Academic, Research, and Independent Projects



Deep Learning

Overview: We have developed algorithms for neural architecture to register CT scan images.

Brain Deep L.

Medical Image Registration and Fusion Processing System

Research Project, Professor Dr. Weiying Dai’s Lab, Binghamton University, NY

February 2018–December 2018

• Developed a Deep Learning tool (Python) for registering Computerized Tomography scan of the kidney

• Built a Registration Framework based on a three-dimensional Convolutional Neural Network that directly learns transformations between pairs of three-dimensional images

Presentation


Machine Learning

Overview: Devloped neural network algorithms for recomendation system, public transportation system, and crop disease detection system.

Recomender system

Recommender System

Academic Project, Binghamton University, NY

January 2018-May 2018

• Developed a recommender system (Java) using the collaborative filtering approach

• Used item-based Collaborative filtering and Adjusted cosine similarity to compute the item similarity

• Predicted the missing values for a user by the Weighted sum approach

Code


crop disese

Intelligent Interpretation System for Diseases on Cash Crops in Vidarbha Region

Academic Project, Sant Gadge Baba Amravati University, Amravati, India

August 2013-May 2014

• Developed a MATLAB application for the detection of 5 common crop plant diseases in India with 91% classification accuracy

• Implemented K-means clustering technique and Multilayer Perceptron Neural Network for clustering and classification of plant diseases

Research Paper 1 | Research Paper 2


Public transport system

Intelligent Public Transport Prediction System using Wireless Sensor Network

Research Project, Sant Gadge Baba Amravati University, Amravati, India

August 2013-January 2014

• Forecasting of vehicle travel time prediction for Jaipur-Delhi Route using FTLR Neural Network Model with Gamma Memory

• Public Transport vehicle failure reporting system

Research Paper 1 | Research Paper 2 | Research Paper 3


Mobile Application

Overview: Developed Android Application using sensors.

Mobile App Dashboard

GPS Coordinates Emergency Alert Application

Academic Project, Binghamton University, NY

August 2018- December 2018

• Developed an Android application with functionality to send a Location Alert message while saving critical time during an emergency

• Used Accelerometer sensor and Kalman Filter Algorithm to recognize Shake Gesture and Google Fused Location Provider API

Video | Application


Computer Network

Overview: Developed algorithms, protocols, and network systems.

Adaptive Headlight

Control Model of Adaptive Headlight System

Independent Work, IETE Cynosure (ICCEEE-2013), Lonere, India

August 2012-December 2013

• Developed an Adaptive Headlight Microcontroller based system (C++) using the CAN bus protocol

• Designed the system to adjust the automobile's headlights to curves in the road based on steering rotation and vehicle yaw rotation

• Awarded the Institution of Electronics and Telecommunications Engineers (IETE) Mumbai Centre’s Young Researchers Award (selected from 110 national and international researchers)

Research Paper | Award


Q-CSMA

Q-CSMA: Queue-Length based CSMA/CA Algorithm

Independent Work

August 2013-May 2014

• Proposed a discrete-time version of the CSMA algorithm (Java), where multiple links are allowed to update their states in a single time slot

• Decreased Delays by 11% (average) while retaining the throughput-optimality property in Wireless Networks

Code


Tech for Social Good

Overview: Implemented design principles, to leverage engineering technology for social welfare.

Baler

Improving Cattle Fodder Storage, Tanzania

International Development Design Summit 2014, MIT D-Lab, USA

● Implemented Human-Centered Design approach to build low-cost Hand Hay Baler and Storage Unit to improve Maasai cattle breeding and caring practices

● Developed a GUI application (in Java) to teach Maasai basic computer skills and English

Article


Gas Monitor

Gas Switch

Research Project, Massachusetts Institute of Technology (MIT) Media Lab, Cambridge, MA

January-February 2014

● Designed a Smart Gas Stove with Burner Dial settings and Timer Smartphone functionality using Raspberry Pi to control Servo Motor by Android Smartphone over Wi-Fi

●Achieved a 30% increase in cooking time efficiency and a 20% decrease in monthly expenditure for Dharavi Catering Businesses

Video