‘Enormous change looks inevitable. Investors hope for billion-dollar health-tech “unicorns”. Payers eye equally sizeable savings. Amid such talk it is worth remembering that the biggest winners from digital health care will be the patients who receive better treatment, and those who avoid becoming patients at all.’ – The Economist
Machine learning and Artificial Intelligence (AI) continue to transform many aspects of our lives. The potential gains in healthcare are enormous. Although investment in digital healthcare start-ups has doubled since 2013, progress is slow, in part because of regulatory and cost hurdles.
Machine learning in healthcare means that organisations can benefit from evolving technological capabilities. This enables accurate accurate and real-time decision making, improving overall efficiency and reducing costs.
From robotic surgery to computer-led well-being, the future could see doctors predicting which of their patients might get ill and taking a more proactive, preventative approach. The digital health revolution is already here but it is unevenly distributed.
DeepMind Health and Data Privacy
Acquired by Google in 2014, DeepMind understands the role data plays in getting patients from test to treatment. DeepMind Health analyses that data and sends it to the right clinician as quickly as possible.
Following criticism in 2016, DeepMind is building a blockchain distributed ledger system to monitor patient data and allow healthcare professionals to ensure records are kept securely.
This allows the system to monitor patients over time. For example, a patient could use DeepMind to track blood test results and see testing profiles. Every entry is allocated a hash value to ensure easy tracking.
No records can be amended secretly as it would alter the hash value and interrupt the data trail. Only authorised healthcare staff can access data and an alarm can be raised if there is any concern about unauthorised access.
Microsoft InnerEye diagnostic imaging
Microsoft’s InnerEye initiative is building sophisticated imaging diagnostic tools using AI. These tools will increase targeted and effective cancer treatment.
‘We are trying to change the way research is done on a daily basis in biology,’ said Jasmin Fisher, a biologist by training who works in Microsoft Research’s Cambridge lab.
Computer vision promises a major breakthrough for machine learning in healthcare, and InnerEye will focus its efforts on treating tumours. Its decision forecasts will assist radiologists and oncologists provide better more consistent care. It will also make these highly-trained professionals more efficient, allowing them to focus more time on patient care.
InnerEye will examine data produced during CT scans. As CT scans produce images that look through bones and tissue, the images are detailed and layered. Identifying concerns in each of those intricate layers could take hours of an oncologist’s time. InnerEye can reduce that time to less than one minute.
A group of ex-Microsoft employees and researchers from the University of Washington are aiming to solve the problem of missed diagnoses with their start-up, KenSci.
By applying machine learning in healthcare to the challenge of patient diagnosis, the digital health start-up believes it can efficiently identify health risks. By identifying patterns and high-risk markers KenSci models disease progression and critical illness within patient groups.
SkinScanner: disease prevention by phone
SkinScanner is an image processing software using machine learning. Its classification algorithm allows patients to submit an image of their skin condition and be compared to 23 disease classes for a visual match.
The deep learning software assists dermatologists in determining a range of conditions, including skin cancer. The company hopes that it will help people who may, at first, be too anxious to visit a doctor: early diagnosis via a smartphone could be lifesaving.
The Stanford researchers that trained the machine learning algorithm aimed to create a technology as reliable as a human dermatologist. Once complete, it was tested against 21 board-certified dermatologists and proved equally efficient in the diagnosis of skin cancers.
RedPixie has launched health and well-being software, Wellband. It is being developed in conjunction with a major healthcare provider to transform the way that they look after their patients.
Through a wearable device, Wellband takes data from a user’s daily behaviour and builds a pattern of normality. When an unusual event occurs, such as a fall or a sudden drop in heartrate, Wellband can call for help.
‘Wellband is all about keeping people well, and when they are not well, doing something with that data’, says Dirk Anderson, RedPixie’s CTO.
The potential implications of Wellband are revolutionary for healthcare in the UK. By knowing more about the health of the market, Wellband is better able to service individuals’ needs and make efficient use of scarce healthcare resources.
Machine learning in healthcare: transformative
‘In healthcare, machine learning could help provide more accurate diagnoses and more effective healthcare services, through advanced analysis that improves decision-making’, according to a report on AI by The Royal Society.
Machine learning in healthcare provides a massive opportunity for advances in wellness. While some of this technology would have seemed like science fiction even a few years ago, it now feels like we’re just at the beginning of a wellbeing revolution.