The Power of Statistics in Digital Health

You are invited to our next Open Network event on 15 September 2016, from 17:30 to 20:00, at the University of Warwick, which will discuss the important role of statistics/data in digital health. Presentations will be delivered by:

 

  • Jonny Edwards, LifeQueue
  • Neil Murphy, Converging Data
  • Prof Tom Nichols, University of Warwick
  • Dr John Thompson, Tracer Measurement Systems

 

The event is open to academia, public sector, industry and the third sector. Networking and refreshments will take place from 17:30, with presentations commencing at 18:00. If you would like to attend, please complete the online booking form by 13:00 on 13 September. Attendance is free and open to all, however, places will be prioritised for WIN members.

Jonny Edwards

Jonny Edwards

Using Machine Learning to Improve the Two Week Wait Pathway in Head and Neck Cancer

 

The two week wait (2ww) pathway exists to improve the clinical outcomes of patients with suspected cancer. In head and neck cancer, a difficult diagnosis domain, only 10% of referrals to the pathway require further treatment, significantly effecting the delivery of this service. At LifeQueue, we hope to build tools to address this problem by enhancing the way GPs refer, and how these referrals are prioritised in secondary care. To achieve this we have applied machine learning methods to data representing the symptoms of 5,000 referrals with 400 positive cancer diagnoses. This talk will elaborate on the methods explored, and discuss how they might be applied in other diagnosis situations.

 

Jonny is the CTO of LifeQueue, a start-up which has recently received VC funding from Dotforge Health and Data. He has twenty years’ experience in machine learning, starting with the neural network boom in the early 90’s. He has worked academically and commercially with images, sounds, natural language and more traditionally formatted data. Jony’s aim, together with his partner and consultant at the Leeds General Infirmary, Jim Moor, is to improve the delivery of oncology services, initially in the area of head and neck cancer. They aim to provide better prioritisation and queuing of patients using machine learning, and have several interesting results, which they are publishing through the summer. These results encompass patient diagnosis in head and neck cancer using classification algorithms, and also a more general study of queuing on cancer pathways. They are particularly interested in extending their work to diagnosis domains with difficult and vague presentation of symptoms.

Neil Murphy

Neil Murphy

Real-time data to run a more efficient hospital

 

NHS organisations are awash with data – but how many have a real-time, end-to-end view of what is happening with their services? Delays in an A&E department, for example, may not be the result of a shortage of frontline resources but due to issues further down the care pathway, for example a lack of beds. Converging Data solves the problem – with a secure, easy-to-use solution that can combine and analyse structured and unstructured data from multiple sources, including but not limited to HL7 data. It gives healthcare managers a real-time view of what is happening across a service, via dashboards that can track anything from admissions and discharges to transfers of patients between wards and services. Unlike other business intelligence tools, our approach is platform-agnostic and does not take your data outside your organisation. It can also handle requests ‘on the fly’ and does not need to have questions pre-programmed into it.

 

Neil is one of the founders of Converging Data, a company started in Australia by two former NHS professionals. They provide Big Data and Analytics services using a Machine Data platform called Splunk, and specialise in the Healthcare and Financial Services sectors. Their work is targeted on real-time data and how it can be applied to deliver Operational Intelligence and support immediate decision making. Neil was formally Lead architect and Strategy lead with the National eHealth Transition Authority based in Brisbane Australia. He joined Nehta from the NHS Technology Office, where he was Head of Platforms and Standards, prior to which he was Director of Business Infrastructure Support with GlaxoSmithKline. He has an MBA from Warwick and it’s a pleasure to be back!

