• Advanced Institute of Engineering and Technology (AVITECH)

  • Seminars

    September 15, 2020: Dr. Luu Manh Ha (AVITECH), DLAD: Image Processing Method based on CNN and Anisotropic Diffusion Filter for Improving Medical Image Compression, Applied for Teleinterventions using 3D Medical Images

    Tele-radiology is increasingly being used on a large scale worldwide. Image compression while preserving the quality of the image is essential in clinical diagnostic and treatment. Compression is especially relevant if the bottleneck in lively viewing the image is in the image transfer over poor Internet connection condition. This talk presents a framework for organ-specific image compression using a proposed image processing method, DLAD, in a preprocessing step to reduce the entropy of the image. The proposed method uses a modern CNN network to extract a probability map of the organ of interest; this probability map guides an anisotropic diffusion filter that smooths the image except at the location of the organ of interest. Subsequently, a compression method, such as BZ2 and HEVC-visual lossless, is applied to compress the image. We demonstrate the proposed method for radio-frequency ablation (RFA) of liver cancer intervention using 3D CT images. To verify the effect of the compression on the quality of the diagnostic and treatment of radiologists, we compare the performance of two certified radiologists on the liver lesion detection and the liver lesion center annotation using 33 sets of the original images and the decompressed images. The statistical analysis shows that the achieved results are promising for teleradiology applications.

    Speaker: Dr. Luu Manh Ha, AVITECH

    Time: 15:30, Tuesday, September 15, 2020

    Venue: G2-315, 144 Xuan Thuy, Cau Giay, Hanoi


    Dr. Luu Manh Ha was born in Hanoi, Vietnam, in 1985. He has been working for the VNU University of Engineering and Technology (VNU-UET) of Vietnam National University, Hanoi (VNU) in Vietnam since 2017 as a lecturer and a researcher. He completed his Bachelor program in Faculty of Electronics and telecommunications at VNU-UET in 2007. He continued the work as a researcher and received his MSc degree in Electronic Engineering from VNU-UET in 2010. He finished his PhD in the BIGR group, Erasmus Medical Center, Rotterdam, the Netherlands in 2017. His research scheme is applications of AI and digital signal processing techniques on telemedicine and telehealth, mainly focuses on abdominal images (CT, US) of the liver, biosignal processing, and choromosome image processing.


    February 2, 2024: Dr. Khoa D. Doan (Vin Univeristy) Toward Reliable and Practical Machine Learning Applications

    While Machine Learning (ML) has rapidly transformed several domains and applications with incredible successes, there are also important areas where the progress is significantly slower. Specifically, there exists a widened complexity gap between the methods currently investigated in research and those used in practice in these areas. One reason is that many algorithms, despite achieving […]

    February 2, 2024: Prof. Heng Ji (University of Illinois at Urbana-Champaign), Combating with Misinformation and Cancer: A Unified Multimodal AI Approach to Healthy and Happy Life

    A research overview of ongoing research projects, especially focusing on two that are most related to the VinUni-UIUC Smart Health Center: (1) Misinformation Detection and Trustworthy Large Language Models; (2) Joint Natural Language and Molecule Learning for Drug Discovery. Unsurprisingly these two seemingly different research problems can be tackled with a unified approach based on […]

    January 11, 2024: Dr. Le Duc Trong (FIT-UET) Resilient Multimodal Learning for Multimodal Emotion Recognition in the Presence of Incomplete Modalities

    Multimodal Emotion Recognition in Conversation (Multimodal ERC) is a critical area of research for interpreting human communication in diverse applications. Nevertheless, the persistent issue of uncertain missing modalities poses a major hurdle, hampering the development of robust Multimodal ERC models. Existing approaches face limitations in effectively leveraging a fusion of diverse data modalities encompassing audio, […]