device (to ensure the breast remains lifted in the supine position), and dedicated stretcher fixed to the WBIS (to carry the patient). We conducted a phantom experiment to evaluate the effect of the HyBIS on breast immobilization during the respiratory cycle.