Email this page
Department of Mathematics
Santosh Singh
Associate Professor,
Department of Mathematics,
School of Natural Sciences
Email Contact:
Education Details
i. Diploma in Electrical and Computer Engineering, Department of Electrical and Computer Engineering, University of Calgary, Canada – 2004
ii. Ph.D., Department of Mathematics, Indian Institute of Technology, Kanpur, India – 2002

Professional Experience
i. Member Technical Staff, Siemens research and Corporate Technology, Bangalore, Indian, 2006-2013
ii. Technical Consultant, Infonics Software Pvt., Bangalore India, 2006
iii. Software Engineer, ReadInk Technologies, Bangalore, India, 2005
iv. Post-Doc Fellow, Department of Electrical and Computer Engineering, University of Calgary, Canada 2002

Research Interests
Medical image analysis, Image reconstruction, Filter bank theory, Computational photography, Light field and Optimization techniques.

Select Publications
i. Santosh Singh and Ankur Gupta, “How a-priori information effects the Human perception” SPIE Medical Imaging, Orlando, Florida, USA February 2013

ii. Santosh Singh, “Three Dimensional computerized tomography image reconstruction: A unique concept of using light field”, IEEE Nuclear Science Symposium and Medical Imaging Conference, Anaheim, CA, USA October 2012

iii. C. S. Sastry and Santosh Singh, “Reconstruction from divergent ray projections”, SPIE Electronic Imaging and Science and Technology, San Francisco, USA, January 2012

iv. Santosh Singh and M. Celenk, “Highly directional and selective three dimensional adaptive IIR filter”, IEEE 52nd International Midwest Symposium on Circuits and Systems, Cancun, Mexico, August 2009

"A System and method for generating the parameters for acquiring a future image" Patent filled 2011E19221N

National and International Recognition
Council of Scientific and Industrial Research (CSIR) Fellowship 1996-2001,

Post-Doc Fellowship, Department of Electrical and Computer Engineering, University of Calgary, Canada 2001-2002

Review committee member for international conferences on Image processing, Fully3D, Midwest Symposium of Circuits and Systems.

Sponsorship chair for IEE International conference on visual iii. Sponsorship chair for IEE International conference on visual and information engineering, Bangalore, India, September 2006.

Served in different committees in image processing, computer vision based internationally reputed conferences.

Invited Lectures
“Multi-rate filter banks for linear trajectory and plane wave Signals", Department of Electrical Engineering and Computer Science, Ohio University, OH, USA, March – 2010.

"Emerging trends in Medical Imaging and Challenges", Conference on emerging trends in Biomedical Signal processing ETBMPS, Bangalore, India April 2008

Teaching Summary
Good teaching and good research reinforce each other. A researcher who teaches is never isolated. Teaching is meant to foster conceptual understanding, critical thinking and creative applications. Teaching encourages regular contacts and collaboration with a steady stream of new students and other faculty members. Good teaching is a combination of classroom work, easy availability of students to answer questions and to offer guidance in course work and reading.

Research Summary
In Image and video quality assessments plays an important role in various image processing applications. A great deal of efforts has been made in recent years to develop objective image quality metrics that correlate with perceived quality measurements. Unfortunately, only limited success has been achieved. The goal of objective image and video quality assessment research is to supply quality metrics that can predict and perceived image and video quality automatically. PSNR and MSE are the most widely used objective image quality/distortion metrics, but widely criticized as well. The reason for such criticism is what we perceived is not correlated with mathematical measures. Human eyes extract structural information from the viewing field, and human visual system is highly adopted for this purpose. Accordingly, the measurement of structural distortion should be a good approximation. But this philosophical observation has many challenges such as how to define and quantify structural distortions? Will there be any mathematical measure, which can be applied to a large group of applications? What is a good working definition for global image and video quality assessment?

My current research focus is to analyze and understand different scenarios based on the kind of diagnosis and anatomical structures, how different doctors and/or medical expert take the decisions. At this stage of research, the main focus is to visually understand, what special features in a particular scan or medical image makes that scan or medical image a good or bad image. Then as a second step, extract those features mathematically so that they can be trained in such a way to classify the medical image as a good or bad. Because of the subjectivity of each individual doctor and/or medical expert, it is almost impossible to come up with a global solution which can provide such a verdict of good and bad for a given medical image. Instead, my main approach is to provide different kind of indexes, which in fact will provide information about the quality of a medical image. The main advantage in such approach is that each individual doctor and/or medical expert can pick up the index according to his choice. In a sense, my aim is not to declare mathematically an image as good or bad, instead provide an index which can fit the subjective choice of the doctor and/or medical expert. In the process of developing such feature extraction techniques, my focus is also to understand as how a-priori information influences the extraction of visual information even if the target object or information is not so clearly visible in the medical images.

Despite major advances in x-ray sources, detector arrays, gantry mechanical design and specially computer performances, but computed tomography (CT) enjoys the filtered back projection (FBP) algorithm as their first choice for the CT image reconstruction in the commercial scanners. Over the years, a lot of fundamental work has been done in the area of finding the sophisticated solutions for the inverse problems using different kinds of optimization techniques. Recent last few years have really been dominated by the compressive sensing techniques and/or sparse reconstruction techniques. Still there is a long way to go for translating these newly developed algorithms in the clinical environment. The reasons are not obvious and seldom discussed. My main focus in past few years is to develop algorithms which can fit the different requirements of CT machines with an advantage of reducing x-ray dosage. For example, how to combine the concepts of few view and limited view reconstruction algorithm techniques in reducing the x-ray dosage during the fractional treatment planning stages. Such reconstruction techniques are mainly based on the concept of sparsity. The biggest challenge in the meaningful implementation of sparsity concept is to come up with the suitable set of basis functions, which itself is an open problem. Recent research efforts have shown promising results as how the sparsity concept can be optimally implemented if substantial a-priori information is available. Of course such algorithms have limitation and do not fit well in every scenario. Also, these reconstruction algorithms are highly computationally intensive and needs special attention to real time implementations. Along with the accuracy of reconstruction algorithms, I am also focusing on the fast implementation of these algorithms for the real time realization. Apart from focusing on the reconstruction algorithms based on different but well established optimization techniques, my focus is also on to investigate completely new paradigm for reconstruction algorithm.

Executive Summary
Santosh Singh has joined Shiv Nadar University in March 2013. He has an extensive experience in medical imaging domain, human perception based image quality assessment. He has published papers and worked in the area of computerized tomography, light field rendering concept, optimization techniques and filter bank theory. Being from the mathematical background, his research is always focused on strong theoretical platform which can be applied to different applications.

Postgraduate Undergraduate
Natural Sciences
Research Postgraduate Undergraduate
Humanities & Social Sciences
Postgraduate Undergraduate
Management and Entrepreneurship
I-Global MBA – 2016-18
Academic Calendar
About the University
Curriculum Approach
Offices and Services
Address and Directions
© Copyright Shiv Nadar University 2017. All Rights Reserved. Disclaimer | Sitemap
The Shiv Nadar University has been established under U.P. Act No 12 of 2011. Shiv Nadar University is UGC Approved