Machine Learning for Bioimage Analysis
Our group aims to develop machine learning approaches for the automated analysis of biological and medical images. Topics of interest include, but not limited to, cellular and tissue image analysis such as segmentation, tracking, and phenotype recognition.
Reconstruction and Tracing Filamentary Structured Objects
The reconstruction and tracing of the filamentary (i.e. thin & long) structured objects is an important problem that are often encountered in biomedical images such as retinal blood vessels, neurons, among others. Exemplar images are e.g. depicted in Fig. 1. For this project, we have established international reputations over the years [3,6,8,9,11] in this research area, and have continuously built-up international collaborations, such as the BigNeuron consortium led by Allen Institute of Brain Science, USA.
3D Behavior Analyses of Articulated Objects
An important aspect of behavior neuroscience is to study the behaviors and social behaviors of human and animals, which play important roles in e.g. understanding neural degenerative diseases. Our group has developed state-of-the-art working systems along this line of researches, some of which have resulted in publications in top venues such as [4,5,7,10,13]. In particular, we develop systems using single 3D cameras that provide 3D posture information of the articulated objects of interest, as e.g. displayed in Fig. 2. More recent results of our demonstration systems can be obtained from related project websites. We have closely collaborated with local research institutes such as IMCB, I2R, NNI, and NUS in this project.
Machine Learning For Bioimage Analysis Members
Dr. Cheng Li
|Dr. CHENG Li||Principal Investigator|
|Dr. WANG Dong||Postdoctoral Fellow|
|Mr. GUAN Zhen||Research Officer|
|Mr. HOE Yew Hock||Research Officer|
|Mr. WU Shuang||PhD student|
This section is still work in progress.