Multi-Class Image Classification Model for Augmented Alzheimer’s MRI Using TensorFlow Take 3

Template Credit: Adapted from a template made available by Dr. Jason Brownlee of Machine Learning Mastery.

SUMMARY: This project aims to construct a predictive model using a TensorFlow convolutional neural network (CNN) and document the end-to-end steps using a template. The Augmented Alzheimer’s MRI dataset is a multi-class classification situation where we attempt to predict one of several (more than two) possible outcomes.

INTRODUCTION: The dataset contains four different classes of Alzheimer’s disease MRI images. The research team collected these images from various websites and manually verified each label. Furthermore, the research team produced many more augmented MRI images and added them to the training dataset to improve the model. The intent is to use the original image dataset for testing the new model’s capabilities.

ANALYSIS: The EfficientNetV2M model’s performance achieved an accuracy score of 99.05% after five epochs using the training dataset. When we applied the model to the validation dataset, the model achieved an accuracy score of 99.75%.

CONCLUSION: In this iteration, the TensorFlow EfficientNetV2M CNN model appeared suitable for modeling this dataset.

Dataset ML Model: Multi-Class classification with numerical features

Dataset Used: Augmented Alzheimer MRI Dataset

Dataset Reference:

One source of potential performance benchmarks:

The HTML formatted report can be found here on GitHub.