AI System Uses Deep Neural Networks to Detect Early Signs of Retinal Degeneration

A team of researchers has developed an Artificial Intelligence (AI) system that uses Deep Neural Networks (DNNs) to detect early signs of age-related macular degeneration (AMD), a common eye condition that can lead to severe vision loss. The AI system identifies a specific biomarker associated with early outer retinal atrophy, a risk factor for AMD progression. The system demonstrated promising results, detecting over 85% of biomarkers with less than one false-positive scan. While further research is needed, this technology could revolutionize AMD detection and treatment, potentially allowing for earlier intervention and slowing disease progression.

What is the Role of Artificial Intelligence in Detecting Retinal Degeneration?

Artificial Intelligence (AI) has been making significant strides in various fields, including healthcare. One such application is in the detection of retinal degeneration, specifically in patients with intermediate age-related macular degeneration (AMD). AMD is a common eye condition that affects the middle part of your vision, usually leaving the side vision intact. It is primarily associated with aging and can lead to severe vision loss in the worst cases.

A team of researchers, including Guilherme Aresta, Teresa Araujo, Gregor S Reiter, Julia Mai, Sophie Riedl, Christoph Grechenig, Robyn H Guymer, Zhichao Wu, Ursula Schmidt-Erfurth, and Hrvoje Bogunovic, have been working on a project that uses Deep Neural Networks (DNNs) to detect early signs of AMD. The team is affiliated with the Christian Doppler Laboratory for Artificial Intelligence in Retina, the Department of Ophthalmology and Optometry at the Medical University of Vienna, the Laboratory for Ophthalmic Image Analysis, the Centre for Eye Research Australia, The Royal Victorian Eye and Ear Hospital, and the Department of Surgery Ophthalmology at The University of Melbourne.

How Does the AI System Work?

The AI system developed by the team uses DNNs to detect and localize a specific biomarker associated with early outer retinal atrophy and a risk factor for progression to geographic atrophy in patients with intermediate AMD. This biomarker is the subsidence of the outer plexiform layer (OPL) on optical coherence tomography (OCT) images. OCT is a non-invasive imaging test that uses light waves to take cross-section pictures of your retina, the light-sensitive tissue lining the back of the eye.

The AI system predicts potential OPL subsidence locations on retinal OCTs. It consists of two modules: a detection module (DM) and a classification module (CM). The DM infers bounding boxes around subsidences with a likelihood score, while the CM assesses subsidence presence at the B-scan level. Overlapping boxes between B-scans are combined and scored by the product of the DM and CM predictions. The volumewise score is the maximum prediction across all B-scans.

What Were the Results of the Study?

The researchers used one development and one independent external dataset in their study, with 140 and 26 patients with AMD, respectively. The results were promising. The system detected more than 85% of OPL subsidences with less than one false-positive scan. The average area under the curve was 0.94 ± 0.03 for volume-level detection. Similar or better performance was achieved on the independent external dataset.

What Does This Mean for the Future of AMD Detection?

The results of this study suggest that DNN systems can effectively support automated detection and localization of OPL subsidence, a key biomarker in AMD. This could potentially revolutionize the way AMD is detected and treated, allowing for earlier intervention and potentially slowing the progression of the disease.

However, it’s important to note that while these results are promising, further research and development are needed to refine the system and validate its effectiveness in a broader range of patients and clinical settings. The researchers are likely to continue refining their AI system, and further studies will undoubtedly shed more light on the potential of AI in detecting AMD and other eye diseases.

How Can AI Improve Healthcare?

The use of AI in healthcare is not limited to AMD detection. AI has the potential to revolutionize many aspects of healthcare, from diagnosis to treatment. In the case of AMD detection, AI can help detect the disease at an earlier stage, potentially slowing its progression and improving patient outcomes.

Moreover, AI can also help reduce the workload of healthcare professionals by automating routine tasks, such as image analysis. This can free up more time for healthcare professionals to focus on patient care.

In conclusion, the study conducted by the team of researchers demonstrates the potential of AI in healthcare, specifically in the early detection of AMD. As AI continues to evolve and improve, its role in healthcare is likely to become even more significant.