SPHERE away day
11th May saw the fourth SPHERE team away day. Held on the magnificent engineering marvel that is the SS Great Britain, 30 + SPHERE researchers gathered to work on the project's next steps. We had a number of sessions on the Clinical uses of the SPHERE technology, hearing about, and discussing Parkinsons, Dementia and Muscular Skeletal Conditions.
We were also set a challenge around 'Ground Truthing' - a challenge of how to simply check if the SPHERE sensors are detecting what we think they are. There was a great response to the challenge, combing the skills of researchers and others from a wide range of backgrounds. Ultimately the blue team were champions, with a simple but effective system. Well done to everyone!
EurValve is leveraging a subset of SPHERE to characterise quality of life at home for heart valve replacement patients before and after surgery. The SPHERE in a Box Kit contains a wrist worn wearable with four gateways and a router. In addition to collecting the patient's movements, the system collects information that can be used to determine the patient's location. Using this data, the project will focus on inferring the patient's activity and level. Trials will begin over the summer with 60 homes targeted.
SPHERE technology update
The goal of SPHERE has always been to test our sensor platform in real-life environments. After months of technology development and a period of testing in the SPHERE house and pilot homes, the technology is now being prepared for wider deployment in up to 100 homes near the University of Bristol.
There will be up to 30 sensors (depending on the size of the home) installed in each of the test homes. We now have the 'deployment' versions of these sensors ready.
These include the SPHERE wearable, and the charging pad. The wearable will come with a choice of colours and straps, and will identify individual members of the household, where they are in the house and some movement information.
The environmental sensor and receiver will collect information about Temperature, light, humidity etc, and receivers will be distributed throughout the house to collect information from the other sensors.
There will be several other sensors in addition to these collecting information on a wide range of activity, including appliance usage, energy usage and quality of movement. We also saw the outputs of some of the sensors, including temperature and humidity.
This is an exciting stage for SPHERE as we start to collect real-life data and develop the sensors ability to recognise different activities. We'll keep you posted on developments!
The First Deployment
On the 15th December 2016 SPHERE had a momentous day as the first deployment in a participants house took place. This was a very exciting day for the team to see the last few years hard work being installed into a participants home. The team look forward to being able to install into further homes in the new year and to start analysing the data collected.
Designing the SPHERE Genie
Kevin Marshall, Sam Mitchell Finnigan, John Vines, & Rob Comber
The Open Lab team at Newcastle University share their experience of designing the User Interface for the SPHERE system.
Beginning in January 2016, a team of researchers from Open Lab, Newcastle University joined the SPHERE team working on the development of the SPHERE Genie, the interface used to control the SPHERE system. Through this blog post, we want to share a little bit about what we have been doing and introduce the SPHERE Genie.
The SPHERE Genie will be the interface through which people taking part in the SPHERE deployments can control the system, pause data recording, and contact the SPHERE team. There will also be research challenges sent out through the Genie, where people can take part in extra activities around the deployments. As the Genie will sit in people’s homes and be used regularly, we want to make sure it works well for everyone and that it is inviting and rewarding to use.
Our work has involved a number of different activities, including running a workshop with members of the research team where they wrote a description of the SPHERE home as if it were a Bed and Breakfast. We wanted to use this to understand how the research team might communicate SPHERE to people who may be interested in living with it. We also had the opportunity to meet many of the Friends of SPHERE in the Knowle West Media Centre and used this as an opportunity to talk about what the walls of our homes might say about us, if they could talk. This gave us a chance to speak to the Friends of SPHERE and, potentially, those who may have the system in their home about any privacy concerns this may bring.
For our final activity, we brought together 11 Friends of SPHERE for a workshop. In this workshop, the Friends of SPHERE completed a data profile of their home, sharing aspects of their life and home with us. We then had a discussion around the type of information people shared and any concerns they may have had about sharing this information. We also provided different visualisations of data taken from the SPHERE house and tried to work together to make sense of what they communicated.
Through these thought-provoking conversations with the Friends of SPHERE, we designed and developed the first prototype for SPHERE Genie, an easy-to-use interface designed for a tablet computer that enables users of the SPHERE system to control it.
It has been a very enjoyable experience working with members of the SPHERE team in Bristol on such an exciting and large-scale research project. The opportunities to meet members of the public and the Friends of SPHERE have been most memorable, with their excitement, interest, and critical input proving highly motivating and engaging for us. We look forward to seeing what happens in the future with SPHERE as it is deployed in homes across Bristol.
My PhD journey so far...
Michal Kozlowski shares his experience working in the SPHERE project
My first contact with SPHERE came when I entered into the Dress/Sense competition, back in 2014.
As part of the team we have developed an idea of one of the school kids which was part of our team – YoBot was an artificial companion used for detection and treatment of acute depression in young adults.
This led to an offer of PhD within the SPHERE’s environmental and on-body sensing group. The summer after my graduation was spent with SPHERE working on a driver for newly-released TI SensorTag 2, specifically the microphone. Having spent so many days working with fellow undergraduates, proceeding to an office full of post-doctoral researchers was daunting at first.
After a while however, I felt the familiar comfort I felt before. Concentrated on a task, I saw no world beyond my monitor and that little PCB. In September 2015, I officially begun my PhD studies. Feeling acclimatised from the previous 3 months, I experienced little change from the usual grind – I begun reading a lot more and slowly constructing the conspectus of my PhD thesis. Passively, I started to notice the way academia is structured and learn the mechanisms responsible for its perpetuation.
Taking an active position in the University’s life, life of a postgrad is much different than undergrad – responsibilities shift, duties pile up, motivation is up to you – and the latter is key.
One of the most brilliant thing about SPHERE is the people – as a PhD in any other research group, I feel, I would not be as involved with the project as I am with SPHERE – collaborating with researchers on a variety of different problems. Personally, I am working on an indoor positioning localisation and mapping platform – for this task we are developing a mapping and localisation system for residential homes. I am part of the team responsible for the indoor localisation of people; for now in the SPHERE house, hopefully used in all of the houses for the 100 home rollout.
SPHERE at CHI
SPHERE at the ACM Conference on Human Factors in Computing Systems (CHI),
San Jose, USA, 5-12 May 2016
Dr Alison Burrows and Dr Em Tonkin represented SPHERE at this year's CHI, one of the top ranked conferences in computer science. CHI brings together several thousand researchers and designers from universities, corporations and start-ups from across the world to discuss the future of how people interact with technology. This year, CHI received over 3500 submissions and accepted over 1000 presentations and events.
With so much going on over the four days, each attendee tailors their experience of CHI according to their research interests and information needs. For instance, Em particularly enjoyed visualisation-centric talks, technology demonstrations, and various alt.CHI presentations such as the use of fiction to support the design process or just how much Robert Pirsig's 'Zen and the Art of Motorcycle Maintenance' can do for us when teaching the principles of user experience and human-computer interaction design.
CHInosaur, the CHI 2016 mascot
Em presented a paper at the pre-conference workshop 'Involving the crowd in future museum experience design'. In her paper, ‘Using the crowd to update cultural heritage catalogues’, she proposed methods to automatically identify and replace obscure and difficult terminology in order to make complex or antiquated texts easier to read. On the surface this topic is not obviously related to SPHERE, but in fact language plays a crucial role in people’s interaction with technology. This workshop was a good opportunity to look for synergies with others working in complementary fields to explore future avenues for research. How can organisations take information that is dry, dense and full of incomprehensible jargon and 'translate' it to the ever-changing language used by the potential audience? Can technology help us to learn more about real-world language use, and to apply that information to our own fields?
