Projects
Localizing epileptogenic zone from spikes and high frequency oscillations using machine learning models in temporal lobe epilepsy (2021): Masters thesis for localizing epileptic seizure occurrence in human brains using EEG signal. As part of the study, I developing a machine learning model capable of identifying the variations and predicting the possibility of the seizure origin in temporal lobe epileptic patients. Supervised by Dr. Harilal Parasuram, Amrita Institute of Medical Sciences and Dr. Manjusha Nair.
Predicting Autonomic Dysfunction in Anxiety Disorder from ECG and Respiratory Signals Using Machine Learning Models (2021): Research project for detection of respiratory rate variability and heart rate variability in patients with an anxiety disorder to identify autonomic dysfunctions using ECG signal. Supervised by Dr. Harilal Parasuram, Amrita Institute of Medical Sciences
Auto-navigation of mathematical modeled quad copter using deep learning (2019): Bachelors degree research project on developing a learning based controller on a mathematically modeled quadcopter for automating the quadcopter entirely and estimate the trajectory on its own. The model was non-linear and the controller was also non-linear, and a classical linear controller (PID) was used for stabilization. The project was simulated using Matlab and SIMULINK, used pretrained convoulutional neural network (Alexnet) for obstacle detection as the learning algorithm. Supervised by Johan Thunberg, Högskolan i Halmstad. (slides & report)
Recommendation System using distributed clustering (2018): Research project (minor project) for recommendation system using a distributed databases and clustering algorithm. Used optics algorithm as the main algorithm for recommendation and experimented with certain other clustering algorithms such as K-means and DBScan. A complete distributed environment was developed for the application. Supervised by Raji Ramachandran, Amrita Vishwa Vidyapeetham.
Other Projects
This section includes web applications, stand-alone application and android application development projects that I have done in the past.
HRV-RRHV-Analysis Tookit (2021): A software developed using Tkinter and PyQT capable of processing and estimating Heart rate variability and respiratory rate variability parameters. The tools is entirely developed in Python language. The tool was developed as part of an initial stage of research based on Autonomic dysfunction prediction based on HRV and RRV estimated features. A research journal article has been published as part of the research done at Amrita Instutite of Medical Sciences and Research.
Calendly clone (2020): A clone of Calendly scheduler application developed as part of a job interview assignment. Application runs REST API in the backend capable of delivering CRUD operations for performing scheduling. Application front-end has some updates that are still undergoing development. Used React.js, Material UI, Calender npm packages for front-end and Django, SQLite, Django REST framework for backend. (source code front-end)
Password Manager (2020) A password manager Android application developed using Java. Basic functionality includes random password generation, storing and retrieving user credentials for individual websites. Uses AES encryption to encrypt the password database. Future addons will be, firebase firestore API integration capable of providing cloud synching and storing of data.
MIHU2019 (2019): An online portal(May I Help You) developed for Amritavarsham66 as part of cultural event organized by Amrita Vishwa Vidyapeetham, Amritapuri campus. Application was developed using the MERN(MongoDB, ExpressJS, ReactJS, and NodeJS) stack with an integration of Material-UI. Served around 3000+ devotees and 1000+ volunteers that participated at the event.
Human Foundation (2019) Developed a web application with an organization for helping the people in Kerala for providing flood relief during the flood. Used ReactJS, and Firebase combination to create a web application that provided real time data processing and extra features such as tracking location, messages etc.