Skip to content

This project was done as part of Capstone Project for PGP in Artificial Intelligence and Machine Learning by Great Learning

Notifications You must be signed in to change notification settings

ramanks19/Capstone-Project-AIML

Repository files navigation

Capstone-Project-AIML

This project was done as part of Capstone Project for PGP in Artificial Intelligence and Machine Learning by Great Learning

📁 Getting Started

The project is built on Google Colab Jupyter Notebook and Kaggle.

🤔 Problem Description

In this capstone project, the goal is to build a Pneumonia Detection System, to locate the position of inflammation in an image. In all, the project objective can be described as:

  • Build a reliable Pneumonia Detection Model which can have a robust backing.
  • Proper pre-processing and meaningful Exploratory Data Analysis.
  • The medical images dataset can be properly trained by a deep learning network with custom architectures
  • Use transfer learning to facilitate training with final layers of the deep network trainable
  • Learn to fine tune the model by trying different optimizers, loss functions, epochs, learning rate, batch size, check pointing, early stopping etc.
  • Read different research papers of given domain to obtain the knowledge of advanced models for the given problem.
  • Advocate a strong backing case for the reliability of the model finally obtained by proposing a use case confidence interval.

📜 Approach

📈 Step 1: Exploratory Data Analysis & Data Preparation

  • Understanding the data with a brief on train/test labels and respective class info
  • Look at the first five rows of both the csvs (train and test)
  • Identify how are classes and target distributed
  • Check the number of patients with 1, 2, ... bounding boxes
  • Read and extract metadata from dicom files
  • Perform analysis on some of the features from dicom files
  • Check some random images from the training dataset
  • Draw insights from the data at various stages of EDA
  • Visualize some random masks generated

Outcome

⚙️ Step 2: Model Building

  • Split the data
  • Use different models to train the data. Here we are using UNet to train our dataset with different backbone structures
  • Evaluate the models (ROC AUC, AP, F1 Score)

Outcome

About

This project was done as part of Capstone Project for PGP in Artificial Intelligence and Machine Learning by Great Learning

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published