PAMBAYESIAN: PAtient
Managed decision-support using Bayesian networks
Queen Mary University of London
EPSRC
Project (Intelligent Technologies to Support Collaborative Healthcare
Programme)
PAMBAYESIAN (Patient Managed Decision-Support using Bayesian Networks) is a 3-year EPSRC
funded project awarded to Queen Mary (June 2017-May 2020) to develop a new generation
of intelligent medical decision support systems. The project focuses on home-based and wearable
real-time monitoring systems for
chronic conditions including rheumatoid arthritis, diabetes in
pregnancy and atrial fibrillation. The project has the potential to
improve the
well-being of millions of people. EPSRC is contributing a
grant of £1,538,497 towards the cost of the project, which is a
collaboration between researchers from both the School of
Electronic Engineering and Computer Science (EECS) and clinical
academics from the Barts and the London School of Medicine and
Dentistry (SMD). The project is also supported by digital health firms
that have
extensive experience developing patient engagement tools for clinical
development (BeMoreDigital, Mediwise, Rescon, SMART Medical,
uMotif,
IBM UK and Hasiba Medical).
The project is led by Prof Norman
Fenton with co-investigators: Dr
William Marsh, Prof Paul Curzon, Prof Martin Neil, Dr Akram Alomainy
(all EECS) and Dr Dylan Morrissey, Dr David Collier,
Professor Graham
Hitman, Professor Anita Patel, Dr Frances Humby, Dr Mohammed Huda, Dr
Victoria Tzortziou Brown (all SMD). The project will also include
four
QMUL-funded PhD students.
Press Release
October 2016: Queen Mary’s new £2 million
project to create intelligent medical decision support systems with
real-time monitoring for chronic conditions.
Queen Mary has been awarded a grant of £1,538,497 (Full economic
cost £1,923,122) from the EPSRC towards a major new collaborative
project to develop a new generation of intelligent medical decision
support systems. The project, called PAMBAYESIAN (Patient Managed
Decision-Support using Bayesian Networks) focuses on home-based and
wearable real-time monitoring systems for chronic conditions including
rheumatoid arthritis, diabetes in pregnancy and atrial
fibrillation. It has the potential to improve the well-being of
millions of people.
The project team includes researchers from both the School of
Electronic Engineering and Computer Science (EECS) and clinical
academics from the Barts and the London School of Medicine and
Dentistry (SMD). The collaboration is underpinned by extensive research
in EECS and SMD, with access to digital health firms that have
extensive experience developing patient engagement tools for clinical
development (BeMoreDigital, Mediwise, Rescon, SMART Medical,
uMotif, IBM UK and Hasiba Medical).
The project is led by Prof Norman Fenton with co-investigators:
Dr William Marsh, Prof Paul Curzon, Prof Martin Neil, Dr Akram Alomainy
(all EECS) and Dr Dylan Morrissey, Dr David Collier,
Professor Graham Hitman, Professor Anita Patel, Dr Frances Humby, Dr
Mohammed Huda, Dr Victoria Tzortziou Brown (all SMD). The project
will also include four QMUL-funded PhD students.
Background
Patients with chronic diseases must take day-to-day decisions about
their care and rely on advice from medical staff to do this. However,
regular appointments with doctors or nurses are expensive, inconvenient
and not necessarily scheduled when needed. Increasingly, we are seeing
the use of low cost and highly portable sensors that can measure a wide
range of physiological values. Such 'wearable' sensors could improve
the way chronic conditions are managed. Patients could have more
control over their own care if they wished; doctors and nurses could
monitor their patients without the expense and inconvenience of visits,
except when they are needed. Remote monitoring of patients is already
in use for some conditions but there are barriers to its wider use: it
relies too much on clinical staff to interpret the sensor readings;
patients, confused by the information presented, may become more
dependent on health professionals; remote sensor use may then lead to
an increase in medical assistance, rather than reduction.
The project seeks to overcome these barriers by addressing two key
weaknesses of the current systems:
Their lack of
intelligence. Intelligent systems that can help medical staff in making
decisions already exist and can be used for diagnosis, prognosis and
advice on treatments. One especially important form of these systems
uses belief or Bayesian networks, which show how the relevant factors
are related and allow beliefs, such as the presence of a medical
condition, to be updated from the available evidence. However, these
intelligent systems do not yet work easily with data coming from
sensors.
Any mismatch between the
design of the technical system and the way the people - patients and
professional - interact.
We will work on these two
weaknesses together: patients and medical staff will be involved from
the start, enabling us to understand what information is needed by each
player and how to use the intelligent reasoning to provide it. The
medical work will be centred on three case studies, looking at the
management of rheumatoid arthritis, diabetes in pregnancy and atrial
fibrillation (irregular heartbeat). These have been chosen both because
they are important chronic diseases and because they are investigated
by significant research groups in our Medical School, who are partners
in the project. This makes them ideal test beds for the technical
developments needed to realise our vision and allow patients more
autonomy in practice.
