Sachin Chandrasekara

Sachin Chandrasekara

Graduate student working on microbiome, expert in bioinformatics, & computational biology.

Brain Tumor Segmentation Using Convolutional Neural Networks

Developed a deep learning-based medical image analysis framework for the detection and segmentation of low-grade glioma brain tumors from MRI brain images. The project focused on automating tumor region identification and tumor area calculation using Convolutional Neural Networks (CNNs) and image segmentation techniques.

The study utilized a publicly available dataset from The Cancer Imaging Archive containing 3,929 MRI images of lower-grade glioma patients. Deep learning models were implemented and trained in Google Colab using Python-based machine learning libraries. The dataset was divided into training, validation, and testing datasets using a 70:15:15 ratio.

Model performance was evaluated using Intersection over Union (IoU) and Sørensen Dice Coefficient metrics. Multiple training epochs were tested, and the best model performance was achieved at 50 epochs with 99.92% training accuracy and 92.64% validation accuracy. The project demonstrated the effectiveness of deep learning approaches for automated medical image segmentation and tumor analysis.

Key Technical Work

  • Developed Convolutional Neural Network (CNN) models for brain tumor segmentation
  • Processed and analyzed MRI brain imaging datasets
  • Implemented automated tumor detection and area calculation workflows
  • Trained and validated deep learning models using Google Colab
  • Performed medical image preprocessing and dataset preparation
  • Evaluated model performance using IoU and Dice Coefficient metrics
  • Tested multiple training epochs and optimized model performance
  • Applied deep learning techniques for medical image analysis and healthcare applications

GitHub Repository

🔗 View Project on GitHub

Technologies

  • Python
  • Convolutional Neural Networks (CNNs)
  • Deep Learning
  • Google Colab
  • TensorFlow
  • Keras
  • Medical Image Processing
  • MRI Image Analysis

Skills

  • Deep Learning
  • Computer Vision
  • Medical Image Segmentation
  • Neural Network Modeling
  • Image Processing
  • Machine Learning
  • Scientific Programming
  • Healthcare AI