1. Techno-Economic Viability for Start-Up Ventures (Trainee - South Indian Bank Pvt Ltd)

Studied the various processes within the bank to come up with a logical business plan for a general start up in a backward economy. Also, recommended processes and methods for staying afloat, reducing the com-mensuration period and capitalizing on the growth phase.

Analyzed general procedures followed by banks in India, to finance budding companies and suggested a common platform to check for their viability.

2. Introduction to Real Time Image Processing (Summer Research Intern - Rutgers University)

First real exposure to MATLAB and the Image Processing Toolbox. Basic operations such as dilation and erosion, object count, boundary measurements. texture analysis etc were performed first on to sample images. Later we shifted to real time images obtained from the University of Medicine and Dentistry at New Jersey. 

Details and codes can be found HERE (temporarily unavailable)

3. Morphological Operations on Cancerous Images (Rutgers-Manipal)

The theoretical morphological operations on images were implemented using MATLAB codes for cancerous image processing. The codes were designed in such a way that they could be used in extremely rural low performance systems by semi trained technicians.

Texture Analysis was performed to differentiate benign tumors from cancerous ones and calcification from tumors. The codes were implemented real time in the Kasturba Hospital on various Mammograms and Ultra-sonograms for a fully automated detection of cancerous tumors. The radius and volume of the tumor was also displayed along with the output. 

Details and codes can be found HERE

4. An Embedded System for Measurement of Movement in Stroke Survivors (Research Intern - Indian Institute of Technology, Kharagpur)
The Hand Glove Apparatus

The need for a cure of paralyzed (hemiplegic and quadriplegic) stroke survivors lead to the application of Functional Electrical Stimulation (FES) therapy. With the application of such a modality, there is a growing need for the quantitative assessment of the change in locomotion before and after the treatment. The goal of the project was to develop a low cost, low power embedded system to measure the movements in patients in terms of angular displacement and angular velocity. GAIT analysis and Force/Flex sensors were also required for a complete assessment of all paradigms. 

Texas Instruments MSP 430 G2452 Micro-controller was used as the brain of the embedded system. It was coded in C++ and Assembly Language using CCS (Code Composer Studio) and TIs Launchpad Environment. Gyroscopes, force/flex sensors, LCD displays, LDOs etc were interfaced with the MC to display results real-time and record the values serially using USB to the computer for further analysis

Undergraduate Thesis can be found HERE (Some of the sensitive results have been removed)
Undergraduate Thesis presentation can be found HERE