In women, reproductive senescence leads to estrogen deficiency and is frequently associated with higher risk of intracranial hemorrhage. Using porcine and mouse experimental models, we are trying to find the effects of vascular remodeling involving vessel destabilization when the ovarian hormone production is removed. Thus, it is important to have a computerized system that detects vasculature geometry automatically and provide quantitative analysis that can help physiology experts. We developed a computerized Dura Mater laminae analysis system for fluorescence microscopy images of brain tissues from porcine and mice. For this purpose, we built a complete system pipeline which has different automatic image processing and analysis components. We provide efficient methods for automatic filtering, segmentation and network extraction followed by medial axis estimation. Quantitative parameters of the micro-vascular network geometry, including curvature, tortuosity and branching angles were computed using the segmentation derived from medial axis and boundary tracing of microvessels. Experiments using epi-fluorescence based high resolution images of porcine and mice micro-vasculature demonstrates the effectiveness of the holistic approach in predicting vascular remodeling and in hormone production.
We proposed robust filtering based segmentation in  and proposed a novel fusion based method  for further improving the segmentation accuracy. Recently, a random forest classifier driven segmentation is proposed in  the feasibility of a user assisted segmentation is shown in , and deep learning in .
Related projects: denoising
Row-wise: Original, Segmentation, Medial axis-Branches, Medial axis-Angles, Curvature.
 V. B. S. Prasath, O. Haddad, F. Bunyak, O. Glinskii, V. Glinsky, V. Huxley, K. Palaniappan. Robust filtering based segmentation and analysis of Dura Mater vessel laminae using epifluorescence microscopy. 35th Annual International Conference EMBS (IEEE EMBS/EMBC), Osaka, Japan. Proc. IEEE, pp. 6055-6058, 2013.
 R. Pelapur, V. B. S. Prasath, F. Bunyak, O. Glinskii, V. Glinsky, V. Huxley, K. Palaniappan. Multi-focus image fusion using epifluorescence microscopy for robust vascular segmentation. 36th Annual International Conference of the Engineering in Medicine and Biology Society (IEEE EMBC), Chicago, IL, USA. Proc. IEEE, pp. 4735-4738, August 2014. doi:10.1109/EMBC.2014.6944682 Available at figshare: doi:10.6084/m9.figshare.1147490
 Y. M. Kassim, V. B. S. Prasath, R. Pelapur, O. V. Glinskii, R. J. Maude, V. V. Glinsky, V. H. Huxley, K. Palaniappan. Random forests for dura mater microvasculature segmentation using epifluorescence images. 38th Annual International Conference EMBS (IEEE EMBS/EMBC), Orlando, USA, Aug 16-20, 2016. Proc. IEEE, pp. 2901-2904. doi:10.1109/EMBC.2016.7591336. EMBS Student Paper Competition Finalist!
 S. Meena, V. B. S. Prasath, Y. M. Kassim, R. J. Maude, O. V. Glinskii, V. V. Glinsky, V. Huxley, K. Palaniappan. Multiquadric spline-based interactive segmentation of vascular networks. 38th Annual International Conference of the Engineering in Medicine and Biology Society (IEEE EMBC), Orlando, FL, USA. Proc. IEEE, pp. 5913-5916, August 2016. doi:10.1109/EMBC.2016.7592074
 Y. M. Kassim, V. B. S. Prasath, O. V. Glinskii, V. V. Glinsky, V. H. Huxley, K. Palaniappan. Microvasculature segmentation of arterioles using deep CNN. IEEE International Conference on Image Processing (ICIP), Beijing, China. Proc. IEEE, pp. 580-584, September 2017. doi:10.1109/ICIP.2017.8296347. Slides available at Sigport.