How AI Helps Detect Glaucoma Before Vision Loss Begins
Artificial Intelligence (AI) is making significant strides in healthcare, particularly in detecting glaucoma. Let us explore how AI is transforming glaucoma detection and what the future of AI holds.
Have you been advised eye surgery elsewhere?
Get a free second opinion from Surya Eye Institute specialists before you decide.
Understanding Glaucoma – The Silent Thief of Sight
Glaucoma is one of the leading causes of blindness worldwide with challenges persisting in early diagnosis, disease progression.
Glaucoma is characterized by progressive structural loss of retinal ganglion cells (RGCs) of the eye. It can cause blindness by damaging the optic nerve, often linked to high intraocular pressure.
Recent population studies have shown that in India, glaucoma in adults aged 40 years and above is estimated to be between 2.7 and 4.3%. Glaucoma is estimated to affect 27.8 million more people by 2040 in Asia, and India will share a major burden out of this. This makes it a disease of concern for our population.
Why Glaucoma Often Goes Undiagnosed
In India, the proportion of undiagnosed glaucoma is estimated at 90%, which means that a major chunk of this population is without proper care and treatment for a diseases they don’t yet know they have
There are 3 main reasons why glaucoma often goes undiagnosed:
- Asymptomatic nature in the early course of glaucoma is the main reason for glaucoma remaining undiagnosed. Until the advanced stages of the disease, there are no symptoms which makes the early detection of glaucoma more challenging.
- In India, there are low levels of awareness and poor access to glaucoma diagnostic and therapeutic services.
- Moreover, considerable expertise is required to perform the appropriate clinical exam and to interpret several specialized tests for detection of glaucoma.
By the time vision loss becomes apparent, significant and irreversible damage may have already occurred, making early detection crucial.
How Early Detection Can Save Vision
Early detection is the key to prevention of glaucoma associated blindness. This condition is particularly concerning as it can lead to complete blindness if not managed properly.
Early detection of glaucoma can save a person’s vision in the following ways:
- Early detection reduces the risk of severe vision loss that can occur if the disease progresses unchecked. This can significantly halt or slow the progression of the disease.
- Earlier detection of the disease facilitates early intervention. This early intervention is crucial in preserving the patient’s vision and preventing further deterioration.
- By reducing the likelihood of severe vision impairment, patients can continue to engage in daily activities and maintain their quality of life
Role of Artificial Intelligence in Glaucoma Detection
Artificial intelligence (AI) is an emerging technological advancement in the field of eye care.
AI uses computational platforms, generation of large annotated ocular images and AI algorithms.
Although glaucoma is asymptomatic in its early stages, structural changes in the macula and retinal nerve fiber layer (RNFL) precede the onset of clinically detectable vision loss. This presents an opportunity for using AI to detect these hidden structural changes. There have been developments of AI strategies using fundus photography, optical coherence tomography (OCT) imaging, and perimetry in glaucoma diagnosis and the detection of glaucoma progression
AI-Powered Optic Nerve Imaging
Optic nerve damage or destruction is a key feature of glaucoma which is one of the foci of AI detection. Optical coherence tomography (OCT) imaging along with AI is currently being used to provide three-dimensional views of retinal layers and optic nerve head structures.
AI helps with measurement of the following indicators of glaucoma related to the optic nerve:
- Quantifying glaucomatous optic nerve injury from fundus photographs : This involves features from spectral-domain OCT images being used to train an AI algorithm to predict neuroretinal damage from optic disc photographs.
- The retinal nerve fiber layer (RNFL) thickness: This remains the most common parameter utilized for glaucoma diagnosis. Analyzing OCT imaging data from peripapillary RNFL thickness maps has been shown to be helpful for discriminating between glaucomatous and normal eyes.
- Volume of the optic nerve head: OCT volumes of the optic nerve head have been able to classify eyes as normal or glaucomatous using AI.
Machine Learning for Visual Field Analysis
Machine Learning (ML) is a subset of AI which provides training to the AI algorithm using the collected information, making it more precise and detail-oriented. ML is an extension of statistical modeling involve the separation of “glaucoma” from “not glaucoma”
Visual field (VF) analysis, which involves the measurement of a person’s eye field of the surrounding area, is a standard assessment key to the detection of glaucoma. Using standard automated perimetry (SAP) perimetry data, AI can classify the severity of field loss from early to advanced damage from a single field analysis.
ML has shown to be capable of the identification of typical patterns of VF loss on par with that seen with clinical experience.
Quantification of VFs in the form of glaucoma-induced patterns of VF loss could also facilitate diagnosis and assist therapy adjustment and prognosis. It could also be used to plan optimization based on the shape, type, and depth of defect with consideration of the patient’s quality of life.
