£1m Funding Award to Improve Primary Care using AI

[Manchester, UK, May 28th] Spectra Analytics is delighted to announce a £1m funding award from InnovateUK to use Artificial Intelligence (AI) to improve shutterstock_284499977NHS Primary Care. The PA
TCHS project, in collaboration with the University of Manchester, Salford Clinical Commissioning Group, and Manchester Health & Care Commissioning, aims to ease access to GP care for urgent cases and reduce GP waiting times. PATCHS is expected to improve patient satisfaction and reduce pressure on staff. Trials are taking place in interested GP practices over the next 2 years.

Primary care is at breaking point nationally as the demand for GP appointments continues to rise whilst under-investment continue
s to squeeze resources. Consequently, waiting times have risen, as have instances of patients being unable to see their GP when required. This is potentially endangering patients and putting immense pressure on GPs.

Effective triage procedures – directing patients to the appropriate care based upon clinical needs – can improve the situation. Research suggests 27% of GP consultations are potentially avoidable with patients better seen by another healthcare professional such as a nurse, pharmacist, or mental health worker.

GP practices do not have standardised triage processes, so the approach can vary significantly between practices. Some have adopted triage technologies but Dr Ben Brown, Wellcome Trust Fellow at the University of Manchester and practising GP, says “Current triage technologies are gnerally not very sophisticated, which means that they either make a large number of incorrect triage decisions, putting patients at risk, or rely heavily on medical staff, increasing workload. Furthermore, the effectiveness of these technologies has not been rigorously evaluated”.

PATCHS uses AI to analyse patient symptoms, medical history, and a range of other factors such as weather and pollution, to determine the type and urgency of a GP appointment request. It then directs patients to the appropriate care professional. CEO of Spectra Analytics, Dr Marcus Ong, says that “This data-driven approach potentially allows the AI to analyse a far broader range of factors that can influence the urgency of a patient’s GP appointment request. It also means that we can tailor triage decisions for individual patients.

Trials are being held in Salford and Manchester over the next two years. Kirstine Farrer, Head of Innovation and Research at Salford CCG, said “Digital innovation is of utmost importance to Salford CCG, and we are keen to support projects that have the potential to improve the efficiency and quality of primary care. We are keen to assess how PATCHS can help manage patient demand and reduce GP workload.” 

If you are interested in finding out more about PATCHS please contact Dr Marcus Ong at Spectra Analytics, email info@spectraanalytics.comor call 0203 968 7800.

ORCA – Your Resilience Tool

  • How resilient is your organisation?
  • Would an outage impact your revenue?
  • What is the impact on 3rd party suppliers?
  • What is the likely impact on your reputation?

These are very hard questions to answer and quantify.

Our new resilience tool is specifically designed to address these challenges offering:

  • Instant insights
  • Peer benchmarking
  • Effective decision making
  • Cost effective mitigations

It is designed to protect your revenue and reputation.

Check out the video below….

Spectra Expands Leadership Team

We are delighted to announce that Spectra are expanding our leadership team. Our Chief Science Officer, Dr Dan Sprague, is extending his role to include Chief Technology Officer, and we have three new team members: Chris Crowther (Chief Information Officer), Steve Williams (Head of Business Development) and Dr Ben Brown (Chief Medical Officer). These appointments will significantly strengthen Spectra’s expertise in Security and Defence, Finance and Healthcare. Their extensive experience will also support Spectra’s ambitious growth plans.

Dr Daniel Sprague

Chief Technology Officer & Chief Science Officer

dan_sprague2In his CTO role, Dan leads Spectra’s technology and software delivery. A full-stack developer, Dan has extensive experience in developing and managing production level systems, database design, system testing architectures, and implementing security protocols. He is also experienced in front-end software development and UX design. As CSO, Dan leads the Data Science / Artificial Intelligence team. He is an expert in statistical modelling and machine learning. He also acts as an external academic supervisor for PhD/MSc students at the University of Warwick.
Dan holds a PhD in Complexity Science from the University of Warwick where he also obtained an MSc in Complexity Science. Previously he read Physics (MPhys) at the University of Oxford. He is also a Certified Prince2 Practitioner with experience of Agile project management.

Chris Crowther

Chief Information Officer

chris_crowther2Chris has over 25 years’ experience in the information assurance and security domain. He is uniquely qualified to understand the evolving threat environment, as well as having an exceptional track record of driving and delivering change in complex organisations. He is a global digital leader with senior experience with the UK military and other Government Departments, US Military and Federal Government, the United Nations, KPMG and Airbus. Amongst a plethora of awards and accolades from the UK and US, Chris’ contribution to the world of Information Risk was recognised in 2016 by his qualification as a CESG Certified Professional Lead Security and Information Risk Advisor. Chris is co-founder and chair for the West of England Cyber Cluster.