Alison presented a paper at the main conference, which also discussed the importance of language in the development of technology. Specifically, the paper entitled ‘Shared language and the design of home healthcare technologies’ (DOI: 10.1145/2858036.2858496) offers a participatory design approach to developing terminology that is meaningful and accessible to all stakeholders involved in an interdisciplinary project like SPHERE. This work was developed with Dr Rachael Gooberman-Hill and Dr David Coyle, within the remit of SPHERE’s User-Centred Design activity. The audience was particularly interested in how this process might be applied in different contexts, including within organisations and with multicultural groups.
Alison presenting her paper in a session on Participating in Well-Being and Family
In addition, Alison presented a poster on behalf of a former MSc student, Xu Luo, whose research focused on applications for Near Field Communication (NFC) technology in the context of home healthcare. This work was accepted to the CHI Late Breaking Work track as a paper entitled ‘SPLASH: Smart-Phone Logging App for Sustaining Hydration enabled by NFC’ (DOI: 10.1145/2851581.2892513), co-authored by Xu and his supervisors, Pete Woznowski (WP6), Alison Burrows (A3), Mo Haghighi (WP3) and Professor Ian Craddock. The project resulted in a smart-phone app that enabled people to monitor their liquid intake throughout the day, through a combination of smart-phone NFC technology and NFC-tagged cups. Over the two days it was exhibited, the poster was well received and sparked several conversations about future applications for this technology.
Ready, Steady, Data collect!
Emma Pritchard reflects on her recent participation in an study in the SPHERE house, collecting data on how participants cook.
Recently, with some trepidation, I accepted a request by a University of Bristol student, Sam and Adeline a SPHERE researcher, to participate in a study in the SPHERE house. They were collecting data exploring the activity of cooking and were looking for participants. After they explained how the study was going to run and what infomation they were going to collect, I accepted the offer.
Sam and Adeline kitted me out with a head-mounted camera and then instructed me to perform some aerobics in the kitchen to allow the wall-mounted camera to recognise me. We were ready! Sam and Adeline were set up in the dining room watching what the camera was catching. We were steady! I am no Ainsley Harriott but when Sam tipped the bag of ingredients out I immediately felt like I was on ready Steady Cook and the dish of the day was tomato pasta. I was cooking! Initially I was conscious that Sam and Adeline would see my poor garlic crushing skills but after five minutes my inner Jamie Oliver came out and I was away.
Having cleaned up I triumphantly exited the kitchen with my bowl of pasta. I was curious to see what they had captured, Sam showed me that when in the room the wall-mounted camera detected me and put a box around me which tracked where I was. During one of my visits to the fridge I gained a twin, my body being in front of the fridge had caused the camera to think it was another person and when I walked away a second box stayed around the fridge. Sam and Adeline are working hard to help the camera distinguish between an actual person and any other household object, which should lead to useful developments in the SPHERE project.
Participation in studies is anonymous. Emma was happy to contribute with her experience in this blog.
Come Dine with SPHERE
Some 42 people arrived for a meal and to find out more about the project – only to discover the venue had been converted into a mock house - where they could visit different rooms to learn about the sensors. Following an introductory talk by SPHERE Director Ian Craddock - visitors divided into groups and were directed round ‘the house’.
In the ‘living room’, complete with couches and TV - Sion from Bristol University gave a demonstration of the video work, looking at people’s movement within the house. Participants could see how the cameras would only record a pinpoint skeleton – so individuals would not be recognisable from the images. They could also discuss how they felt about cameras installed in homes and find out in which rooms they would be set up. People also wanted to know if there was the option to turn them off if they wished.
Seated around a large table - Rob and Fontas from Bristol University along with Sue from KWMC showed examples of sensors that would be installed in people’s homes. These included the Passive Infra-Red (PIR) sensor, a sensor that records light, noise levels and temperature as well as humidity and air pressure, which would be fitted in every room, and it would look as small as a smoke alarm – all agreed they would not mind this in their homes and would probably forget about it.
Other examples shown included appliance and water sensors and the wearable sensor to record people’s activity levels. Questions ranged from: “Who is going to see the data? and “Is the wearable
waterproof?” to “Could this system be used to find out what pets are doing?”
The ‘kitchen’ was manned by Rob and Kevin from Newcastle University who are developing the interface for the project. They were asking what information participants would like to see on a tablet or computers. Some suggested it would be good to know how the different rooms in their house were being used and the interaction between inhabitants. Others asked: "Could the interface help motivate people to move if they were in rehab?" , "could it flag up things that needed looking at in the house?", or show if health was deteriorating?" and "would it show energy monitoring and who was leaving the lights on in the house?"
At the end of the event, visitors carried on conversations while eating Moroccan stew and salads. There was a lot of positivity about the project and lots of simultaneous group discussions.
Matthew said the best thing was “developing an understanding of how healthcare and technology will be integrated in the near future", and added "but I’d like more of an understanding of how the data is stored, interpreted and used by clinicians".
Sam commented: “I liked the openness of researchers to ideas, criticism etc. I would have liked more time to talk!”
And Jackie added: “The whole thing was very interesting and gives hope for the NHS of the future.”
There were some excellent discussions taking place, and we'd love to continue with them. More events will be advertised soon. In the meantime, please do sign up below for our newsletters, or continue the conversation on our facebook or twitter pages.
Activity and Behaviour Recognition
Collaboration with Rostock University
SPHERE started a new international collaboration with the Mobile Multimedia Information Systems Group at the University of Rostock, for two projects on activity and behaviour recognition. A Bristol master student, Sam Whitehouse, has been invited to spend 3 weeks at Rostock, where he is now designing a model of cooking activities. Dr Adeline Paiement also visited Rostock for a couple of days, to help Sam settling in and to discuss more collaboration possibilities. Dr Kristina Yordanova will in turn join us at Bristol for a month in a few days to work on an activity models for the full house. Stay connected!
Citizen Sensing Initiative – Visit to Chicago
Dr Xenofon (Fontas) Fafoutis represented SPHERE in University of Bristol delegation to Chicago. The visit was sponsored by UKTI (UK Trade & Investment) and the FCO (Foreign & Commonwealth Office), and was led by Dr Theo Tryfonas, Senior Lecturer in Systems Engineering with the department of Civil Engineering of the University of Bristol. The other members of the delegation were: Dr Philippa Bayley, Manager of the Cabot Institute, Ms Rebecca di Corpo, Knowledge Exchange Project Manager with RED, and researchers Dr Aksel Ersoy and Dr Georgios Z. Papadopoulos.
Photo by Philippa Bayley
On the first day, the UoB delegation visited Lane Tech High School - who are piloting sensing technology in their school environment. Lane Tech is the largest high school in Chicago and one of the largest in the country with 4500 students. The school is collaborating with the School of the Art Institute of Chicago and the Array of Things project at Argonne Labs on their own pilot project called the 'Lane of Things'. The idea is for students to build and programme their own wireless sensors to be deployed around the school and to address issues that concern and interest them.
