Multi-modal classifier fusion with feature cooperation for glaucoma diagnosis

Abstract

Background: Glaucoma is a major public health problem that can lead to an optic nerve lesion, requiring systematic screening in the population over 45 years of age. The diagnosis and classification of this disease have had a marked and excellent development in recent years, particularly in the machine learning domain. Multimodal data have been shown to be a significant aid to the machine learning domain, especially by its contribution to improving data driven decisionmaking. Method: Solving classification problems by combinations of classifiers has made it possible to increase the robustness as well as the classification reliability by using the complementarity that may exist between the classifiers. Complementarity is considered a key property of multimodality. A Convolutional Neural Network (CNN) works very well in pattern recognition and has been shown to exhibit superior performance, especially for image classification which can learn by themselves useful features from raw data. This article proposes a multimodal classification approach based on deep Convolutional Neural Network and Support Vector Machine (SVM) classifiers using multimodal data and multimodal feature for glaucoma diagnosis from retinal fundus images from RIM-ONE dataset. We make use of handcrafted feature descriptors such as the Gray Level Co-Occurrence Matrix, Central Moments and Hu Moments to cooperate with features automatically generated by the CNN in order to properly detect the optic nerve and consequently obtain a better classification rate, allowing a more reliable diagnosis of glaucoma. Results: The experimental results confirm that the combination of classifiers using the BWWV technique is better than learning classifiers separately. The proposed method provides a computerized diagnosis system for glaucoma disease with impressive results comparing them to the main related studies that allow us to continue in this research path.

SCRD-Net: A Deep Convolutional Neural Network Model for Glaucoma Detection in Retina Tomography

“…ese results show that it is beneficial to treat different model features independently. Moreover, HRT data with multiple modalities plays a crucial role in clinical diagnosis [58]. More importantly, our proposed SCRD-Net model achieved a sensitivity of 78.3% at a specificity of 0.95, demonstrating an improvement of 7.3% when the structural and textural features are combined.…”

A Comparative Analysis of Glaucoma Feature Extraction and Classification Techniques in Fundus Images

“…The classification was performed with Randon Forest (RF). Another study that combined nonstructural information with features from CNNs was Benzebouchi et al [10], in which the authors proposed a multimodal representation based on extracting features from different CNNs with features non-structural from GLCM, Hu Moments, and Central Moments.…”

Multimodal Information Fusion for Glaucoma and Diabetic Retinopathy Classification

“…The intermediate fusion strategy fuses features before the final decision layer. In contrast, late fusion fuses the decision results, ignoring any correlation between the different modalities [3].…”