Steve Williams

Head of Business Development

steve_williams2Steve is an established leader with over 30 years experience in the capital markets and an extensive background in technology infrastructures. He is the General Manager of DXC Fixnetix. Previously Steve was Global COO for Equities Electronic Markets at Citigroup and General Manager for BNP Securities where he oversaw the firm’s Japanese business and headed all equities trading across the region.

Dr Ben Brown

Chief Medical Officer

ben_brown2Ben is a practising General Practitioner (GP), and expert in developing advanced analytic clinical software. In addition to his medical training, Ben has qualifications in health informatics (PhD), public health (MPH), and leadership (MSc). Ben’s unique career provides him with insights into the challenges of day-to-day clinical practice, combined with the skills to harness data science to address them. This has allowed him to build a strong track record of developing software implemented into NHS clinical practice. His work has been published in over 40 peer reviewed scientific papers, and has won awards from the International Medical Informatics Association, British Computer Society, and Royal College of General Practitioners.

CogX – The festival of all things AI

CogX 2018,cogxlogo the “festival of all things AI”, drew over 6000 attendees and 300 speakers from across technology industries and expertise, ranging from company CEOs to early career research students. Included in this attendees was one of our resident data scientists, as well as many students from the University of Warwick doctoral training centres in Complexity Science and Mathematics of Real World Systems – the very PhD schemes that have given rise to many members of Spectra, as well as being the birthplace of our company itself.

CogX Robot

Companies big and small exhibited content. The expo part of the conference contained a startup village showcasing a variety of new companies in AI and robotics. Larger companies also exhibited content, such as Microsoft Azure and data analysis platforms being used to recommend people their perfect juice (unfortunately including beetroot). Tesla were also showing off their famed car. Google, IBM, and BT each hosted pavilions to demonstrate their latest technologies. Newer dedicated data companies such as Pivigo, Seldon, and Prowler.io were also showing off their various data-focused product platforms.

Seldon was also one of the many companies giving talks throughout the conference. Seldon, Nvidia, and others spoke of the various infrastructures they’ve developed, including data source and optimised software made available to push forward deep learning research in the case of Nvidia. The Financial Times hosted a stage all of their own. Representatives from many companies and academia were involved in many talks and discussions, including AI specific ones such as BenevolentAI and the Open Data Institute, and broader ones such as Deutsche Bank and the Financial Conduct Authority. Government representatives, such as Matt Hancock the Secretary of State for Digital, Culture, Media and Sport, were in attendance. The Turing Institute were also hosting research specific talks, including that of Warwick Complexity alumni Merve Alanyali.

The content of talks were wide ranging, but largely focused on the impact of AI and the broad issues in the future of AI. Ones of particular interest included discussions on the impact AI will have on the economy, mostly on the impact of automation on employment and whether we may be able to achieve a dream future where AI take on the mundane jobs and humans are left with loftier goals. There CogX 2018 Talkwere also discussions on fake news, the many different kinds and motivations that exist, and how organisations such as Full Fact are attempting to combat it. Finally, there was a great deal of talk on the use of AI in seeking a cure for cancer, and in extending life, concluding that complex treatments for complex diseases need machine learning to draw on a wealth of information, as the expertise of a single doctor can only go so far.

If nothing else, this event very effectively showcased the excitement, interest, and range of talent currently in AI. AI, machine learning, and data science are constantly developing tool sets that can give immense utility to any and all businesses. To begin taking advantage yourself, contact Spectra Analytics for a free consultation to see how we can apply our cutting edge data science and business intelligence techniques to help you grow your business.

Modelling fertility in rural South Africa by Rob Eyre, March 2018

Individuals living within the study area of the Agincourt HDSS. Image by A Khosa, courtesy of agincourt.co.za.
Individuals living within the study area of the Agincourt HDSS. Image by A Khosa, courtesy of agincourt.co.za.

One of our data scientists Rob Eyre recently published a paper in Emerging Themes in Epidemiology on modelling fertility in a poor rural region of South Africa using an innovative non-linear approach (the full paper can be found here).

A common issue throughout much of quantitative Public Health research is the application of a range of standardised statistical methods even when such methods are not appropriate. Such standard methods often assume the relationships being modelled to be linear, despite this assumption often being unjustified. One such area where this is the case is in the modelling of how fertility changes over different socio-economic characteristics such as age, education, and social status.

A core aspect of the work we do here at Spectra Analytics involves using more modern, sophisticated, and well-thought-out methods that provide better results to our clients. In line with this, Rob’s research used an innovative combination of a non-linear parametric model of fertility over age, with the use of the highly flexible semi-parametric machine learning method of Gaussian process regression to bring in further variables such as socio-economic status for which no established fertility pattern model exists.