AI Prediction of Risk Based on Retinal Scans
OCT provides substantial retinal structural information in three dimensions and its quantification can be highly useful. For these reasons, OCT quantification and interpretation has always been an active area of research. These OCT scans when combined with AI can be used to identify risk of developing glaucoma or risk of progression of glaucoma by studying the anatomical structures involved.
AI has demonstrated increasing superiority in the analysis of genetic (molecular and omics) data to facilitate glaucoma diagnosis and risk stratification. This uses Ribonucleic acid (RNA) sequencing datasets to identify the genes (ENO2, NAMPT, and ADH1C) relevant to glaucoma.
Benefits of AI Screening for Patients and Doctors
Artificial intelligence offers a potential paradigm shift in glaucoma care. AI systems can learn complex patterns from large datasets and might detect subtle glaucomatous changes beyond the threshold of human observation, this offering several benefits for both patients as well as doctors.
Faster, More Accurate Diagnoses
With the increasing prevalence of glaucoma, there is a critical need for advanced diagnostic tools that can identify the disease earlier and more accurately.
AI tools promote objectivity, improve consistency in glaucoma assessment. There is also the potential outperformance of clinicians by a trained AI algorithm in making a diagnosis of glaucoma.
Affordable Screening for High-Risk Populations
The detection of glaucoma progression at an earlier stage may enable earlier intervention and therefore further reduce the risk of patients developing glaucoma-related visual impairment .
There is a huge opportunity to optimize both resource utilization and the workload of clinicians, by offering affordable screening for high-risk populations thus enabling the provision of high-quality glaucoma care.
AI algorithms may help to augment referral priorities in order to efficiently triage those patients who need to be seen by a specialist, and those who do not.
AI as a Support Tool, Not a Replacement for Experts
The algorithms reported to date in research studies have been trained and validated on specific patient cohorts or on collections of disc photographs. They have been shown to perform comparably to clinicians with specificities of 90% and 91%, respectively,
As glaucoma is a progressive disease, absolute confirmation of diagnosis may only be possible in some cases through the evaluation. Furthermore, current published research has not been designed to account for the natural variability that exists within populations, including the impact of ethnicity, extremes of refractive error, and age. That’s why AI should be used as a support tool rather than a replacement for experts for glaucoma detection.
Future of AI in Glaucoma Care
In the future, AI may become an essential adjunct to glaucoma diagnosis, which will not replace the clinical skills but facilitate decision making.
AI algorithms may be developed to serve as a glaucoma referral refinement scheme to manage referrals from community based screening programs and optometrists
In cases of an established diagnosis of glaucoma, AI strategies may have the potential to function as an additional adjunct to the glaucoma assessment in making a clinical diagnosis in more challenging cases by helping to support the diagnosis or reject it.
How Hospitals Like Surya Eye Are Adopting These Innovations
India currently has a lack of availability of eyecare services and also poor utilisation.
Surya Eye hospital has been implementing strategies to improve access to glaucoma care, in fact eyecare as whole, and more specifically affordability issues in glaucoma. By incorporating the latest advances in imaging technology and perimeter into AI algorithms, many eye hospitals have integrated these innovations to transform eye care for their patients.
ML approaches have been successfully employed in discriminating between normal eyes and those with glaucoma progression with fewer tests and in a shorter timescale. Thus, early detection of glaucoma is being made possible.
Global Research and Indian Success Stories
Global research on AI tools in glaucoma detection have focused on various aspects.
These include:
- comparison of the accuracy of machine learning algorithms against gonioscopy in detecting angle closure in patients with glaucoma.
- quantification of the performance of artificial intelligence (AI) in detecting glaucoma with OCT
- comparison of machine learning versus ophthalmologists in glaucoma diagnosis on fundus examinations
While the research on AI tools in glaucoma detection is still underway, there have been massive strides in AI technology made.
Conclusion – Smart Technology for a Brighter Future
Blindness from glaucoma is preventable. In India, where the majority of glaucoma still remain undiagnosed, AI offers tremendous potential in detection.
Artificial intelligence has increasingly contributed to glaucoma management, with promising developments in early detection, monitoring of disease progression. Ultimately, AI-driven tools are expected to refine precision medicine in glaucoma and improve long-term patient outcomes, this benefitting both the patients and the doctors.
FAQs:
Yes, AI can detect glaucoma early by identifying the minute structural changes that indicate damage to the eye structures which are easily missed in the early stages.
AI has shown to be accurate in diagnosing glaucoma with a percentage of over 90% in most research studies.
Yes, AI glaucoma screening is currently available in India at Surya Eye hospital.
AI offers several advantages over traditional eye tests such as higher accuracy, faster detection and improved accessibility.
Glaucoma can be detected using AI technology in its early stages i.e. as soon as the minute structural changes in the eye structures occur.
