[Paper Review] Detection of Alzheimers Disease from MRI using Convolutional Neural Network with Tensorflow

GuruRaj Awate, Sunil L. Bangare+2 more|arXiv (Cornell University)|Jun 26, 2018

Brain Tumor Detection and Classification41 citations

TL;DR

The paper proposes using a convolutional neural network implemented in TensorFlow to classify Alzheimer's disease from MRI scans, trained on the OASIS dataset.

ABSTRACT

Nowadays, due to tremendous improvements in high performance computing, it has become easier to train Neural Networks. We intend to take advantage of this situation and apply this technology in solving real world problems. There was a need for automatic diagnosis certain diseases from medical images that could help a doctor and radiologist for further action towards treating the illness. We chose Alzheimer disease for this purpose. Alzheimer disease is the leading cause of dementia and memory loss. Alzheimer disease, it is caused by atrophy of the certain brain regions and by brain cell death. MRI scans reveal this information but atrophy regions are different for different people which makes the diagnosis a little trickier and often gets miss-diagnosed by doctors and radiologists. The Dataset used for this project is provided by OASIS, which contains over 400 subjects 100 of which having mild to severe dementia and is supplemented by MMSE and CDR standards of diagnosis in the same context. Enter CNN, Convolutional Neural Networks are a hybrid of Kernel Convolutions and Neural Networks. Kernel Convolutions is a technique that uses filters to recognize and segment images based on features. Neural Networks consist of neurons which are loosely based on human brains neuron which represents a single classifier and interconnected by weights, have different biases and are activated by some activation functions. By using Convolutional Neural Networks, the problem can be solved with minimal error rate. The technologies we intend to use are libraries like CUDA CuDNN for making use of GPU and its multiple cores-parallel computing to train models while giving us high performance.

Motivation & Objective

• Motivate automatic diagnosis of diseases from medical images to assist doctors and radiologists.
• Apply CNNs to MRI data to detect Alzheimer's disease.
• Leverage GPU acceleration (CUDA CuDNN) for efficient training.
• Demonstrate feasibility of hybrid image analysis techniques for brain atrophy patterns.

Proposed method

• Train a Convolutional Neural Network on MRI scans using TensorFlow.
• Utilize CUDA CuDNN to exploit GPU parallelism for faster training.
• Rely on the OASIS dataset, which includes MMSE and CDR standards for diagnosis.
• Explain core CNN concepts: filters, feature maps, and learned weights for image classification.

Experimental results

Research questions

• RQ1 Can a CNN trained on MRI data accurately differentiate between Alzheimer's disease and non-demented cases using the OASIS dataset?
• RQ2 How does GPU-accelerated CNN training perform for this medical-image diagnosis task?
• RQ3 What role do MMSE and CDR standards play in validating the CNN's predictions?
• RQ4 What are the limitations and diagnostic challenges due to brain region atrophy variability among individuals?

Key findings

• The study demonstrates using a CNN with TensorFlow to tackle Alzheimer's detection from MRI scans.
• The OASIS dataset provides over 400 subjects, including roughly 100 with dementia, along with MMSE and CDR diagnoses.
• The approach leverages GPU acceleration (CUDA CuDNN) to enhance training performance.