UK-based universities are working to create artificial intelligent (AI) devices that can improve treatment for cancer, intensive care, and chronic wound patients. They plan to develop continuous monitoring devices for critically ill patients that can administer medicines or adjust treatment automatically, using feedback from sensors. With such tools, researchers from Nottingham, Oxford, and Warwick Universities say they could provide better personalized, accurate, and timely care – ultimately improving patient outcomes.
Over the next three years, Stephen Morgan, from the faculty of engineering at Nottingham University, will lead a network of experts in healthcare technologies, focusing on sensing technologies, clinical care, control and modeling to discover tools and care methods.
“Although this is a very simple systems model, in practice, implementing personalized and optimized treatment using closed loop control is a challenging engineering and physical sciences problem requiring a multidisciplinary approach,” explains Morgan, a biomedical sensing expert.
The idea is to create a closed loop control system of feedback and intelligence, coupled with algorithms used in machine learning for AI. High-tech sensors will monitor patient vital signs while mathematical models inform designers as to how the body is predicted to work or how a disease model behaves.
The need for such technology is palpable, both globally and locally in the UK, say researchers, who cite some telling statistics:
- Over 100,000 patients are admitted to intensive care units in the UK per year;
- Survivors of critical illness commonly have a care requirement post-discharge from hospital and many experience reduced cognitive function;
- Meanwhile, a substantial proportion of the worldwide burden of cancer could be prevented with early detection and better treatment;
- Similarly, 200,000 patients in the UK have a chronic wound with the cost to the NHS conservatively estimated at £3 billion ($4.4 billion) per year, around 3 percent of the total estimated expenditure on health. With proper diagnosis and treatment, much of this burden could be avoided.
Morgan says the timing is right in terms of technology. “The advent of massive computer power, highly sensitive, specific and flexible sensors, and precisely delivered treatments has finally allowed closed-loop control systems to offer a revolutionary leap in medical treatment.”
He notes that hospital systems are also poised for such a disruption. “The intensive care unit, for instance, provides a highly controlled and technology-friendly environment that favors the development of closed loop control and so there are excellent opportunities to make rapid progress in optimizing treatment and advancing the proposed technology.”
Disrupting hospital treatments
The technology could have far-reaching applications, notes Morgan. A potential game-changer in intensive care could be a device that uses sensors to continuously monitor a patient’s breathing for irregularities and automatically tailor the amount of oxygen being provided, and to personalize the forces used to ventilate the patient’s lungs, in order to improve their condition.
Smart wound dressings, comprising fabric embedded with sensors that check for infection and even administer antibiotic drugs to accelerate healing, represent another example of the technology that may meet future healthcare challenges.
Calling all stakeholders
The Universities will host a series of three Grand Challenge workshops early in the project for academic experts, clinicians, patients, and other stakeholders. The workshops, in Wound Care, Critical Care and Cancer Treatment, will develop a shared understanding of the clinical challenges, technology, and techniques available.
They will also highlight challenges faced by patients and identify the most appropriate monitoring and treatments for closed loop control.
The goal of the workshops is to produce a roadmap for the development of closed loop control systems for optimizing treatment with a view to developing new research that address major healthcare challenges. Where there are gaps in knowledge or proof of concept data does not exist then these will be addressed through a series of eight feasibility projects funded by the network.
The network has wide support from seven U.K. academic institutions (Cranfield, Kings College London, Leicester, Manchester, Nottingham, Oxford and Warwick), three international academic institutions, four hospital trusts, five industry partners and other stakeholders such as patient groups, Medilink East Midlands, the East Midlands academic health sciences network and the Woundtec healthcare technology cooperative.