It’s often been said that big data holds the keys to the transformation of healthcare delivery and disease management. And, it’s no surprise that healthcare systems across the world are looking closely at where data and advanced analytical techniques can have the biggest impacts on patient outcomes. Epilepsy is a very good example of a condition where the cure is likely to involve data and life sciences working closely together.
Epilepsy is a challenge in many regards. Despite having been first documented in 400BC, there’s still no absolute cure, and it continues to affect over 60 million people worldwide. Doctors must rely on patient-reported seizures and anecdotal evidence and often will need EEGs, MRIs and other kinds of scan for proper diagnosis – which can take 4-5 years. And, there are big gaps in what the medicines available today can achieve. One in three sufferers find that no medication works for them: there is currently no hope of becoming “seizure free”.
For those burdened by the disease, and especially for that “final third” of patients who can’t currently be treated, life is characterised by lack of control and unpredictability. Patients don’t know when they might experience a seizure and can’t control how people around them might respond when it happens. What’s more, though the healthcare system tries its best, it often falls short of patients’ expectations.
The big challenge in many areas of medicine – not least neurology – isn’t the lack of data, but an abundance of it. A significant, but largely overlooked phenomenon in medicine is the sheer volume of research literature and trial and health record data that is published every year. There is so much content that keeping abreast of the science is practically impossible:
If published data is one side of the treatment coin, the patient’s own perspective is the other. Factors such as how the patient is feeling that day, what they’ve had to eat, blood chemistry, and so on, are very important for treating neurological conditions like Parkinson’s disease and epilepsy. Wearables, mobile apps, and other forms of monitoring are invaluable in these cases, but again, this increasing patient generated data adds to the volume of data physicians have to wade through.
It is safe to say that advanced analytics, AI, and machine learning will have a vital role to play in the health system of the future. They can help doctors and other healthcare providers deliver that tailored experience to patients, draw on the latest science, and drive out uncertainty. It’s some way off, but it’s likely that this sort of data will be using this information via digital health solutions to improve outcomes and healthcare system efficiencies.
Where ‘dry’ data ‘and wet’ science meet
UCB, the biopharma company where I work, has a novel approach to developing medicines. Everything we do starts with one simple question: “how will this make a difference to patients’ lives?”. We start by analysing patient populations – and even ‘sub-populations’ – where there is a significant unmet need that we can address, and then research exactly what will maximise the value to patients on their individual journeys.
We’re very excited by the possibilities of combining the “wet” science of pharmaceuticals and human anatomy with the “dry” world of big data and IT. To explore how we can harness the power of big data and computational science, we are working with for instance Georgia Tech in the U.S. to couple that data and science with our established clinical expertise in epilepsy in an effort to find ways to identify the right solution for patients faster.
To better capture patients’ day-to-day experience of epilepsy, we’re investigating the use of wearable sensors to monitor for epilepsy episodes far more effectively than with EEGs alone. By measuring just three things – key brain signals, heart rate and accelerometry, clinicians can keep track of their patients’ conditions in real time. The SeizeIT consortium, in which UCB is a partner, plans to produce a wearable ear device which will stream data live to doctors and patients and help them predict when a seizure is likely to occur. We hope the added transparency means they spend more time living, and eventually less time in hospital.
It’s now possible to gather data from within the body, too. “Digestible sensors”, whose microelectronics can be harmlessly broken down by the body, can be incorporated into drug capsules, so patients and doctors can see exactly how the body absorbs the medication – helping patients see the optimal time to take their pills. Easy-to-follow but inaccurate dosing instructions, like “take twice a day with food”, are likely to become a thing of the past.
We’re drawing on all these initiatives – and others – to establish a comprehensive epilepsy patient database to both advance epilepsy science and improve the patient experience. We’re designing the database to encapsulate everything we need to measure and describe both the patient’s experience – and the biology – of their epilepsy. This includes demography, physiology (EEG and other sensors), medical history, molecular and genetic markets, imaging, medication, and other indicators.
The AI will see you now?
Are machine-based systems likely to replace human clinicians any time soon? The short answer is no: quite apart from the limitations of the technology available to us today, the simple fact is that disease is part of the human experience. We believe strongly that epilepsy sufferers are far more likely to gain when the treatment is discussed and agreed with human doctors. Yet, big data and advanced analytics must be part of our arsenal if we are to add value to patients along their patient journey and bring seizure freedom to the “final third” of patients – and add epilepsy to the growing list of conditions human ingenuity has overcome.