The next day, the delegation was hosted at the Argonne Labs, who are developing the Array of Things (AoT) city sensor technology jointly with the University of Chicago, School of the Art Institute of Chicago and the City of Chicago. AoT is funded by a $3.1M fund to develop open source and open hardware outdoor sensing technologies and deploy 500 in the city of Chicago. The prototype sensing nodes incorporate various sensing technologies, including temperature sensors, humidity sensors, light levels sensors, infrared light sensors, pressure sensors, vibration sensors, air quality sensors, and particulate matter sensors. They also incorporate a 5 MP HD camera. Pilot deployments outside of Chicago are organised including Bristol in collaboration with the University of Bristol and Bristol is Open. Discussions have also been initiated on the potential collaboration with SPHERE by measuring the air quality outside the SPHERE houses.
Photo by Xenofon (Fontas) Fafoutis
Over the next days, the delegation visited the School of the Art Institute of Chicago, a collaborator in the Array of Things project; and Arup, a British construction firm that specialises in Smart Buildings. The delegation also attended an event on Smart Water infrastructures organised by BuiltWorlds.
On the last day, the UoB delegation visited Matter, a Chicago-based incubator dedicated to healthcare technologies. It was a unique opportunity to discuss the potential challenges and benefits of modern technologies to healthcare systems, and identify potential partnership models between Matter and the University of Bristol.
This blog first appeared here in http://blog.yhat.com/posts/model-based-machine-learning.html
Friends of SPHERE
Putting on an event like Friends of SPHERE for the first time we were really dipping our toes in the water to gage public interest in the project. Bristol University chose the perfect venue – At Bristol is very central and attractive to families and the promise of a Planetarium Show afterwards was an added bonus. But the fifty or so people who turned up on a sunny Saturday afternoon to find out more about SPHERE were extremely interested in the project itself – and how the technology in homes could benefit health in the future.
Asked what attracted them – replies were: “Well, it sounds very interesting.” And pensioner Linda said: “My friend and I try and do different things at the weekend and this sounded fascinating.”
From the moment they were greeted at the door – there was a chance for children to attend a popular robot-programming workshop if they wished - run by Knowle West Media Centre. The laughter generating from the room proved that science could be fun – and these could be potential researchers of the future. Adults were given refreshments and the chance to chat to others before Director Ian Craddock gave an introduction to the project – with statistics about our ageing population in the future and the predicted strain on the NHS.
SPHERE is divided into work packages and attendees were able to visit different tables for demonstrations of the technology. They ranged from the wearable devices - with a pancake flipping trial to measure energy use - to movement sensors – where you saw a pinpoint skeleton of yourself.
Nina who had travelled all the way from Knowle West for the afternoon said: “I came because I’ve been part of a project before with sensors in the home and I’m interested in this one.”
The afternoon discussions on different tables with researchers gave people the chance to ask questions in small groups and find out more about the project – but also to come up with useful feedback. For example, one table discussed extra things that people would like SPHERE to do - such as detecting damp homes or helping predict falls.
Researcher, Alison Burrows, had created some postcards for people to write on – such as “If your house could talk to you what would you like it to say?”
There was no shortage of ideas in the discussion – with people being concerned about air pollution and also using the system for alerts to remember keys or locking their door on leaving; and keeping elderly parents safe or alerts when away and a problem occurred such as water leakage. “But I think your house should be your friend and not too critical,” Emma said when the conversation turned to reminders about healthy eating and exercise.
At the close of the afternoon Bob, who is already part of the SPHERE public advisory group, remarked: "I thought it was excellent – it was a great opportunity to discuss with both the people who are working on SPHERE and with new people – so other issues were raised".
Bob also commented: "In our group (wearables) we brought up the fact there was no emergency system on the device if anyone wanted to get someone urgently and some modifications were needed. I thought being divided into groups was good so that certain issues could be highlighted. I think SPHERE is a very good template for people to follow in other regions and countries too."
The finale of the Planetarium Show - with a spectacular scene of the winter sky - was the icing on the cake. It was also very timely with live footage and photos of British astronaut Tim Peake on the International Space Station.
A mum who had brought her two children remarked: “They’ve loved it – I was worried they might be bored but we’ll definitely come back to another SPHERE event and I’ve learnt a lot about the project too.”
If you want to join Friends of SPHERE and receive more info about future events, add your details in our
'Get Involved' page.
What is Model-Based Machine Learning?
by Tom Diethe
If you haven’t had your head in the sand, you’ll realize that Machine Learning is no longer a niche research area, it is now a mainstream technology being applied by engineers right across the spectrum.
"I think it's the dawn of an exciting new era of info and computer science … It's a new world in which the ability to understand the world and people and draw conclusions will be really quite remarkable… It's a fundamentally different way of doing computer science." -Steve Ballmer
However, in practice there are pitfalls and dangers, especially when employing machine learning for the first time. It’s easy to become swamped by the sheer number of methods, there is a whole new vocabulary to learn, and it is often difficult to choose an algorithm for a given problem. Often, especially when data is unstructured (which is increasingly the case) it’s hard to work out which off-the-shelf method fits the problem, and instead one has to resort to coercing the data to fit the problem. It’s also often not clear how do to deal with noisy, or missing or corrupted data.
This post is about a different viewpoint called “model-based machine learning” , which tackles these difficulties, can solve problems in a wide range of application domains, and covers most existing machine learning algorithms as special cases. It will also allow us to deal with uncertainty that we encounter in real-world applications in a principled manner.
Since the invention of the Perceptron algorithm , huge numbers of algorithms have been created to solve various specialist tasks. A common approach to solving problems involves trying a few different algorithms, often in practice guided by familiarity, or due to the presence of a particular toolbox in the language being used, rather than it being the most appropriate for the problem. Each algorithm will also have parameters that often require careful tuning, and these often don’t map to intuitive concepts. As a result, practitioners often resort to using the defaults set by the authors of a given toolbox, or exhaustive searches of the parameter space.
If you can’t find an algorithm that fits your problem, you are left with two options: modify your problem until it fits some standard framework, or invent a new algorithm. Whilst the latter may get you a NIPS or ICML paper, this is not a viable option for most. Model-based machine learning instead offers tailored solutions.
The central idea underpinning the model-based approach to directly encode any problem-specific assumptions, with any available prior knowledge, in the form of a (mathematical) model. These include the number and types of variables in the problem domain, and the factors that determine their interaction. A model-specific algorithm is then (automatically) generated. This approach can be used to do any standard machine learning task, such as classification, regression, or clustering, whilst improving understanding and control over how these tasks are accomplished.
If we take the example of a network of different kinds of sensors in a real world environment, these will introduce different sources of uncertainty. We might have sensors that are simply not working, or that are giving incorrect readings. More generally, a given sensor will at any given time have a particular signal to noise ratio, and the types of noise that are corrupting the signal might also vary.
As a result we need a principled framework for quantifying and computing with uncertainty. In the model-based approach we build a model of how the data was generated, which can directly incorporate the noise models for each of the sensors. It is easy to see that probabilistic Bayesian graphical models are a natural fit to the model-based framework.
“The subjectivist states his judgements, whereas the objectivist sweeps them under the carpet by calling assumptions knowledge, and he basks in the glorious objectivity of science.” -I.J. Good
The key insight that is often overlooked is that any off-the-shelf algorithm has underlying assumptions of its own (intentionally or not), although often these are ill-defined. The result is that the algorithm behaves like a “black box”, meaning empirical comparisons are necessary, for example by nested cross-validation. This is laborious and inefficient, and there are many pitfalls to this approach . If no algorithm gives adequately good results the way forward is even more unclear.
