Research Projects
CT image analysis for liver cancer intervention using convolutional neural networks
Liver cancer is the sixth most common cancer worldwide and the most common cause of death from cancer in Vietnam. In this project, we focus on applying AI in order to improve liver interventions for liver cancer treatment using the radiofrequency ablation (RFA) and Radioembolization (TARE).
Two main problems we are solving are:
- Improving the compression ratio of 3D CT images of the liver for teleinverventions
- Reducing the radiation exposure by optimizing scan range of the liver in CT imaging.
In the former problem, efficient compression of images while preserving image quality has the potential to be a major enabler of effective remote clinical diagnosis and treatment, since poor Internet connection conditions are often the primary constraint in such services. We improve the conventional image compression method by using CNNs and filtering technique such that the compression ratio is increase while remaining the image quality for the clinical purpose. In the latter problem, the manual CT scanning of the liver in the interventions often contain excess scan range. Therefore, we aim to use CNNs to predict scan range of the liver in the interventions and subsequently allow to reduce the excess radiation dose manually performed by CT radiographer.
This project is under collaboration of University of Engineering and Technology, University of Technology, Sydney (Australia), the Erasmus Medical Center (The Netherlands) and Bach Mai Hospital (Vietnam).
Examples of the decompressed CT images in RFA liver intervention using BZ2 and HEVC method. By filtering the unnecessary in information outside of the region of organ of interest, the entropy of the image is reduced, enabling to archieve higher compression ratio while the quality of the image is preserved for clinical purpose. You can find a demo Matlab code here.
Selected publications
Luu, H. M., van Walsum, T., Franklin, D., Pham, P. C., Vu, L. D., Moelker, A., … & Trung, N. L. (2021). Efficiently compressing 3D medical images for teleinterventions via CNNs and anisotropic diffusion. Medical Physics, 48(6), 2877-2890.
Luu, M. H., van Walsum, T., Mai, H. S., Franklin, D., Nguyen, T. T. T., Le, T. M., … & Trung, N. L. (2022). Automatic scan range for dose-reduced multiphase ct imaging of the liver utilizing cnns and gaussian models. Medical Image Analysis, 78, 102422.
Loc, P. X., & Ha, L. M. (2023). Impact of Image Denoising Techniques on CNN-based Liver Vessel Segmentation using Synthesis Low-dose Contrast Enhanced CT Images. REV Journal on Electronics and Communications, 12(3-4).
Hoang, H. S., Pham, C. P., Franklin, D., van Walsum, T., & Luu, M. H. (2019, September). An evaluation of CNN-based liver segmentation methods using multi-types of CT abdominal images from multiple medical centers. In 2019 19th international symposium on communications and information technologies (ISCIT) (pp. 20-25). IEEE.
Trung, N. T., Hoan, T. D., Trung, N. L., & Ha, L. M. (2019, December). Robust Denoising of Low-Dose CT Images Using Convolutional Neural Networks. In 2019 6th NAFOSTED Conference on Information and Computer Science (NICS) (pp. 506-511). IEEE.
Other information
PI: Dr. Luu Manh Ha, AVITECH