 

www.convergingdata.com

Thomas Nichols

Professor Thomas Nichols

Machine Learning Methods for Population Neuroimaging

 

Thomas Nichols is a Wellcome Trust Senior Research Fellow in Basic Biomedical Science, a Professor and the Head of Neuroimaging Statistics at the Institute for Digital Healthcare, holding a joint position between Warwick Manufacturing Group & the Department of Statistics. Before joining the University of Warwick he was the Director of Modelling & Genetics at the GlaxoSmithKline Clinical Imaging Centre at Hammersmith Hospital in London, where he worked on statistical methods for fMRI in the context of clinical trials, and integrating genetic data into brain image analyses. Before coming to the UK he was an Associate Professor of Biostatistics at the University of Michigan, and in 2001 received his Ph.D. in statistics from Carnegie Mellon University where he also trained in cognitive neuroscience. He has been active in the field of functional neuroimaging since 1992, when he worked at the University of Pittsburgh’s PET Center as a programmer and statistician. Dr. Nichols’ research focuses on modelling and inference of neuroimaging data, including PET, fMRI & M/EEG.

John Thompson

Dr John Thompson

“Truth, damned truth and statistics” – How statistics can help you get real value from your data

 

This is an introductory overview of the various roles of statistics, not only in analysing both quantitative and qualitative data but also in designing experiments, clinical diagnostics and therapeutics trials, device testing, evaluation and validation, sampling strategy, in reliability and maintainability engineering, in quality and process control, in audit, forensically, etc. John will talk about modern statistical tools and methods, including graphical ones, in a simple way with practical examples, comparing their strengths and weaknesses with those of conventional, classical methods. His approach is designed to be understandable by those with little or no experience of statistics. John will briefly and graphically illustrate his talk about the various tools and methods of statistics and how he has applied and developed these with practical examples from some of his previous work, for example:

 

  • Developing methods for the HSE for exploring and analysing a large database of occupational exposures to dusts and toxic gases in the UK foundry industry over a 30yr period, including evaluating the effects of changes of exposure limits and their efficacy in different occupational practices in foundries.
  • Developing methods of data smoothing and fitting for a radically new type of a prototype high sensitivity and very rapid response lung function test device.
  • Evaluating reliability of haemodialysis equipment in a major West Midlands hospital, e.g., patterns of failures of equipment from different manufacturers over the full working life of such equipment to understand design flaws and the impact of user training or otherwise on failure patterns.

 

John is an entrepreneur, scientist, engineer and applied statistician who has worked in large and small industry, academia and the NHS. He has been involved in teaching, training, research and consultancy in applied statistics for 47 years, using it across a wide range of applications in medicine, industry, scientific and engineering R&D, quality/process control, environmental regulation, etc. He has over 70 peer reviewed publications, including five book chapters on applied statistics and chemometrics and a major HSE research report on statistical explorations of a large database of occupational exposures in the UK foundry industry. His areas of expertise include exploratory, robust, resistant and non-parametric statistics, statistical graphics, experimental design, statistical sampling strategy design, clinical diagnostics and therapeutics trial design and analysis, computer intensive resampling methods and, recently, compositional data modelling and analysis.

 

Email: tracerms@btinternet.com or statisticstraininguk@btinternet.com

Tel: 07787 552913

17:30

 

18:00

 

18:05

 

 

18:30

 

 

18:55

 

19:10

 

 

19:35

 

 

20:00

Arrivals and networking

 

Welcomes

 

Dr John Thompson

“Truth, damned truth and statistics” – How statistics can help you get real value from your data

 

Professor Thomas Nichols

Machine Learning Methods for Population Neuroimaging

 

Refreshments and networking

 

Neil Murphy

Real-time data to run a more efficient hospital

 

Jonny Edwards

Using Machine Learning to Improve the Two Week Wait Pathway in Head and Neck Cancer

 

End

Dr John Thompson

“Truth, damned truth and statistics” – How statistics can help you get real value from your data

Download the PDF

Watch the video on YouTube

 

Professor Thomas Nichols

Machine Learning Methods for Population Neuroimaging

Download the PDF

Watch the video on YouTube

 

Jonny Edwards

Using Machine Learning to Improve the Two Week Wait Pathway in Head and Neck Cancer

View the slides online

Watch the video on YouTube

 

Neil Murphy

Real-time data to run a more efficient hospital

Download the PDF

Watch the video on YouTube

 



Posted on by Martin Rowland

home about projects education research members news events contact login

The West Midlands Health Informatics Network is hosted by the Institute of Digital Healthcare (IDH) at University of Warwick.