From a model-based viewpoint, to make predictions using the model we need to plug the observed data into model, and compute the probabilities of the possible values a variable can take after the relevant evidence is taken into account - a process known as “inference”.
Exact and Approximate Inference
By separating model and inference in this manner, the same method of inference can be applied to a wide variety of models, or alternatively different inference methods can be used for the same model. Perhaps the simplest approaches to inference are simulation based approaches such as Markov chain Monte Carlo (MCMC), which can shown to be correct in the long run, but slow to converge. For any “realistic” model with more than a trivial quantity of data, we may hit the limits of computational tractability. Deterministic approximate inference methods, such as Expectation Propagation and Variational Bayes, make it possible to learn models by trading off computation time for accuracy.
"An approximate answer to the right problem is worth a good deal more than an exact answer to an approximate problem." -John Tukey
Because we have clearly laid out the assumptions when creating models, it becomes easier to compare them both qualitatively and quantitatively. This is especially important if there are parts of the modelling process that are difficult to pin down.
“Statisticians, like artists, have the bad habit of falling in love with their models.” -George Box
If we can compute the probability of a model given the the data (not assuming any particular model parameters), this is the “Model Evidence”](https://en.wikipedia.org/wiki/Marginal_likelihood) (or marginal likelihood). This quantity can then be used to compute the so-called “Bayes factor”, which can be thought of as a Bayesian alternative to classical hypothesis testing [4, 5], by taking the ratio of the evidence for each model. For models where the evidence is too costly to evaluate numerically, approximate Bayesian computation can be used instead.
Case Study: The SPHERE Project
Many countries are experiencing the effects of an ageing population, which coupled with a rise in chronic health conditions is encouraging a shift towards the managing health related issues in the home. The SPHERE (a Sensor Platform for HEalthcare in a Residential Environment) project  has designed a multimodal sensor system and analytics platform for this purpose. Naturally, the SPHERE setting presents many sources of uncertainty. Firstly, we are dealing with multiple sensor modalities (environmental, body-worn, video), each of which will have different noise profiles and failure modes. Secondly, annotated or labelled data is expensive and intrusive to acquire, and the resulting labels are potentially noisy and inaccurate. Lastly, patterns of human behaviour are subject to many factors that may or may not be attributed to the particular health context of a given individual. In this project we are making use of model-based methods throughout.
Tools for Model-Based Machine Learning
Because of the separation of the model from the method of inference, it also becomes possible (if by no means trivial) to create software that is able to take the model, as specified using some form of modelling language or API, and then automatically generate inference routines (possibly even by automatically generating source code!) to solve a wide variety of models. This allows a new breed of engineer - effectively a “modeller”, who does not need to know about the specifics about the inference method being used. Some examples of software packages that seek to achieve this are:
Infer.NET. A software framework developed at Microsoft Research Cambridge for running Bayesian inference in graphical models. It can also be used for probabilistic programming.
BUGS. A Bayesian modelling framework using MCMC methods.
Church. A probabilistic programming language designed for expressive description of generative models .
Stan. A probabilistic programming language implementing full Bayesian statistical inference with MCMC sampling, approximate Bayesian inference with Variational inference and penalized maximum likelihood estimation.
GPy. Gaussian processes framework in python, from the Sheffield machine learning group.
PyMC. A python module that implements Bayesian statistical models and fitting algorithms, including MCMC.
See the 'early access' model-based machine learning book at http://www.mbmlbook.com.
Machine learning is being successfully applied to a large number of real-world problems as you read this post. In some cases, all that is needed is the “best guess answer”, and there also happens to be an off-the-shelf tool for the job. However, there are also a large class of other scenarios: where either no such tool exists, where quantifying the uncertainty is of great importance or where you want to be able to introspect on the answers given by the system. Model-based machine learning provides a compelling path to tackling all of these scenarios.
. Winn, J., Bishop, C.M., Diethe, T. (2015). Model-Based Machine Learning. Microsoft Research Cambridge. http://www.mbmlbook.com.
. Rosenblatt, F. (1957). The perceptron, a perceiving and recognizing automaton. Report 85-460-1. Cornell Aeronautical Laboratory.
. Krstajic, D., Buturovic, L. J., Leahy, D. E., & Thomas, S. (2014). Cross-validation pitfalls when selecting and assessing regression and classification models. Journal of cheminformatics, 6(1), 1-15.
. Goodman, S. N. (1999). Toward evidence-based medical statistics. 1: The P value fallacy. Annals of internal medicine, 130(12), 995-1004.
. Goodman, S. N. (1999). Toward evidence-based medical statistics. 2: The Bayes factor. Annals of internal medicine, 130(12), 1005-1013.
. Zhu, N., Diethe, T., Camplani, M., Tao, L., Burrows, A., Twomey, N., Kaleshi, D., Mirmehdi, M., Flach, P. & Craddock, I. (2015). Bridging e-Health and the Internet of Things: The SPHERE Project. Intelligent Systems, IEEE, 30(4), 39-46. http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=7156004
Skeleton-Free Body Pose Estimation From Depth Images for Movement Analysis
SPHERE at ICCV 2015, Santiago, Chile from the 11th to 18th of December 2015
Adeline Paiement and Ben Crabbe presented their paper "Skeleton-Free Body Pose Estimation From Depth Images for Movement Analysis" at the ChaLearn Looking at People workshop that was held on the 12th of December at the International Conference on Computer Vision (ICCV) 2015. In her presentation, Adeline describe their last work on human body pose estimation, and received positive feedback.ICCV is a major event in the Computer Vision community, which brings together students, academics and industry researcher. ICCV 2015 was held in Santiago, Chile from the 11th to 18th of December.
SPHERE at the 2nd IEEE World Forum on the Internet of Things, MILAN
Images X. Fafoutis
SPHERE was extremely well represented in the technical sessions of the 2nd IEEE World Forum on the Internet of Things, Milan, Italy, in December with 7 papers accepted. In addition to this wide representation, Director of SPHERE, Professor Ian Craddock gave a key note speech.
Professor Craddock talked about how the Internet of Things will transform health. He explained how the rising cost of healthcare is being driven by long-term health conditions in people of all ages. Professor Craddock went on to discuss how tackling this issue will require new models of care, underpinned by new "Internet of Things" (IoT) technologies that allows the clinical environment not merely to be extended to the community, but also to the home, the workplace and everywhere that the "patient" goes. Professor Craddock went on to talk about the implications for the engineers and scientists that are developing these new technologies.
In addition to Professor Craddock’s keynote speech, SPHERE was represented by Dr Robert Piechocki, Dr Bo Tan, Dr Xenofon (Fontas) Fafoutis, Mr Paul Worgan, and Dr Mo Haghighi, Former SPHERE Research Assistant.
Dr Xenofon (Fontas) Fafoutis explains, “The emerging era of the Internet of Things will revolutionise everyday life and the future of healthcare. It was a unique opportunity to share the research outcomes of SPHERE with a diverse academic and industrial audience”.
