Projects
Actual learning experience comes by actively engaging in putting to action the theoretical knowledge gained for practical implementation to design and build solutions. It helps gain conceptual insights as well as practical expertise.
I always try to put my learnings into practical solutions to create something useful.
Here are the projects I have worked on...
Artificial Intelligence and Deep Learning
Perform a survival analysis based on the time-to-event (death event) for the subjects to assess the likelihood of a death and severity of the heart failure condition.
Key technologies: Python 3.8.8, Numpy, Pandas, Matplotlib, Plotly, Kaplan-Meier estimates, Cox regression models XGBoost, and LightGBM
This study aims to build a time efficient system by applying snapshot ensemble techniques based on deep learning to automate the detection of the parasite using whole slide images of thin blood smears.
Key technologies: Python, Tensorflow, OpenCV, Scikit-image, Matplotlib, Numpy, Pandas
A systematic review and implementation of the UNet++, a convolutional neural network (nested U-Net architecture) dedicated for biomedical image segmentation.
Key technologies: Python, Tensorflow, OpenCV, Numpy, Matplotlib
A systematic review and implementation of the UNet segmentation model, a convolutional neural network dedicated for biomedical image segmentation. Application of the model to segment and identify the cell nuclei.
Key technologies: Python, Tensorflow, OpenCV, Numpy, Matplotlib
Diagnose the presence of skin cancer using CNN & explain the reasoning to arrive at the decision utilizing the Grad-CAM technique by using the gradients flowing into the final convolutional layer and highlighting the important regions considered for arriving at the decision.
Key technologies: Python, Tensorflow, OpenCV, Scikit-learn, Numpy, Pandas
A Deep Learning based AI solution to help diagnose the presence of Covid-19 infection by analyzing the chest X-Ray images.
Key technologies: Python, Tensorflow, OpenCV, Scikit-image, Matplotlib, Numpy, Pandas
AI based application to automatically detect if a person is wearing mask or not. It will be beneficial in places which deploy automated entry gates at malls, theatres, stadiums, indoor arena, gyms, station, airports.
Key technologies: Python, Tensorflow, Matplotlib, Numpy, Scikit-image
A comprehensive analysis on the spread of the corona virus worldwide. The statistical analysis shows the spread using interactive charts and figures. The geographical analysis shows the spread on the geo maps of the majorly affected countries.
Key technologies: Python, Matplotlib, Plotly, folium, Numpy, Pandas
Web Apps - React JS & Ruby
A search filter for a list of items, such as a list of products or contacts. The filter updates the displayed items based on the search input in real-time.
Key Technologies Used: ReactJS and hosted on netlify.
Creates a pagination component for a list of items fetched from an API. User navigates between pages, and ensure the data is displayed correctly using the Progressive & COntrolled ways of pagination
Key Technologies Used: ReactJS and hosted on netlify.
A pagination component with ability to bookmark items in the paginated list & able to navigate to the bookmarked item on click of the corresponding bookmark reference.
Key Technologies Used: ReactJS and hosted on netlify.
Paper Publications & Reviews
This paper discusses about the progress AI has made in the healthcare domain and what the future of AI looks like. It also discusses about the implementation opportunities various AI technologies provide in various fields of healthcare & clinical workflow, how AI systems will boost healthcare to restore human touch in patient-physician encounters.
A systematic review and implementation of the Inception v1, v2, and v3 papers.
Key technologies: Python 3.8.8
A systematic review and understanding of the following papers -
[1] Batch Normalization: Accelerate Training by Reducing Internal Covariate Shift
[2] How Does Batch Normalization Help Optimization?
Key technologies: Python 3.8.8
A systematic review and implementation of the UNet++, a convolutional neural network (nested U-Net architecture) dedicated for biomedical image segmentation.
Key technologies: Python, Tensorflow, OpenCV, Numpy, Matplotlib
A systematic review and implementation of the UNet segmentation model, a convolutional neural network dedicated for biomedical image segmentation.
Key technologies: Python, Tensorflow, OpenCV, Numpy, Matplotlib
A systematic review and implementation of the Grad-CAM: Visual Explanations, uses the gradients of any target concept, flowing into the final convolutional layer to produce a coarse localization map highlighting important regions in the image for predicting the concept.
Key technologies: Python, Tensorflow, OpenCV, Numpy, Matplotlib
Python Frontend using PySimpleGUI
A notepad like application developed using the Python and PySimpleGUI framework.
Key technologies: Python, PySimpleGUI, TKinter
The classic tic tac toe board game developed using PySimpleGUI and TKinter python libraries.
Key technologies: PySimpleGUI, TKinter
A simple image viewer application to load and display all images in the selected directory.
Key technologies: PySimpleGUI, TKinter