"AN-EYE-4-DEMENTIA: Development of AN innovative intelligent EYE imaging solution for screening of DEMENTIA "
Lead Research Organisation:
University of Liverpool
Department Name: Eye and Vision Sciences
Abstract
Dementia is a major burden on global healthcare systems. Currently, there are ~50 million people with dementia worldwide which is projected to increase to 152 million by 2050. The total cost of dementia care in the UK is projected to increase from £34.7 billion (2019) to £94.1 billion (2040). Proposed diagnostic criteria focuses on the prodromal disease stage, namely mild cognitive impairment (MCI) (or early dementia) which occurs prior to functional disability and overt Alzheimer's dementia (AD). Timely diagnosis of MCI is crucial for effective treatment and the prevention of progression. Due to the subtle onset of symptoms and gradual cognitive decline, diagnosis remains challenging. Unfortunately, existing clinical tests for AD are either slow, costly, invasive, or detect the disease at an advanced stage.
The cornea is the most densely innervated tissue in the human body and can be directly visualised with specialist equipment. Our team and others globally have demonstrated nerve loss and additional pathological features in the cornea in neurodegenerative diseases, thus demonstrating its role as a biomarker in MCI/AD. The current method to visualise corneal nerve is by corneal confocal microscopy (CCM). However, CCM is technical, requiring specialist skills and direct contact with the cornea with topical anaesthesia, which limits its ubiquitous acceptance. Multiple studies including our published data have demonstrated that artificial intelligence (AI) greatly improves the diagnostic utility of CCM (corneal) images for the diagnosis of diseases including MCI and AD with high sensitivity and specificity (>0.8-0.9). An objective, non-invasive, non-contact, rapid and readily accessible method to detect MCI and predict AD is required to improve the management and clinical outcomes of patients.
Our interdisciplinary team of engineers, scientists and clinicians are exceptionally well placed to develop a game-changing integrated diagnostic imaging solution tailored to the needs of people with MCI/AD. Our objectives for this project are:
To develop a novel robust optical coherence tomography (OCT) imaging device for the detection of MCI and prediction of dementia. OCT is widely available but there is no commercial system capable of imaging the corneal nerves. Built on our recent prototype achieved by an MRC Confidence in Concept Award, this new device will be rapid (~2 seconds), non-invasive, and able to image the corneal nerves over a large field of view and depth of field whilst being tailored to nuances of patients with MCI/AD e.g. inattention, slower saccadic eye movements.
To develop new automated AI algorithms to diagnose MCI, and to predict dementia from MCI by analysing corneal nerve images. We aim to enhance our recent patented AI algorithms and produce a microservice capable of differentiating between patients with MCI and without MCI, and predicting the progression of MCI to AD at the time and place of patient care.
To validate the new technologies in a clinical study to ensure optimal diagnostic performance. Our integrated imaging and diagnostic solution will be externally validated in 154 people living with MCI and age-related cognitive decline (n=77 each cohort) with the MCI cohort followed-up over 18 months for AD. The diagnostic/predictive accuracy of the solution will be determined.
Detection and timely treatment of MCI prevents disability, benefiting both UK and global society. A robust commercialisation plan will be developed to accelerate the translation of this innovative solution for the benefit of patients, the NHS and the UK economy.
The cornea is the most densely innervated tissue in the human body and can be directly visualised with specialist equipment. Our team and others globally have demonstrated nerve loss and additional pathological features in the cornea in neurodegenerative diseases, thus demonstrating its role as a biomarker in MCI/AD. The current method to visualise corneal nerve is by corneal confocal microscopy (CCM). However, CCM is technical, requiring specialist skills and direct contact with the cornea with topical anaesthesia, which limits its ubiquitous acceptance. Multiple studies including our published data have demonstrated that artificial intelligence (AI) greatly improves the diagnostic utility of CCM (corneal) images for the diagnosis of diseases including MCI and AD with high sensitivity and specificity (>0.8-0.9). An objective, non-invasive, non-contact, rapid and readily accessible method to detect MCI and predict AD is required to improve the management and clinical outcomes of patients.
Our interdisciplinary team of engineers, scientists and clinicians are exceptionally well placed to develop a game-changing integrated diagnostic imaging solution tailored to the needs of people with MCI/AD. Our objectives for this project are:
To develop a novel robust optical coherence tomography (OCT) imaging device for the detection of MCI and prediction of dementia. OCT is widely available but there is no commercial system capable of imaging the corneal nerves. Built on our recent prototype achieved by an MRC Confidence in Concept Award, this new device will be rapid (~2 seconds), non-invasive, and able to image the corneal nerves over a large field of view and depth of field whilst being tailored to nuances of patients with MCI/AD e.g. inattention, slower saccadic eye movements.
To develop new automated AI algorithms to diagnose MCI, and to predict dementia from MCI by analysing corneal nerve images. We aim to enhance our recent patented AI algorithms and produce a microservice capable of differentiating between patients with MCI and without MCI, and predicting the progression of MCI to AD at the time and place of patient care.
To validate the new technologies in a clinical study to ensure optimal diagnostic performance. Our integrated imaging and diagnostic solution will be externally validated in 154 people living with MCI and age-related cognitive decline (n=77 each cohort) with the MCI cohort followed-up over 18 months for AD. The diagnostic/predictive accuracy of the solution will be determined.
Detection and timely treatment of MCI prevents disability, benefiting both UK and global society. A robust commercialisation plan will be developed to accelerate the translation of this innovative solution for the benefit of patients, the NHS and the UK economy.