The SPHERE researchers contributed to the conference proceedings with 7 papers that cover a wide range of IoT challenges for future healthcare including reliable energy-constrained wireless communications, inductive power transfer for wearables and health trackers, classifier-based residential floor map construction, and passive non-intrusive human motion sensing.
The list of the presented papers includes:
E. Tsimbalo, X. Fafoutis and R. Piechocki. Fix It, Don't Bin It! - CRC Error Correction in Bluetooth Low Energy. In Proceedings of the 2nd IEEE World Forum on Internet of Thing (WF-IoT), Milan, Italy, December, 2015.
X. Fafoutis, E. Mellios, N. Twomey, T. Diethe, G. Hilton and R. Piechocki. An RSSI-based Wall Prediction Model for Residential Floor Map Construction. In Proceedings of the 2nd IEEE World Forum on Internet of Thing (WF-IoT), Milan, Italy, December, 2015.
E. Tsimbalo, X. Fafoutis, E. Mellios, M. Haghighi, B. Tan, G. Hilton, R. Piechocki and I. Craddock. Mitigating Packet Loss in Bluetooth Low Energy. In Proceedings of the 2nd IEEE World Forum on Internet of Thing (WF-IoT), Milan, Italy, December, 2015.
B. Tan, A. Burrows, R. Piechocki, I. Craddock, Q. Chen, K. Woodbridge and K. Chetty. Wi-Fi Based Passive Human Motion Sensing for In-Home Healthcare Applications. In Proceedings of the 2nd IEEE World Forum on Internet of Thing (WF-IoT), Milan, Italy, December, 2015.
M. Haghighi, P. Woznowski, N. Zhu, E. Tsimbalo, S. Hannuna, A. Burrows, B. Tan, L. Tao and R Piechocki. Agent-based Decentralised Data-acquisition and Time-synchronisation in Critical Healthcare Applications. In Proceedings of the 2nd IEEE World Forum on Internet of Thing (WF-IoT), Milan, Italy, December, 2015.
M. Haghighi, K. Maraslis, T. Tryfonas, G. Oikonomou, A. Burrows, P. Woznowski and R. Piechocki. Game Theoretic Approach Towards Optimal Multi-tasking and Data-distribution in IoT. In Proceedings of the 2nd IEEE World Forum on Internet of Thing (WF-IoT), Milan, Italy, December, 2015.
P. Worgan, O. Pappas, T. Omirou and M. Collett. Flexible On-Body Coils for Inductive Power Transfer to IoT Garments and Wearables. In Proceedings of the 2nd IEEE World Forum on Internet of Thing (WF-IoT), Milan, Italy, December, 2015.
The HealthCom 2015
The 17th International Conference on E-health Networking, Application and Services(HealthCom), sponsored by the IEEE, took place in Boston, USA, on 14-17 October 2015. The purpose of HealthCom is to bring together interested parties from around the world working in the healthcare field to exchange ideas, discuss innovative and emerging solutions, and develop collaborations.
SPHERE WP2 presented a paper entitled "A Comparative Home Activity Monitoring Study using Visual and Inertial Sensors” in the presentation session of the main conference. The presentation included a brief introduction to SPHERE project and the comparative study with a dataset for the fully automated, sample-accurate recognition of common home actions in the living roomenvironment using commercial-grade, inexpensive inertial and visual sensors. The presentation went very well and received valuable and constructive feedback.
The issue of energy
Do you wish you didn’t have to charge your various devices quite so often? SPHERE will work best if it’s as passive as possible – which is why SPHERE is working towards discarding the charger for good.
SPHERE researcher Jerry Luo, based in the University of Southampton, has been exploring a novel approach to harvesting energy – using his shoes…
How it works:
Using the SPHERE Wearable developed in SPHERE project, for every single footstep, the transmitter gains sufficient energy from the energy harvesting insole and is able to send 3 to 4 packages of 32-byte wireless data to its receiver.
The transmitted data can be used for gait analysis, personal identification, indoor localization and in-room localization.
Soon your shoes could be working for you…. Watch this space.
SPHERE wins at REFRAME
SPHERE Phd student Hao Song wins data challenge at European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases
Our congratulations go to Hao Song, SPHERE Phd student, who recently came first in a competition at the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases.
Conference attendees were challenged to explore bicycle hire data for Valencia and calculate availability for the up-coming months. After some hard work and complex calculations, Hao’s work beat the competition and he was crowned winner of challenge. Well done Hao!
Friends of SPHERE
The SPHERE technology is developing fast. With this progress comes the need to increase the number of studies and the amount of testing that we need to undertake. Engaging with the public has always been very important to SPHERE - there are many questions around user acceptance, ethics and design, for example, that SPHERE will have to thoroughly engage with the public on to succeed.
The next big step for SPHERE is to test the technology in real homes. We've been busy attending events across the city, including three 'Make Sunday Special' events to talk to the public about SPHERE, and to let them know about what we'll be doing next.
If you'd like to find out more about how you could potentially get involved in working with SPHERE to give feedback on the technology, please visit our 'Friends of SPHERE' page.
Vast Quantities of Data!
August 2015 - Ruth White
The SPHERE project will enable us to collect a vast quantity of data from a combination of environmental, video and wearable sensors in a home environment. What will all of this data mean though? Researchers may be happy to trawl through lots of ones and zeros and graphs of accelerometer data but even we get bored of that eventually and it certainly holds little meaning for the end user of the system! The person living in the house and the doctors looking after them want to be able to quickly use the data to make real decisions. There must be a more efficient way … this is where my research comes in!
I am looking at how all of this data can be interpreted to provide useful information. Ok, so it’s not just me dealing with all of this data alone, first I rely on my colleagues to do the initial stages of the processing. They are able to take the raw data and determine what activities someone is doing in the house, for example, sitting watching the tv or washing the dishes. I then try and find patterns in the list of activities that someone is performing to discover what their routine behaviours are. For example, the routine of dinner time for one person might involve them chopping up vegetables and meat, cooking on the stove, eating at the table, washing up the dishes, drying up the dishes and tidying up; whereas for someone else it might involve heating a carton of soup in the microwave, eating on the sofa whilst watching the tv and tidying up.
Although each person will have a different normal behaviour, using these discovered routines it is possible to then look for patterns in how one person’s routine changes. So, over a long period of time, perhaps a year, if someone’s routine changes dramatically then this will be noticed. The user of the system and the doctor can then review this information to determine if it is significant and indicates a deterioration in health.
I am also investigating using the same techniques to find patterns in people’s eating habits. Although the SPHERE project does not yet automatically collect detailed information about what someone is eating, this is something that could be included in the future. Therefore it’s important to start thinking about how this data can be used. I am currently using food diary data collected as part of a National survey to test my ideas.
SPHERE house experiments begin!
July has been an important month for SPHERE. After weeks and months of testing and installation, ethics approvals and decorating, experiments are starting in the SPHERE house.
Initial tests will establish 'ground truth' allowing the system to recognise normal activities, such as making a cup of tea.
As these experiments develop and bring in data, the expriments will move on to recording day to day life for a series of volunteers who will stay in the house. It's an exciting time for SPHERE and we're looking forward to collecting a lot of data!
SPHERE at Red Maids' School
SPHERE and Toshiba researchers recently visited Redmaids School to run a competition to design wearable technology with a health benefit.
