AI Research Detects Glaucoma with 94 Percent Accuracy

Glaucoma affects more than 2.7 million people in the U.S. and is one of the leading causes of blindness in the world. To study how deep learning can help…

NVIDIA logo
NVIDIA
•
Nefi Alarcon
•2 min read•intermediate•
--
•View Original
KerasTensorFlow

Overview

Researchers from IBM and New York University have developed a deep learning framework that detects glaucoma with 94 percent accuracy using raw optical coherence tomographic (OCT) imaging. This advancement aims to improve the efficiency of glaucoma diagnosis, which currently affects over 2.7 million people in the U.S.

What You'll Learn

1

How to utilize deep learning frameworks for medical image analysis

2

Why accurate glaucoma detection is critical for patient outcomes

3

When to apply convolutional neural networks in healthcare diagnostics

Prerequisites & Requirements

  • Understanding of deep learning and neural networks
  • Familiarity with NVIDIA Tesla GPUs and Keras/TensorFlow frameworks(optional)

Key Questions Answered

What is the accuracy of the new deep learning framework for glaucoma detection?
The new deep learning framework developed by researchers achieves a glaucoma detection accuracy of 94 percent, significantly improving upon the previous state-of-the-art framework, which had an accuracy of 89 percent.
How does the deep learning framework detect glaucoma?
The framework detects glaucoma directly from raw optical coherence tomographic (OCT) imaging, which uses light waves to capture cross-section images of the retina, allowing for automated disease identification.
What dataset was used to train the neural network?
The neural network was trained on hundreds of OCT images from 624 patients, divided into 888 training samples, 112 validation samples, and 110 test samples, ensuring that eyes from the same patient were not included in different datasets.
What are the future research directions mentioned in the article?
Future work will focus on investigating potential biomarkers of glaucoma, which could enhance the understanding of the disease and improve diagnostic methods.

Key Statistics & Figures

Glaucoma detection accuracy
94 percent
Achieved by the new deep learning framework developed by researchers.
Previous state-of-the-art accuracy
89 percent
The accuracy of the previous framework for glaucoma detection.
Number of patients in the dataset
624 patients
The dataset used to train the neural network.
Training samples
888 training samples
Part of the dataset split for training the model.

Technologies & Tools

Some links below are affiliate links. We may earn a commission if you make a purchase.

Hardware
Nvidia Tesla Gpus
Used for training the deep learning framework.
Software
Keras
A deep learning framework used in conjunction with cuDNN for model training.
Software
Tensorflow
Another deep learning framework utilized for training the neural network.

Key Actionable Insights

1
Implementing deep learning frameworks like Keras and TensorFlow can significantly enhance the accuracy of medical diagnostics.
By leveraging advanced algorithms, healthcare professionals can achieve better diagnostic outcomes, particularly in areas like glaucoma detection.
2
Utilizing NVIDIA Tesla GPUs can improve the efficiency of training deep learning models.
These powerful GPUs are designed for high-performance computing tasks, making them ideal for processing large datasets in medical imaging.
3
Incorporating previously unused image regions in analysis can lead to improved diagnostic accuracy.
This approach can help uncover new insights and patterns that traditional methods may overlook, enhancing overall patient care.