Core members

 

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).

Extended picture archiving and communication systems

In the age of digitalization, PACS has been used widely in hospitals and medical centers. In this research, we establish a extended picture archiving and communication systems to assist radiologist and oncologist in accessing the medical image. The extended PACS is builded based on an open source PACS which is compatible to standard communication protocols of medical images. The extended features allow clinicians to access and view medical images via mobile devices such as laptop and tablet based on web and android operation system, enabling the clinicians to diagnose patient diseases remotely. In addition, the extended PACS is designed as an open system which enables to extend in a large scales and integrate with medical image processing modules. The aim is to connect multi hospital and satellite medical centers, allowing to share and reuse the medical image resources.

Improving karyotyping of human chromosomes using deep learning

Karyotyping based on biomedical image processing acts an crucial role in gene analysis, enabling geneticist to diagnose several genetic diseases and genetic disorder such as Down syndrome, leukemia, etc. In current workflow, geneticist use conventional biomedical image processing method to separate the chromosomes either manually or semi-automatically. Karyotyping often requires geneticist to analysis dozens of metaphase images and reorganize the chromosome into 23 pairs and thus is very time consuming. In this research, we building a tool which applies AI to automatically classify the chromosomes, aiding geneticist to detect abnormality in the karyogram of the patients.

Signal and data processing for vital sign data

The outbreak of infectious diseases is threatening global health. Especially, in the South-East Asia region have been at serious risk. At mass gathering places, such as, airport quarantine facilities, public health centers, and hospital out patients units, rapid and highly reliable screening methods of infection are urgently needed during the epidemic season for preventing the spread of infection.

EEG signal processing from the computer vision perspective

In this work, we study the processing of EEG signals in the image domain, i.e., visualizing the signal in the binary image and applying computer vision techniques for signal smoothing and classification. Some research questions are being investigated:

1. How well can we smooth the signal in the image domain? Here, we plot the interested signal in binary representation at different line-width levels, then “thin” using some recent effective skeletonization algorithms. The process consists of plotting the signal at a high-level line width (upscale) and then at a unit line width (downscale); hence, the name Upscale and Downscale Representation (UDR) comes into play. The skeletonized signal is then converted back to the time domain for correlation evaluation.

2. How can UDR be implemented in classification models to improve detection accuracy? Here, we investigate the feasibility of implementing UDR as a smoothing filter for data pre-processing or in a shallow layer of a CNN to enhance the signal classification accuracy further using deep learning.

Other information

Contact: Dr. Le Trung Thanh