Signal and Image Processing Group

Research Overview

Signal and Image processing (SIP) Group’s main focus is on the development of signal and image processing systems and algorithms. Combining machine learning techniques, biological vision models and signal statistics, the group develops advanced image analysis tools for SBIC. The SIP actively participates in ongoing and future projects of SBIC and actively collaborates with all the groups. SIP also focuses on development of novel algorithms for post-processing to extract meaningful information which enables detection, characterization and quantification of clinically significant information. SIP aims to develop new skills and knowledge in advanced multi-modal image acquisition (MR, CT, PET, Optical), signal, and image processing, contribute to IP and publish in prestigious journals.


SIP is currently focused on using machine learning and deep learning for MR data processing applications such as Image super-resolution, Image reconstruction from partial k-space acquisition and generation of quantitative maps. This work is aimed at reducing imaging related costs, patient discomfort, enabling new diagnostic biomarkers and increasing the imaging throughput efficiency. Apart from the above, we are also working on automatic brain extraction and segmentation of marmoset images, Identification of ROIs involved in remembering and recollection process from fMRI images, Stroke lesion segmentation from unenhanced CT and segmentation of ventricular system and brain structures for NPH.

1

We also collaborate internally (within SBIC) and externally on MR imaging and Image processing projects. Some of the ongoing projects are

Projects (Ongoing)   Collaboration with  
 Imaging Vulnerable Carotid Atherosclerotic Plaques TGIC, SBIC
 Automatic monitoring and classification of mouse behaviour NMNCG, SBIC
 Postnatal Radiation Exposure-Induced Changes in Brain Development, Cognition and Aging SNRSI, NUS
 Alterations in brain structure and functions due to TBI and correlation with AD biomarkers QBI, Australia
 Fetal Imaging – Sequence and post-processing development MRMethods group, KK Hospital and Siemens

Automatic segmentation and characterization of Traumatic Brain Injury using multi-parametric MRI

2
3
4

Texture based classification & grading of paediatric Brain Tumour

5
6

Our past projects:

Project (Completed) Collaboration with 
 Multimodal characterization of Brown adipose tissue 
 Abdominal fat quantitation in rodents and Standardization of Fat-Water MIG Group
 Whole brain temperature measurement and analysis in TBI DSO
 Automatic quantification of human abdominal visceral, superficial & deep subcutaneous fat compartments SICS & SAMS
 Adipogenesis & Fluorescent Image Analysis SCFMG
 Texture based classification & grading of paediatric Brain Tumour KK Hospital
 Marmoset Brain extraction and segmentation SigN & KK Hospital



Mission

The Signal and Image Processing (SIP) group is a supporting group whose purpose is to create, develop and implement state-of-the-art signal and image processing resources to SBIC; educate the SBIC’s scientific groups in signal and image processing; and provide a platform for researchers of SBIC to interact and exchange ideas in post-processing. 

Vision

The Signal and Image Processing (SIP) group aims to be the distinguished information and resources group in signal and image processing to SBIC. We achieve this by providing a variety of state-of-the art algorithms to SBIC research problems that are customized to each group; adapting/updating to a rapidly changing technical know-how; and being closely involved in the post-processing activities of each group at all levels.



People