Retinal Image Analysis

Many important eye diseases as well as systemic diseases which lead to blindness manifest themselves in the retina. Retinal fundus images which shows presence/absence of abnormalities are widely used by ophthalmologists for diagnosis purpose. Hence, Automatic retinal image analysis can be used for diagnosis of  these diseases.

Our current work focuses on the following areas:
  • Disease Analysis: Developing techniques for detection of abnormalities such as microaneurysms, exudates, haemorrhages, cotton wool spots etc.
  • Content Based Image Retrival (CBIR) of Retinal Images: Pathology / Anatomy based retinal image retrival from large databases.

People Involved:
Project Supervisor: Dr. Manesh B. Kokare
Research Scholars: Prasanna Porwal, Ravi Kamble.

Automatic Lung Field Segmentation In CT Scan Images

Automatic lung filed segmentation is very important since it is fundamental step for further quantitative lung analysis in computer aided diagnosis (CAD) from chest computed tomography (CT) scans. Segmenting lung field from CT scans is an important task for the analysis, diagnosis and treatment of pulmonary disease. Many methods for the automatic lung segmentation have been proposed and the segmentation of the lungs without abnormalities in scans of good image quality is possible with high accuracy in most cases. But the segmentation of the lungs in cases containing pathological abnormalities remains challenging and all of the proposed methods will likely fail in a subset of cases.
Our current work is directed towards developing automatic lung field segmentation technique in case of severe pathological abnormalities present in the CT scan images.

People Involved:
Project Supervisor: Dr S.N.Talbar
Reseach Scholar: Ganesh Singadkar

Underwater Acoustic Signal Processing

Ambient noise variability is a critical challenge encountered by multiple stakeholders, including sonar designers and operators. The tropical waters in the Indian Ocean Region (IOR),present random fluctuations in the surface parameters, namely the wind speed,surface temperature, wave height, etc. resulting in variations in the ambient noise characteristics. The site specific surface fluctuations in the tropical regions restrict the possibility of generalized algorithm design to mitigate the ambient noise impact.
The work attempts to characterise ambient noise and to design mitigation algorithms for tropical littoral water.

People Involved:
Project Supervisors:Dr. S. S. Gajre, Dr. Y.V.Joshi and Dr. Arnab Das (Acoustic Research Lab., Tropical Marine Science Institute, National University of Singapore & adjunct faculty at SGGSIE&T).
Research Scholar: Mr. Piyush M. Asolkar