In 2017, dementia overtook heart disease to become the biggest killer in the UK resulting in calls for increased funding into research for the condition.
Indeed, with soaring rates and a lack of clinical research when compared to other diseases such as cancer and heart disease, dementia statistics are likely to worsen as little progress is made into the condition.
Currently, one of the best ways to fight the condition is through early diagnosis, which, if accurate, can help slow down its progression and gives patients the information and support needed to prepare to live with dementia.
Now, a new start-up hopes to use its technology to improve dementia diagnosis rates across the UK and hopefully help improve the lives of dementia patients. Cognetivity has developed an integrated cognitive assessment (ICA) test which challenges large portions of the brain with natural images, and which is designed as a quick, easy-to-use tool for clinicians.
The idea is that the test, which only takes five minutes, will help clinicians diagnose patients with dementia earlier, enabling quicker treatment and ongoing monitoring.
Here, Digital Health Age sits down with Sina Habibi, founder and CEO of Cognetivity to discuss how the company’s test works and the state of dementia in the UK.
Could you tell me about how the Cognetivity test works?
Cognetivity’s ICA has a couple of key differences to existing cognitive tests, both in the design of the task itself, and by the use of Artificial Intelligence (AI) to process the results. In essence the test works by engaging a large proportion of the brain in a single challenging task – and this is very much in line with latest neuroscience theory, which has moved away from using tasks such as testing memory to the idea that in order to detect fine changes in performance you have to test large bodies of cells simultaneously. The ICA exposes the subject to an image for a very short period of time, and the subject has to determine something about the content of the image. The images used vary in composition and have a number of mathematical attributes, ranging from easy to process to difficult. The subject’s speed and accuracy are related to the properties of the images (in essence how difficult the image is to process in a small amount of time) and used to create a measure of the subject’s ability to process information. Also key to the approach is the use of Artificial Intelligence (AI) to analyses the data, and to compare the subject to known groups of patients, giving a likelihood of a subject belonging to a particular group, be that “normal” for the age group, suffering from Mild Cognitive Impairment (MCI) or displaying the characteristics of mild Alzheimer’s disease (AD) amongst others. The beauty of the AI approach is not only that it improves the test’s sensitivity, but also that the system can learn from further data, essentially improving its sensitivity and ability to categorise a subject as the system learns.
Cognetivity is used to detect early signs of dementia but could it (or similar platforms) be utilised to help treat the disease or improve cognitive function?
Currently we have no data to suggest that using the ICA leads to improvements in cognitive function; our focus is very much on detecting earlier than currently possible and the significant help this would allow people to receive from existing healthcare pathways and treatment methods. However, there is evidence that regular brain “exercise”, whether through traditional activities such as learning a language or a musical instrument, or through electronically delivered games and puzzles, can be helpful in enhancing a patient’s cognitive reserve and their coping strategies.
At what stage does dementia have to be detected for patients to benefit the most?
Really as early as possible – the current earliest stage that is considered detectable is the MCI stage, and the ICA has demonstrated its ability to detect the impairment associated with this disease stage. A diagnosis of MCI is certainly early enough to allow a patient the best chance of effective treatment to delay the onset and progression of symptoms, but if we are to really detect earlier we need to be able to pick up on the pre-MCI stage. Even in the absence of a disease modifying therapeutic there is much that can be done to help patients if diagnosed sufficiently early, with significant personal, social and economic benefits. Furthermore, when disease modifying drug treatments are eventually approved it will be vital that they are given to patients at the earliest stage of the disease, as all therapeutics currently in development are designed to stop or slow the progression of the disease rather than re-grow damaged cells associated with significant damage evident in the later stages of the disease.
What’s the current state of diagnosis and treatment for dementia like in the UK?
