Sonal Kothari

Current Status: pursuing PhD in electrical engineering and GRA at Bio-MIBLab
B.E in Electronics and Communication engineering from M.B.M Engineering College, Jai Narain Vyas University , Jodhpur, India (2004 – 2008 )

Cell cluster segmentation and informative cell feature extraction (in collaboration with Center for Cancer Nanotechnology Excellence CCNE) - Pathologists often depend on parameters such as the number, shape and size of cells in a tissue sample to make important diagnostic decisions. In healthy conditions, nuclei in cells are mostly distinct and parameters can be determined by direct image segmentation methods such as region-based methods, histogram-based methods and edge detection based methods. However, in pathological conditions, individual cells come close together and nuclei form dense clusters. Therefore, accuracy of cell-counting, cell shape and size determination depends on the segmentation of these dense clusters. We developed a method involving detection of concavities on cell cluster edges and segmentation at these concavities by ellipse fitting. The features estimated include nuclei count, nuclei size distribution, nuclei eccentricity (roundness) distribution, nuclei closeness distribution and cluster size distribution. These features can be used for classification different image types and subtypes.
Comparison of performance of various classifiers for microarray data - The goal of this project is to compare the performance of various classifiers in classification of microarray genomic data with varying classifier parameters.

S. Kothari, Q. Chaudry, and M. Wang, “Automated cell counting and cluster segmentation using concavity detection and ellipse fitting techniques,”, IEEE International Symposium on Biomedical Imaging (ISBI’09), in press.
S. Kothari, Q. Chaudry, and M. Wang, “Extraction of Informative Cell Features by Segmentation of Densely Clustered Tissue Images”, IEEE EMBS’09 Conference, in press.
Chaudry Q, Raza SH, Sharma Y, Kothari S, Young AN, and Wang MD; “Automated renal cell carcinoma subtype classification using cellular features of elliptical models of segmented nuclei clusters”, submitted for Journal of Computerized Medical Imaging and Graphics.

Internship at Tata institute of fundamental research, department of astronomy and astrophysics, Mumbai, India (May 2006 - June 2006)
Internship at Indian institute of science, CEDT, Bangalore, India ( June 2007 – July 2007)