The event was launched by SPHERE director, Professor Ian Craddock. The pupils split into 5 groups to design and build a wearable device with a health benefit. They came up with some fantastic ideas, ranging from a mood sensing device for those effected by autism, to a uv sensor to indicate when it was time to add sun block.
SPHERE researcher Ni Zhu, said
"'What the girls achieved suprised me not only in terms of their well-trained skills in Arduino and programming, but also the capabilities of discovering daily living needs from a scientific angle as well as enabling technologies to realise their ideas.'
Mrs Bramley-Dymond was also delighted with the day
"It was really fantastic day, and the support that Russ, Rachel, Lily and Ni offered us was perfect. I don't think we would have done it without them.
The girls were thrilled and overwhelmed by the generous prizes and I can't say thank you enough.
The girls had a valuable journey today and came up with some interesting ideas."
Prof Ian Craddock
As Director of the SPHERE project I was very happy to be asked to talk about The Future of Health Ethics and Privacy at the Cheltenham Science festival.
I drew uponsome concepts around Internet of Things (IoT) privacy and future health uses cases I’d recently also presented at events in Sweden and Portugal.
It was also good to work with my friend and colleague Madeleine Murtagh, Professor of Social Studies of Health Science, who speaks with such eloquence and understanding on the huge social science aspects of these technologies - I certainly learnt a lot from her talk.
One of the most remarkable things about "e-health" or "IoT for Health" - or whatever you want to call it - is that it generates an equal level of interest and insight in both specialist and non-specialist audiences. This is a godsend for an engineer who worked much of his career in the highly mathematical black art of electromagnetics.
As we saw at my talk, a non-technical audience has a very clear concept of privacy. They have a nuanced understanding of the trade-offs between locking data up in vaults and making it available for research and health. Thanks to the rapid uptake of fitness bands – there’s a ready appreciation of how technology will inevitably transform healthcare for us all.
The Cheltenham Science Festival is undoubtedly one of the best places in the UK to engage with audiences and I very much enjoyed my visit - especially the excellent debate that we had after the talk. People had lots of questions - from which medical conditions could benefit from the SPHERE technology to who owns the data.
It was great to have the opportunity to hear people’s views and think about how we can take them on in our research.
Advisory Group Update
SPHERE needs to talk. The technology is coming along very well. But, as we all know, just because something works, it doesn't mean people will take it on. Why are we not all driving around in Sinclair C5s? And how many of us listened to Minidiscs? Just because it looks great on paper, and works, doesn't mean it will succeed. SPHERE has to talk to people. If we don't understand what our potential users want, and what they'll be prepared to allow in terms of, perhaps, sharing data, or allowing new technology to be installed in their houses, we won't succeed.
This is why our advisory groups play such an important role in the project. This week, the group met with Fontas and Lindsay to talk about wearables.
Should they be waterproof?
Would you wear more than one at a time?
Will they interfere with pacemakers?
Will people with arthritis be able to do them up?
These were among the many questions and thoughts raised by the advisory group exploring the development of SPHEREs wearable.
As one of SPHERE's researchers put it "When concentrating on solving technical problems, it's easy to forget the use to which the technology is to be put - and meeting [the advisory group] reminds us of that"
Six Months at SPHERE
So, what has SPHERE achieved in the last 6 months? Well, it’s been an incredibly busy time, with the research really taking off. Amongst other things…
The first SPHERE Bluetooth LE wearable sensor has been produced. SPHERE’s own ultra low power design is uniquely optimised for long-term data collection and will therefore underpin research both in and outside of SPHERE for years to come.
The “Dress/Sense” wearable technology competition, launched by the President of Singapore (left) on his state visit to the UK, produced eight high-quality entries. George Ferguson, Mayor of Bristol, awarded team ‘Yo’ the winning prize for their Cognitive Behaviour Therapy device.
SPHERE conducted a study involving five men and women with Parkinson’s Disease, all in their 70’s and all at high risk of falling. This identified a number of activities on which to focus future work.
The energy harvesting team have demonstrated how to run wearable devices without batteries by transferring energy safely and efficiently through the air. In scenarios where patients cannot be expected to regularly recharge batteries, this will be a game-changer for personal health devices.
The SPHERE living lab (house) has successfully been instrumented with sensing technology and the various data sensor sources have been integrated. Initial studies to calibrate the system are now underway.
To find out more see our newsletter.
SPHERE away day
How would you write the manual for the first people to have SPHERE installed in their homes? How would you write the installation manual? How do we present SPHERE at conferences? How quickly can SPHERE researchers run 100 meters in a hamster wheel?
These were all important questions that were asked at the SPHERE away day 2015. SPHERE is a large (and growing) team of researchers and collaborators. Spread across several universities and sites, it is of great importance for the team to meet up, take a step back, and assess our progress and our next challenges.
Researchers presented their work from over the last year, giving everyone a great insight into just how much SPHERE has achieved and how exciting the future looks.
There was, of course, also time to develop as a team, including the hamster wheel....
Dress/Sense film launch
SPHERE rolled out the red carpet on Wednesday 25th March to welcome back the participants of SPHERE's Dress/Sense competition. Film-maker Azita Ghassemi has documented the wearable technology competition from day one.
The film traces the journey the particpants took from the launch of the competition to the winning team's prize being awarded. It was fantatic to see that journey, and to hear how the participants have progressed, with participants moving on to PhD's, work experience and even a presentation in the US.
The opportunities raised by bringing together such a broad range of people was summed up perfectly by one of the participants "The fact different specialties can offer equally to produce an end product. Each different professional group spoke a different language that meant there was a massive increase in communication overall.”
Communicating SPHERE - Algorithm or Calculation?
SPHERE's technology is developing fast. We are going to be working more and more closely with the public to test in real-life situations over the coming months. So what is the best way to communicate what we are doing? On 12th March a large group of SPHERE researchers joined together to explore how we can clearly explain what we are doing, without bombarding the public with jargon.
Algorithm or Calculation?
Sensors or Equipment?
Camera or movement sensor?
We developed a glossary using a set of 'personas' of people who are likely to benefit from SPHERE. It is was the start of work the will undoubtedly be ongoing throughout the development of SPHERE. Please do contact us if you have any thoughts.
Do the SPHERE sensors work in the Garden?
In early March we were lucky enough to have a visit to the SPHERE test home by members of PEP-R. PEP-R stands for Patient Experience Partner in Research. The group meet regularly to input their experience into the research undertaken on musculoskeletal conditions at Southmead Hospital.
The PEP-R group had a tour of the SPHERE test house and the opportunity to talk about what they saw with Pete, a post-doctoral researcher on the project.
It was great to get the group into the house to give use their perspective on first encountering the SPHERE technology.
We were asked many questions, and also given useful thoughts.
"If someone has a fall in the garden that isn’t picked up by the wearable, could the other sensors detect an anomaly and trigger an alarm?"
This kind of engagement is vital for the SPHERE project to help create a technology that not only technically works, but also reflects the needs and worries of potential users.
Thank you very much to PEP-R and Amanda Burston for providing us with such useful feedback.
Dress/Sense competition - what's been happening...
Medics, Engineers, Designers, School pupils.... The Dress/Sense competition was unique in its approach of bringing together an incredible array of talents to create innovative and amazing wearable technology.
