My portfolio items and demos below highlight my ability to deliver top-notch work in the domain of NLP, Computer Vision, Deep Learning, and Machine Learning. There are projects spanning different verticals like healthcare, remote sensing, chatbots, vision automation, etc.
Pdf Chat
- Developed AI-powered pdf chat which can understand any pdf with tables, forms, and any type of miscellaneous data and serve answers
- Used LangChain for the main logic
- Gradio for UI
- OpenAI API for answering Questions
Object Detection
- Developed a TFLite model for the detection of police cars, bikes, and vans.
- Deployed the model in the Android app
Lung Cancer Detection
- My most significant portfolio item in healthcare automation
- Data Pre-processing: Resized, adjusted the luminance, rescaled, and converted data to uint8
- Training Nodule Detector: A 3D Faster R-CNN Network is used as the detector
- Proposals: The model trained was tested on all data, giving all suspicious nodule locations and confidence
- Training Nodule Classifier: Sampled proposals according to confidence, and for each proposal, cropped cubes, centered at the proposed center
Diabetic Retinopathy Classification
- Developed a deep learning model to classify Diabetic Retinopathy using CNN and transfer learning
- Tech-Stack: Python, Keras
Information-Theoretic Feature Selection from Hyperspectral Images
- Proposed greedy feature selection algorithm using mutual information, Relevancy, Redundancy, and Representability
- Carried out experiments on two hyperspectral images i.e. Indian Pines and Salinas Valley
- Observed a considerable increase in accuracy (5-6%) compared to the baseline accuracy.
- It outperforms existing methods such as mRMR, PCA, and NMIFS even at lower dimensions
- Tech-Stack: Python, Keras
- Publication