Rob and his research colleagues – Thomas House of the University of Manchester, F. Xavier Gómez-Olivé of the Agincourt research unit in South Africa, and Frances Griffiths of the University of Warwick – successfully applied this method to data from the Agincourt Health and Socio-Demographic Surveillance System (HDSS), run by the Medical Research Council/University of the Witwatersrand Rural Public Health and Health Transitions (Agincourt) Research Unit. This is an annual census performed in a poor rural region of South Africa, collecting information on births, deaths, migration, and many different health aspects. The results of this analysis provided more robust and reliable estimates of the fertility patterns within the Agincourt study area that are free from unjustified assumptions of linearity.

The researchers hope this work will encourage others working in fertility modelling to look beyond standard methodology and be more thoughtful about what methods they use and the assumptions they make when using these methods.

Spectra join British delegation to Cyber Tech Tel Aviv

cybertechWe were very excited to win a place on the British delegation to Cyber Tech Tel Aviv 2018. This was part of the InnovateUK Global Business Accelerator Programme organised by the Enterprise Europe Network and Business West. The aim was to develop closer relationships with the UK and Israel in the field of Cyber Security.

The UK and Israel are two of the leading countries in the field of Cyber Security so this was a fantastic opportunity for the countries to share ideas and learn from one another. Cyber Tech Tel Aviv, the largest Cyber Security event outside the USA, showcased some of the top technology from global providers such as Microsoft and Dell to more than 100 start-ups.

This was a hugely valuable experience for Spectra as we developed some great connections both in Israel and within the UK delegation. This culminated in a fabulous dinner with the British Ambassador which was a great opportunity to network with some of the more influential members of the community.

 

 

Spectra win InnovateUK Grant

Image result for innovateuk We are excited to announce that Spectra has won an InnovateUK grant to develop AI triage for primary care with the University of Manchester. The aim is to reduce GP workload and alleviate the mounting pressures on NHS primary care.

Primary care is the foundation of the NHS, accounting for 90% of all NHS contacts; over 340m consultations per year. Over recent years it has come under significant  strain due to increasing demand, a hiring/retention crisis, and budget constraints. The pressure is exacerbated by avoidable GP consultations where patients should more appropriately self-care or consulting other healthcare professionals such as a nurse or pharmacist. This puts GP practices under unnecessary strain, and endangers patients who cannot see a GP when required. This problem is caused by a lack of effective triage processes – deciding where to send patients – which often rely on non-clinical staff such as receptionists. An estimated 27% of consultations are avoidable, with 6% of patients seen by another professional within the practice and 4% seeing pharmacists or using self-care [1].

The implementation of effective triage processes could dramatically reduce the workload on GP practices. In one approach, ‘Telephone First Triage’, GPs triage patients during a callback phone call when an appointment is requested. However, new research found that whilst face-to-face appointments fell 38% on average, telephone consultations increased 12-fold, increasing average GP  workload by 8% [2]. In addition to its questionable effectiveness, telephone triage is not scalable as it requires constant support from GPs.

An alternative approach is to use Artificial Intelligence (AI) based triage. AI triage is well-suited to primary care due to the large amount of patient data, which allows the algorithm to identify both common more unusual ailments. In this feasibility study, we intend to evaluate the efficacy and impact of AI triage in primary care.

[1] Primary Care Foundation & NHS Alliance, 2015, Making Time in General Practice
[2] BMJ, 2017, Evaluation of telephone first approach to demand management in English general practice

A path to causal inference

The aim of data analytics is to infer the relationships between variables in a system in order to predict and/or control said system. For example, we may wish to understand the relationship between a stock’s return and its volatility in order to profit from changes in these variables or to reduce risk. For instance if we knew that when volatility went down the stock price would go up we could profit by buying the stock when we first saw signs of a drop in stock volatility.

Unfortunately, it is often not possible to directly infer causation because we are (usually) unable to directly perturb a given system; the only way to unambiguously determine causation. Consequently, we often settle for analysing simple correlations which act as a crude proxy for causation. This can be seen in the chart from Spurious Correlations which shows the total revenue generated by arcades in the US versus the number of computer science doctorates. Whilst there is a 98.5% correlation there is no plausible causal mechanism.

from http://www.tylervigen.com/spurious-correlations
from http://www.tylervigen.com/spurious-correlations

There are alternative approaches to inferring causation such as developing a mechanistic model as is done in epidemiology or building an experiment as is done is behavioural economics. The first of these requires a good understanding of the underlying process whilst the latter requires strict controls to prevent external influences and to make it as realistic as possible. Another approach is to consider causation in a statistical sense as is done with Granger causality.