"Each different professional group spoke a different language that meant there was a massive increase in communication overall. " Participant
As well as producing some truly innovative designs, the project gave the rare opportunity for people from different backgrounds to share their expertise for a wider benefit.
“I love when you do it and it’s all swell that you’ve actually programmed it, then you actually do it and you’re like YEAH! I really like working with a group of people who can actually get it done, cause I sometimes have ideas, but there is no way that I can do it, especially because I am quite you and I can’t start a business or anything, so it’s really nice to have an opportunity where something that we’ve worked on and ideas come through it actually be produced and maybe finalised.” Participant
“It’s really good to see the students integrating with other students from the University and people from industry it is something they don’t get to do within school. “ Teacher
“I wanted to do dress sense because it’s the first chance I’ve had to do a proper inter-disciplinary project at the University. I spend a lot of time with computer scientists, electronics engineers and you can end up being quite focused, quite blinkered, so it’s interesting to hear other people’s engineering problems. So to have medics come up to me and say I want to know before hand when someone’s going to have a fit due to epilepsy, or over time how someone’s posture changes, so if someone’s depressed they are going to be into themselves and maybe looking at their feet a lot, how do you detect that over time, that’s a very interesting problem.” Participant
The legacy of the project has lived on beyond the end of the competition. One of the teams, Jurojin, continues to develop their device. Several participants are interested in taking up MSC project, PhD's or other opportunities within SPHERE.
The School children are inspired to continue and develop their skills in engineering and coding. The project will build and develop on the foundations of multi-disciplinary working to see how far this work can develop.
And we'll leave you with the booklet produced by team 3 Yo! Click on the image below to see more.
SPHERE's Work Package 1 - Home Enviroment Sensing - Update
SPHERE environmental sensors, as the most fundamental elements of our smart home technology, are designed as a a general-purpose platform with various sensing capabilities, whereby their functionality within the home environment can be customised as needed. Environmental sensors constantly capture parameters and events of interest from their surrounding areas and inform the SPHERE home gateway of any changes taking place in the ambient environment. They are spread across different parts of the house, continuously monitoring power consumption of different appliances; water flow; humidity; temperature; luminosity; air quality, noise level; door movements; and finally human physical movements in different parts of the house.The cooperation of those sensors with other SPHERE sensing technologies, enables monitoring human behavior to significant details and register the pattern of their interactions with their residential environment in micro-levels.
Variation of temperature and humidity in different parts of the house, taking shower with cold or hot water, switching on the kettle, and walking in different parts of the house are typical examples of such activities that may be recorded with their accurate timing in their given space. In Work package 1, we have designed our sensor platform to be adaptable to various home infrastructures with regards to the size and layout of the property, and their available resources. Therefore, the network of our environmental sensors smartly scales in and out according to their operating environment.
SPHERE environmental sensors are equipped with a range of technological advances with respect to their communication and sensing capabilities. The backbone of such sensors is also very flexible in terms of adding more functionalities and customising their roles in the network. Such flexible features are due to a powerful software layer, in which the operating system takes advantage of the latest research in middleware technologies.
The second version of SPHERE environmental sensors, which is currently under development, is based on the most popular and energy-efficient ARM processor architecture.
The technical advantage behind such sensors enabled them to function in an ultra low-power mode, whereby their required energy is harvested from their surroundings. Their wireless communication capability including ZigBee and Bluetooth Low Energy (BLE), also allows them to secure data communication, minimize power consumption and cover large areas with no necessary modification to the existing infrastructure.
SPHERE's very own Aardman Animation!
With Aardman Animation's 'Shaun the Sheep' currently getting rave reviews it seems like a good time to remind everyone about SPHERE's very own Aardman Animation!
Way back in 2013, SPHERE was lucky enought to work with Aardman Animations to create this incredible short film talking through the benefits and questions raised by the SPHERE project. The audio was recorded by local people, and the animation was created right here in Bristol. Why not take a few minutes, sit back, and enjoy!
SPHERE's PhD students - what are they working on?
Hi, my name is Paul Worgan and I am undertaking a human-computer interaction PhD with the SPHERE project. My PhD is focussed on providing power to wearable health sensing systems in a flexible and convenient manner. As well as being a SPHERE researcher, I am undertaking my doctorate in the Centre for Doctoral Training in Communications (http://www.bristol.ac.uk/cdt-communications/) and I am based within the Bristol Interaction and Graphics laboratory (http://big.cs.bris.ac.uk/).
In my year and a half spent as a PhD on the SPHERE project I have been exposed to a fantastic range of disciplines including human-computer interaction, human factors engineering, energy management and healthcare. I have also learnt more skills than I thought possible; including a familiarity with accelerometers, gyroscopes and Arduino prototyping, 3D printing, laser cutting, Bluetooth communication, wireless power transfer systems and experimental design.
Initially I developed a wearable electronic knee goniometer for use in exercise rehabilitation programs. During the project I got to work alongside a very talented physiotherapist, who assisted in bringing the project to fruition.
My current research focusses on providing power to wearable health sensing systems in a novel and flexible way. Asking patients to wear sensing systems can be enough of a burden without having to invest effort into recharging or changing batteries. My research focusses on providing ambient wearable health system recharging and considers the human factors and safety challenges associated with this.
Outside of research I have had the opportunity to assist as a postgraduate teaching assistant in a masters level computer science module and an undergraduate electronics course.
My experience of life as a SPHERE PhD student has been very positive. A particular highlight of the PhD for me has been assisting in the SPHERE wearable competition Dress/Sense (http://www.dress-sense.org.uk/) and seeing the fantastic range of creative ideas generated from a diverse and talented group of people.
Bristol taps away to be first smart city
SPHERE recently featured in an article in the Sunday Times. The quotes SPHERE director Professor Ian Craddock describing the SPHERE house as a 'Big Sister House' and predicts "the public will embrace the technology, just as many people track their daily fitness regines on mobile phone apps". 2014 drew a great deal of interest in the SPHERE project, and we look forward eagerly to developments in 2015, as the SPHERE house runs it's first tests!
SPHERE presented two papers at the recent PowerMEMS conference in the Awaji Yumebutai International Conference Centre, Japan.
The two papers presented were Experimental Study of RF Energy Transfer in Indoor Environment (S-E Adami, P P Proynov, B H Stark, G S Hilton and I J Craddock) and Low-power Methods of Power Sensing and Frequency Detection for Wideband Vibration Energy Harvesting (P Proynov, B H Stark and N McNeill)
December 5th 2014
The Dress/Sense competition has been an incredible Journey. From the launch by the president of Singapore in October, to the prize-giving on December the 5th, we continued to be surprised and delighted with what we've seen.
The Watershed played host to the final chapter of this tremendous competition, with the Mayor of Bristol, George Ferguson awarding Team 3 - 'Yo' a cheque to recognise the huge amount of work they put in to their innovative and unique winning entry.
The evening was very well attended with everyone greatly impressed with the winning team's skill and imagination in creating the 'Yo' device - a device designed to help those undergoing Cognitive Behaviour Therapy.
The competition has produced so many devices of such a high standard that this won't be the end. We are inviting all teams in to the SPHERE offices to explore how we can continue to work together.
The evening reflected the cheerful and exciting nature of the competition and was thoroughly enjoyable.
A huge thank you goes out to our funders, EPSRC, with support from: Toshiba, Kinneir Dufort, the West of England Academic Science Network (AHSN) and Designability. Thank you all.
