Skip to content

Latest commit

 

History

History
53 lines (32 loc) · 2.97 KB

README.md

File metadata and controls

53 lines (32 loc) · 2.97 KB

NeuroDetect: MRI‑Based Alzheimer’s Disease Detection using DNN

🧠 Overview

Training a Deep Neural Network (DNN) to classify brain MRI scans for detecting Dementia versus Non-Dementia patients. This project aims to use a simpler architecture with fewer parameters while still achieving good accuracy. We focus on a data-centric approach, emphasizing data augmentation and fine-tuning hyperparameters rather than complex model architectures.

image

The dataset is sourced from Kaggle and can be found at Alzheimer's Dataset

🎯 Objective

The main objective of this project was not to create a model with the highest state-of-the-art (SOTA) accuracy but to understand how data affects model performance. By defining a simple CNN model and keeping it fixed, we focused on using data augmentation techniques and hyperparameter tuning to improve accuracy.

💡 Skills

  • PyTorch
  • Python
  • MLFlow
  • Data Augmentation
  • Hyperparameter Tuning

⚖️ Dealing with imbalanced classes

The dataset contains images for 4 classes:

  1. Mild Demented
  2. Moderate Demented
  3. Non Demented
  4. Very Mild Demented

The class distribution is as follows:

Techniques to handle class imbalance

The dataset exhibits significant class imbalance. To address this, I employed and compared two data augmentation techniques:

  1. SMOTE (Synthetic Minority Over-sampling Technique): Equalized the instance counts across all classes.

  2. Manual Augmentation: Boosted the number of instances in underrepresented classes while retaining the overall class distribution. This approach maintained the relative differences in instance counts, offering a better representation of real-world conditions

Result: The model performs better when the class distribution closely mirrors real-life scenarios, providing a more accurate reflection of actual conditions.

🛠️Model

Several models developed on this dataset leveraged pre-trained architectures such as RESNET [1], VGG, and others, achieving remarkable accuracies nearing 94%. Conversely, a different approach employed a more straightforward Convolutional Neural Network model, yielding a validation set accuracy of approximately 80% [2]. This project endeavors to craft a simplified DNN model, boasting fewer parameters than expansive architectures like RESNET, all while upholding high accuracy standards.

References

[1] Alzheimer's Detection using DL

[2] Class Prediction with PyTorch