To advance the technology, we will design ways to create belief
networks for the different intelligent reasoning tasks, derived from an
overall model of medical knowledge relevant to the diseases being
managed. Then we will investigate how to run the necessary algorithms
on the small computers attached to the sensors that gather the data as
well as on the systems used by the healthcare team. Finally, we will
use the case studies to learn how the technical systems can integrate
smoothly into the interactions between patients and health
professionals, ensuring that information presented to patients is
understandable, useful and reduces demands on the care system while at
the same time providing the clinical team with the information they
need to ensure that patients are safe.
Summary
Medical decision support systems based on Bayesian Networks (BNs)
– that combine expert judgment with data and incorporate causal
knowledge – consistently outperform data-driven
‘score’ based systems commonly used for risk assessment.
Such BN models have been developed for a wide range of medical
applications and, in theory, could replicate much routine analysis
currently undertaken by GPs examining patients in a surgery. It is
increasingly well understood how the expert driven BN decision-support
models can be built, and progress is being made with their introduction
into routine clinical use. However, they remain primarily tools for
medical staff, rather than for patients, used as part of existing care
pathways with visits to a clinic, GP surgery or hospital. Moreover, the
data used is ‘low frequency’ (such as when a doctor enters
an observation) and decision-making is infrequent (often even just
‘one-time only’ such as whether or not a treatment should
be applied), at intervals determined by medical staff.
In parallel with the BN developments, it has become feasible to
continuously monitor patients with portable or wearable sensors,
covering an increasing range of physiological parameters. In theory
these monitoring devices, along with other home treatment devices,
could be provided with the 'intelligence' of an expert-built BN model
to support patients directly with more care at home and fewer visits to
medical centres. However, there are a range of complexity challenges
that must be overcome to build the necessary bridge between BN models
and devices usable by patients in the home. Thus, the goal of this
project is to overcome these complexity challenges to develop a
framework for a new generation of intelligent medical decision support
systems including their real-time monitoring – based on expert
built BNs.
The research framework will be informed by, and validated with, two
major medical case studies of chronic conditions where there is
potential for self-monitoring and treatment in the home, namely:
musclo-skeletal conditions and diabetes. A third case study on cardiac
arrhythmia will be used to show that the new framework is widely
applicable.
The project is a collaboration involving a diverse team of senior
researchers in the Faculty of Science and Engineering (who will lead
the different strands of the ICT research) and clinical academics from
the Barts and the London School of Medicine and Dentistry School (who
have proposed, and will supervise, the case studies, will support the
pathways to impact and engagement with medical practitioners, and
provide analysis of the potential benefits). QMUL’s belief in
both the importance of the research and capability of the team is
demonstrated by its commitment to fund four full time PhD students to
work on the case studies if the proposal is successful.
To ensure optimal pathways to impact, we will use a participatory
design approach to involve patient groups from the start, as we know
that technology must be usable by non-specialists. Also, safety (from
the perspective of the human interface of devices and their intrinsic
functionality) is a core research objective, meaning that the
traditional ‘regulatory barrier’ to actual deployment will
be easier to penetrate. We have also planned Patient and Public
Involvement workshops, focus groups, and consultation events with
medical practitioners with special focus on GPs. The team is also
internationally renowned for its strength in public engagement, and
will produce publications pitched at medical practitioners, regulators,
and the general public.
The Team
Prof Norman
Fenton (EECS, S&E): PI. He is Director of the Risk and
Information Management Research Group, whose expertise is in
risk
assessment and decision analysis
using BN models with data and judgment
Dr William Marsh
(EECS, S&E) co-PI leading the overall model
integration task
Prof Martin Neil
(EECS, S&E) co-PI leading the inference algorithm
work
Prof Paul Curzon (EECS, S&E) co-PI leading the HCI design work
and public engagement
Dr Akram Alomainy(EECS, S&E) co-PI leading the sensor
integrations work
Dr
Dylan Morrisey (SMD):
Consultant Physiopherapist in the Centre for Sport and Exercise
Medicine. Co-PI leading the MSK case study
Prof
Graham Hitman(SMD):
Director of the Blizard Institute.
Co-PI leading the diabetes case study
Dr
David Collier(SMD):
William Harvey Research Institute
(Heart Centre). Co-PI leading the heart-monitoring case study and GP outreach
Prof Anita Patel (SMD):
Health Economist in the Blizard
Institute. Co-PI providing a
study of the health economics aspects of potential implementations