While the treatment of patients in the UK is of a high standard, given the tools available to specialist clinicians, there is an issue with diagnosis at primary healthcare level. The majority of dementia sufferers are either diagnosed too late for effective treatment to be administered, or often are not diagnosed at all. There is a major unmet need for an effective and easily administered cognitive test that can allow the early detection of the cognitive problems associated with dementias such as Alzheimer’s disease. If more patients were diagnosed earlier then the specialist treatment pathways already in place in the NHS would be much more effective at providing meaningful care to patients using existing methods.
At what level do you want Cognetivity to be deployed across say the NHS?
Ideally Cognetivity’s ICA would be used in a similar way to how blood pressure tests are currently used to screen patients for issued such as cardiovascular problems. The ICA test could be taken as part of routine assessments that can be carried our easily and cheaply, and which provide GPs with meaningful information that can be used to make a diagnosis and to refer to the appropriate specialist treatment. In the case of the ICA the test could be taken when a patient belonging to an at-risk group (for example over the age of 65, or even earlier if we are to truly detect the earliest stages of disease) visits the GP surgery for a general appointment. The test can then be taken unsupervised while the patient waits for their appointment, the results can then be immediately uploaded to Cognetivity’s cloud-based servers where the AI engine can analyse the results and transfer these directly to the patient’s records in the GP’s computer system. The GP can then make an assessment based on meaningful information about the patient’s cognitive state in the context of the other information they have. This would massively improve the rates of early detection of dementia at the first point of contact with the health system for all patients.
Following on from that, is there room for Cognetivity to be utilised in other sectors separate from healthcare, like in the workplace or other industries?
Absolutely. We are evaluating the ICA’s potential to be used as a fatigue monitor, allowing workers in dangerous or challenging conditions to be warned of the increased potential for accidents due to mistakes caused by fatigue. This could have a significant impact on preventing avoidable accidents that are due to tiredness in a whole swathe of industries and situations.
What barriers currently exist for companies wanting to develop solutions for dementia?
In order to properly develop a solution there are a number of regulatory hurdles that must be cleared if the solution is to be used in a clinical environment – that is to be used by doctors. There is the need to properly validate any solution scientifically though well designed and repeatable studies, and to publish these at conferences and peer reviewed journals. There is also the need to set up the quality-control infrastructure required to achieve regulatory approval, as well as the design and carrying out of the required clinical studies. Even in the case of diagnostic device there is a considerable burden of approval work that needs to be undertaken, often taking years and large amounts of money. The overall process involves a considerable amount of work on a number of fronts – so it’s not easy!
Is there a reason that digital solutions for dementia have been fairly non-existent when compared to other conditions like diabetes?
Dementia is not a single disease, rather a family of diseases of which Alzheimer’s disease is the commonest and best known. As a condition it has been traditionally underreported (basically explained away as being “just part of ageing”) and from a scientific perspective disease has also been poorly understood in terms of its mechanism and contributing factors. The fact that it has historically proven to be so hard to diagnose, even by experts, has made it a challenge to develop meaningful solutions to measure and manage. The whole picture is only recently beginning to become more clear and its complexity still leaves challenges to properly understand how best to help sufferers. Conditions such as diabetes have disease mechanisms and treatments that have better understood for much longer, that need relatively simple measurements to monitor – basically a much easier proposition technically to develop solutions for. The more that dementias are understood and more treatments become available the more effective solutions will be developed to help sufferers.
What do you hope to achieve with Cognetivity?
We basically want to help as many people as possible. There are far too many people suffering unnecessarily due to late diagnosis, or even no diagnosis at all, which leads to patients’ disease progressing more quickly than necessary, and untold anxiety for families and friends that is avoidable given an early diagnosis and proper medical support. If, through the unique technology we have developed, we can provide the “killer app” that allows widescale screening so that no-one slips though the gaps then we will have achieved something. We also think we have a role to play in the process of developing drugs to treat these diseases, by offering a sensitive and repeatable tool to detect the subtle and often hard to detect effects of drug candidates in clinical trials. There is a massive need for breakthroughs on this front and if we can be a small part of that it would be a major achievement for everyone involved.