Imagine you’re a detective trying to solve a complex case. You have different pieces of evidence scattered throughout the city, each holding a clue to unlock the mystery. Now, think of retinal OCT images as the evidence and doctors as the detectives. Optical Coherence Tomography (OCT) is like a high-tech magnifying glass that allows doctors to examine the eye’s layers, helping them detect diseases like retinopathy. However, not all doctors have the same level of expertise, and some regions may lack medical resources. That’s where AI comes in! Researchers have developed a powerful algorithm called FN-OCT (Fusion Network for OCT) that uses artificial intelligence to analyze retinal OCT images with speed and accuracy. By combining three deep learning algorithms and using fusion strategies, FN-OCT improves the adaptability and precision of traditional classification algorithms. The results are impressive – FN-OCT achieved a 5.3% increase in prediction accuracy compared to previous methods. In addition, it accurately classified common retinal OCT diseases with an impressive accuracy rate of 92%. Gradient-weighted class activation mapping (Grad-CAM) was used to visualize the algorithm’s effectiveness. This breakthrough offers doctors a valuable tool and theoretical support to assist in the diagnosis of retinal OCT diseases. Check out the full research article to dive deeper into this amazing fusion of AI and healthcare!

Optical coherence tomography (OCT) is a new type of tomography that has experienced rapid development and potential in recent years. It is playing an increasingly important role in retinopathy diagnoses. At present, due to the uneven distributions of medical resources in various regions, the uneven proficiency levels of doctors in grassroots and remote areas, and the development needs of rare disease diagnosis and precision medicine, artificial intelligence technology based on deep learning can provide fast, accurate, and effective solutions for the recognition and diagnosis of retinal OCT images. To prevent vision damage and blindness caused by the delayed discovery of retinopathy, a fusion network (FN)-based retinal OCT classification algorithm (FN-OCT) is proposed in this paper to improve upon the adaptability and accuracy of traditional classification algorithms. The InceptionV3, Inception-ResNet, and Xception deep learning algorithms are used as base classifiers, a convolutional block attention mechanism (CBAM) is added after each base classifier, and three different fusion strategies are used to merge the prediction results of the base classifiers to output the final prediction results (choroidal neovascularization (CNV), diabetic macular oedema (DME), drusen, normal). The results show that in a classification problem involving the UCSD common retinal OCT dataset (108,312 OCT images from 4,686 patients), compared with that of the InceptionV3 network model, the prediction accuracy of FN-OCT is improved by 5.3% (accuracy = 98.7%, area under the curve (AUC) = 99.1%). The predictive accuracy and AUC achieved on an external dataset for the classification of retinal OCT diseases are 92 and 94.5%, respectively, and gradient-weighted class activation mapping (Grad-CAM) is used as a visualization tool to verify the effectiveness of the proposed FNs. This finding indicates that the developed fusion algorithm can significantly improve the performance of classifiers while providing a powerful tool and theoretical support for assisting with the diagnosis of retinal OCT.

Read Full Article (External Site)