Working in the feild of Visual, Audio and Audio-Visual Speech Recognition System under Dr Rajiv Ratna Shah(Director of MIDAS LAB), Prof. Changyou Chen(Professor at University of Buffalo) and Yaman Kumar Singla(Google PhD fellowship 2020)
Working in the feild of hemodynamic shock prediction in Intensive Care Unit. At Tavlab I am mentored by Dr Tavpritesh Sethi(Founder of TAVLAB), Dr. Piyush Mathur, MD and Ridam Pal
Working in the feild of NLP, building models for rumour detection task. This internship lead to a publication. Dr. Preeti kaur mentored me throughout this research.
Worked with the NGO based in Nigeria to help them in developing website for a project. I was resposible for the frontend and was tasked with making the website fully responsive. The website was for bring volunteers to schools where there service can do a lot of good.
I contributed to the development of a website that helps offset CO2 emission. I was tasked with building the frontend and help backend developer. I used HTML, CSS, Bootstrap, JavaScript and Flask(Backend).
A novel technique to detect rumors from tweets. We used BERT, RoBERTA, ALBERT, and DistilBERT model to represent source and comment tweets. Our approach was able to produce better precision, recall, and F1 score over the state-of-the-art classifier that uses Conditional Random Fields (CRFs) to learn the context during the event.
Patients with long-standing diabetes often fall prey to Diabetic Retinopathy (DR) resulting in changes in the retina of the human eye, which may lead to loss of vision in extreme cases. The aim of this study is two-fold: (a) create deep learning models that were trained to grade degraded retinal fundus images and (b) to create a browser-based application that will aid in diagnostic procedures by highlighting the key features of the fundus image. In this research work, we have emulated the images plagued by distortions by degrading the images based on multiple different combinations of Light Transmission Disturbance, Image Blurring and insertion of Retinal Artifacts. InceptionV3, ResNet-50 and InceptionResNetV2 were trained and used to classify retinal fundus images based on their severity level and then further used in the creation of a browser-based application, which implements the Integration Gradient (IG) Attribution Mask on the input image and demonstrates the predictions made by the model and the probability associated with each class.