Granger causality is based upon the premise that the process X strictly Granger causes another process Y if future values of Y can be better predicted using the past values of X rather than only the past values of Y. This notion was originally introduced by Wiener (1956) and later formalized in terms of linear auto-regression by Granger (1969). As stated by Barnett et al. (2009), “identifying Granger causality is not identical to identifying a physically instantiated causal interaction in a system; this can only be unambiguously identified by perturbing the system. Instead, it is a causal relation in a statistical sense.” A major problem with Granger causality is that most real problems, such as the return–volume relation, are nonlinear (Hiemstra and Jones, 1994; Chuang et al., 2009). This led researchers such as Baek and Brock (1992) and Hiemstra and Jones (1994) to develop nonlinear extensions to Granger causality. The Hiemstra and Jones (1994) test is now the most commonly used method among practitioners in finance and economics. Unfortunately, Diks and Panchenko (2005) show that this measure may not actually test Granger causality and identify numerous situations in which the test actually fails. An alternative approach is to use a truly nonlinear and nonparametric method such as information theory.

Information theory was originally developed to examine the properties in signal processing, such as data compression, by Shannon (1948). However, it is now widely used in the physical sciences for problems such as statistical inference due to its ability to analyse nonlinear statistical dependencies and higher order moments of the distribution.

Mutual information (MI) is a popular measure in the field of information theory. MI gives the mutual reduction in uncertainty of one variable given another. For example, one can calculate the reduction in uncertainty of the daily return at time, t, by knowing the daily volume at time, t. If there is no reduction in uncertainty, then the daily returns and volumes are statistically independent. Unfortunately, since this measure is symmetric under the exchange of variables, it is only able to determine if two variables are related. However, if one wishes to imply causation one can simply add a time lag to one variable; this assumes that the causal effect cannot back propagate through time. For example, one can find the reduction in uncertainty in the daily return at time, t+1, given the daily volume at time, t, and vice versa. If there is only a reduction in uncertainty in one direction or one is substantially larger, then one variable must be strongly influencing or causing the changes in the other variable.

One can use an asymmetric measure such as Transfer Entropy (TE) (Schreiber, 2000), an information theoretic measure of time-directed information transfer between jointly dependent processes. Barnett et al. (2009) state that TE is not framed in terms of prediction but in terms of resolution of uncertainty. The TE from Y to X is the degree to which Y disambiguates the future of X beyond the degree to which X already disambiguates its own future. This parallels the notion of Granger causality. In fact, Barnett et al. (2009) show that TE is equivalent to Granger causality for Gaussian distributed variables and Hlaváčková-Schindler (2011) extended this to variables distributed as exponential Weinman’s, log-normal’s and certain parametrizations of Generalized Gaussian’s.

This article is based upon a recent paper I published in the Journal of Applied Economics entitled “An information theoretic analysis of stock returns, volatility and trading volumes”; Ong (2015). In this paper I used information theory to show that the observed negative correlation between a stock’s returns and its volatility (known as the Leverage Effect, Black (1976)) is driven by trading volumes; this is supportive of previous research by Avramov et al (2006). This is important for trading and risk management purposes and supports the idea of a behavioural based explanation for the Leverage Effect.

References

Avramov, D., Chordia, T. and Goyal, A. (2006) The impact of trades on daily volatility, Review of Financial Studies, 19, 1241–77

Baek, E. and Brock, W. (1992) A general test for nonlinear Granger causality: bivariate model, Working Paper, Iowa State University and University of Wisconsin at Madison

Barnett, L., Barrett, A. B. and Seth, A. K. (2009) Granger causality and transfer entropy are equivalent for Gaussian variables, Physical Review Letters, 103, 238701

Black, F. (1976) Studies in stock market volatility changes, in Proceedings of the 1976 Meeting of the Business and Economics Statistics Section, American Statistical Association, Alexandria, VA, pp. 177–81

Chuang, -C.-C., Kuan, C.-M. and Lin, H.-Y. (2009) Causality in quantiles and dynamic stock return-volume relations, Journal of Banking & Finance, 33, 1351–60

Diks, C. and Panchenko, V. (2005) A note on the Hiemstra-Jones test for Granger non-causality, Studies in Nonlinear Dynamics and Econometrics, 9, 1558–3708.

Hiemstra, C. and Jones, J. D. (1994) Testing for linear and nonlinear Granger causality in the stock price-volume relation, The Journal of Finance, 49, 1639–64.

Hlaváčková-Schindler, K. (2011) Equivalence of Granger causality and transfer entropy: a generalization, Applied Mathematical Sciences, 5, 3637–48

Ong, M. (2015) An information theoretic analysis of stock returns, volatility and trading volumes, Applied Economics, 47, 36, 3891-3906

Schreiber, T. (2000) Measuring information transfer, Physical Review Letters, 85, 461–4

Shannon, C. E. (1948) A note on the concept of entropy, Bell System Technical Journal, 27, 379–423