• The Journal of Korean Society for Radiation Therapy
• /
• v.23 no.1
• /
• pp.13-19
• /
• 2011

Purpose: It's essential to minimize the tumor motion and identify the exact location of the lesions to achieve the improvement in radiation therapy efficiency during SBRT. In this study, we made the established compression belt to reduce respiratory motion and evaluated the usefulness of clinical application in SBRT. Materials and Methods: We analyzed the merits and demerits of the established compression belt to reduce the respiratory motion and improved the reproducibility and precision in use. To evaluate the usefulness of improved compression belt for respiratory motion reduction in SBRT, firstly, we reviewed the spiral CT images acquired in inspiration and expiration states of 8 lung cancer cases, respectively, and analyzed the three dimensional tumor motion related to respiration. To evaluate isodose distribution, secondly, we also made the special phantom using EBT2 film (Gafchronic, ISP, USA) and we prepared the robot (Cartesian Robot-2 Axis, FARARCM4H, Samsung Mechatronics, Korea) to reproduce three dimensional tumor motion. And analysis was made for isodose curves and two dimensional isodose profiles with reproducibility of respiratory motion on the basis of CT images. Results: A respiratory motion reduction compression belt (Velcro type) that has convenient use and good reproducibility was developed. The moving differences of three dimensional tumor motion of lung cancer cases analyzed by CT images were mean 3.2 mm, 4.3 mm and 13 mm each in LR, AP and CC directions. The result of characteristic change in dose distribution using the phantom and rectangular coordinates robot showed that the distortion of isodose has great differences, mean length was 4.2 mm; the differences were 8.0% and 16.8% each for cranio-caudal and 8.1% and 10.9% each for left-right directions in underdose below the prescribed dose. Conclusion: In this study, we could develop the convenient and efficient compression belt that can make the organs' motion minimize. With this compression belt, we confirmed that underdose due to respiration can be coped with when CTV-PTV margins of mean 6 mm would be used. And we conclude that the respiratory motion reduction compression belt we developed can be used for clinical effective aids along with the gating system.

• Journal of Broadcast Engineering
• /
• v.27 no.1
• /
• pp.44-55
• /
• 2022

In order to automatically provide information on fruits in agricultural related broadcasting contents, instance image segmentation of target fruits is required. In addition, the information on the 3D pose of the corresponding fruit may be meaningfully used. This paper represents research that provides information about tomatoes in video content. A large amount of data is required to learn the instance segmentation, but it is difficult to obtain sufficient training data. Therefore, the training data is generated through a data augmentation technique based on a small amount of real images. Compared to the result using only the real images, it is shown that the detection performance is improved as a result of learning through the synthesized image created by separating the foreground and background. As a result of learning augmented images using images created using conventional image pre-processing techniques, it was shown that higher performance was obtained than synthetic images in which foreground and background were separated. To estimate the pose from the result of object detection, a point cloud was obtained using an RGB-D camera. Then, cylinder fitting based on least square minimization is performed, and the tomato pose is estimated through the axial direction of the cylinder. We show that the results of detection, instance image segmentation, and cylinder fitting of a target object effectively through various experiments.