Weakly-supervised detection of AMD-related lesions in color fundus images using explainable deep learning
Authors:
José Morano,
Álvaro S. Hervella,
José Rouco,
Jorge Novo,
José I. Fernández-Vigo,
Marcos Ortega
Abstract:
Age-related macular degeneration (AMD) is a degenerative disorder affecting the macula, a key area of the retina for visual acuity. Nowadays, it is the most frequent cause of blindness in developed countries. Although some promising treatments have been developed, their effectiveness is low in advanced stages. This emphasizes the importance of large-scale screening programs. Nevertheless, implemen…
▽ More
Age-related macular degeneration (AMD) is a degenerative disorder affecting the macula, a key area of the retina for visual acuity. Nowadays, it is the most frequent cause of blindness in developed countries. Although some promising treatments have been developed, their effectiveness is low in advanced stages. This emphasizes the importance of large-scale screening programs. Nevertheless, implementing such programs for AMD is usually unfeasible, since the population at risk is large and the diagnosis is challenging. All this motivates the development of automatic methods. In this sense, several works have achieved positive results for AMD diagnosis using convolutional neural networks (CNNs). However, none incorporates explainability mechanisms, which limits their use in clinical practice. In that regard, we propose an explainable deep learning approach for the diagnosis of AMD via the joint identification of its associated retinal lesions. In our proposal, a CNN is trained end-to-end for the joint task using image-level labels. The provided lesion information is of clinical interest, as it allows to assess the developmental stage of AMD. Additionally, the approach allows to explain the diagnosis from the identified lesions. This is possible thanks to the use of a CNN with a custom setting that links the lesions and the diagnosis. Furthermore, the proposed setting also allows to obtain coarse lesion segmentation maps in a weakly-supervised way, further improving the explainability. The training data for the approach can be obtained without much extra work by clinicians. The experiments conducted demonstrate that our approach can identify AMD and its associated lesions satisfactorily, while providing adequate coarse segmentation maps for most common lesions.
△ Less
Submitted 4 December, 2022; v1 submitted 1 December, 2022;
originally announced December 2022.
Improving AMD diagnosis by the simultaneous identification of associated retinal lesions
Authors:
José Morano,
Álvaro S. Hervella,
José Rouco,
Jorge Novo,
José I. Fernández-Vigo,
Marcos Ortega
Abstract:
Age-related Macular Degeneration (AMD) is the predominant cause of blindness in developed countries, specially in elderly people. Moreover, its prevalence is increasing due to the global population ageing. In this scenario, early detection is crucial to avert later vision impairment. Nonetheless, implementing large-scale screening programmes is usually not viable, since the population at-risk is l…
▽ More
Age-related Macular Degeneration (AMD) is the predominant cause of blindness in developed countries, specially in elderly people. Moreover, its prevalence is increasing due to the global population ageing. In this scenario, early detection is crucial to avert later vision impairment. Nonetheless, implementing large-scale screening programmes is usually not viable, since the population at-risk is large and the analysis must be performed by expert clinicians. Also, the diagnosis of AMD is considered to be particularly difficult, as it is characterized by many different lesions that, in many cases, resemble those of other macular diseases. To overcome these issues, several works have proposed automatic methods for the detection of AMD in retinography images, the most widely used modality for the screening of the disease. Nowadays, most of these works use Convolutional Neural Networks (CNNs) for the binary classification of images into AMD and non-AMD classes. In this work, we propose a novel approach based on CNNs that simultaneously performs AMD diagnosis and the classification of its potential lesions. This latter secondary task has not yet been addressed in this domain, and provides complementary useful information that improves the diagnosis performance and helps understanding the decision. A CNN model is trained using retinography images with image-level labels for both AMD and lesion presence, which are relatively easy to obtain. The experiments conducted in several public datasets show that the proposed approach improves the detection of AMD, while achieving satisfactory results in the identification of most lesions.
△ Less
Submitted 22 May, 2022;
originally announced May 2022.
