Cross-Sectional Ultrasound Image Generation for Ultrasonographer Training System Using Mixed Reality
Khanabhorn Kawattikul, Kazuhiko Hamamoto · 2024
This research presents a method for generating cross-sectional ultrasound images for ultrasonographer training sys-tems. The work mainly focuses on developing an efficient technique that can produce application-quality cross-sectional ultra-sound images from 3D volume data, which allows multi-angle cross-section generation (2D images). The proposed technique tackles volume and surface reconstruction techniques to effective-ly generate both 3D and 2D datasets. To evaluate the performance of the proposed technique, we employed a subset of the FPUS32 dataset as ground truth and compared the generated cross-sectional images with the original ultrasound images. The results demonstrate that the proposed technique is capable to produce images of comparable quality, highlighting its potential for effective application in ultrasound diagnosis, research, and training.