Chief Medical Officer for England visits the SPHERE house
26 November 2014
Prof Ian Craddock, Director of SPHERE, Dame Sally Davies, Prof Jeremy Tavare, Director of the Elizabeth Blackwell Institute for Health Research and Patty Holley, Project Manager
The team was delighted to receive the visit of Professor Dame Sally Davies, Chief Medical Officer for England. She was invited to have a look at the SPHERE house, a highly instrumented space designed to test the sensor platform designed by SPHERE to monitor the wellbeing of people at home.
Demos using ambient sensors and off-the-shelf cameras showed the possibilities of using technologies to analyse data related to changes in behaviour and how this may be linked to health. Dame Sally Davies was accompanied by Dr Tony Soteriou, Research Infrastructure and Growth Senior Manager. SPHERE plans to use the house not only to test the accuracy of the sensors but also to record the experiences of people living in a monitored environment.
Her visit has been organised by the Elizabeth Blackwell Institute, named after one of Bristol's unsung pioneers in improving public health. It has recently announced a £3 million funding boost to continue its work to bring together leading researchers from a range of different fields to help solve the most pressing health problems of the 21st century.
Professor Jeremy Tavaré, Director of the Elizabeth Blackwell Institute, said: "It's a huge honour to welcome Dame Sally to Bristol and to have this opportunity to show her the work going on, both at the University and in the wider city, to tackle various pressing health problems faced by the nation as a whole”.
SPHERE was invited to participate in the UK-India Workshop on Affordable Assisted Living Technologies (ALT) for the Elderly
Bangalore, Wednesday 19th November 2014
India, as the rest of the world, is facing a change in population structure with increased longevity and therefore growing needs in healthcare especially in the ageing population.
New technologies play a role in addressing some of the needs of these changes in society. SPHERE was invited to participate in a workshop organised by the Indo-British Geriatric Association, the theme of the workshop was using affordable technologies in the Indian context.
Key to the development of all these new technologies and devices is the participation of the user in the generation of ideas and the design of the devices. At the same time local and national governments should be involved in developing pathways so the users (patients, carers, health professionals) can access new technologies.
Finally, India has a high number of trained professionals that would benefit from interchanges with researchers in the UK. Co-funded PhD studentships will be an excellent way to promote collaboration between these two nations. SPHERE will be very interested in hosting these exchanges. The University of Bristol offers a wide range of expertise in areas that would benefit the development of projects to promote healthy ageing.
SPHERE at the ‘Long Term Care Revolution’ organised by Innovate UK
Dr Dritan Kaleshi presented SPHERE to Innovate UK’s Long Term Care Revolution event on 25th November, which was organised to launch of a series of consultations by Innovate UK ALIP team to announce a call in April 2015 for innovative solutions in long-term care.
The presentation was well received, with lots of useful feedback for the SPHERE team in attendance. There was a strong focus on approaches and robots at the event, and calls for ‘consumer-driven solutions’ – a great event and plenty to think about!
Dress/Sense Final Session
Saturday 22nd November saw the final session of SPHERE’s Dress/Sense competition. The teams now have till Friday to complete their wearable device. The day say the teams working hard to make the most of their precious time together.
There was an amazing diversity of activity – from soldering, to knitting, CAD design to filmmaking, all competitors seemed to be deeply involved. So, who will win? Watch this Space……
Bristol: The UK's most technically smart urban environment?
SPHERE project is based in Bristol, with partners and collaborators across the country and the world. Why Bristol? A recent article in the Guardian Newspaper Networked and super fast: welcome to Bristol, the UK’s smartest city highlights how SPHERE is perfectly placed to work in a University from a city that is riding on the crest of a wave of technological innovation.SPHERE's Public advisory Group
The SPHERE project is well on the way to developing the technology to successfully monitor people's health in their home. The technology does, however, raise a number of questions. To ensure that the technology will work for people as well as in its own right, SPHERE is engaging with the public to get their opinions, concerns and ideas. One way in which SPHERE is doing this, is through the public advisory group. There are two SPHERE advisory groups; a public group and a social care professional group. Each group has around 8 members, who come from a wide range of backgrounds. They are invaluable in helping the project reflect on those questions that we need to consider for the project to work in the real world. The most recent meeting took place in November in the SPHERE research home. Thoughts included: Can the sensors be linked to an automatic entry system for social care workers? How will a user interface consider the needs of older people, people with disabilities and learning difficulties? Is there potential for the system to help monitor children (an advanced 'baby listening' device?
Dress/Sense Session Two
The second day of SPHERE's Dress Sense competition was all about the ideas. The day got off to a flying start with the participants having a chance to meet others and discuss their ideas over spaghetti and jelly beans.
After that was out the way, the group had two really useful and insightful talks on dementia and diabetes by Dr Jude Hancock and Professor Julian Hamilton-Shield, which greatly inspired the group. Ideas generation followed, with over 20 innovate ideas being put forward.
The group voted on these - those that they thought should go forward and those that were great for another time but not this competition. Teams were then formed, and the planning began. Now we only have to wait another couple of weeks to see the final products. We can't wait!
SPHERE meets the President of Singapore
SPHERE has recently celebrated its first birthday, and it has been doing it in style. The President of the Republic of Singapore visited SPHERE on October 23rd. The President was in Bristol as part of the Singaporean State visit to the UK. He visited the University of Bristol to see the technology being developed by SPHERE, and witnessed the signing of a memorandum of understanding between NTU Singapore and Bristol University. Both the University of Bristol and NTU are interested in how to help aging with the effects of an ageing population, and there are many opportunities to explore.
The highlight of the visit was the launch of SPHERE’s Dress/Sense competition – a competition to design and build wearable technology with a health benefit. Pupils form Bristol Free School and Redmaids' School were enthusiastically demonstrating the Arduino kits that they had been learning to programme. The President took a great interest in their work, and how it could, in the future, help the SPHERE project.
Greg Clark MP, Minister for Universities, Science and Cities gave a speech on the importance of exploring the role of technology in the future of healthcare. SPHERE researchers had created interactive exhibits for the guests to explore, and the scope and inventiveness of the project is really starting to show itself.
SPHERE has recently celebrated its first birthday, and it has been doing it in style.
The first year of SPHERE has produced some truly innovative developments in technology, has brought together sometimes quite unexpected partners, and has left us all feeling really excited about what the future holds!
SPHERE DRESS/SENSE competition gets off to flying start!
After the launch of the SPHERE Dress/Sense competition on Thursday 23rd October 2014 by The President of the Republic of Singapore, Saturday saw the first day of the competition. The day got off to a flying start, with 60 competitors turning up full of ideas and energy. The atmosphere was alive with ideas, and there was a lot of excitement in the group, which spanned from 12 year old school children, medical and engineering students to NHS professionals.
The competition challenges teams of scientists, medics, engineers, designers and textile artists to design a piece of wearable technology with a health benefit. The event takes place over three days and the teams will be competing for a £5,000 prize, with the winning team receiving their prize from the Mayor of Bristol at a prize giving on 5th December.
After the first day, the competitors were buzzing with ideas and had created their first piece of wearable tech.
We can’t wait to see what happens next….Body Area Networks Conference
SPHERE’s Evangelos Mellios from Work Package 3 (on-body sensing and wireless communications) recently presented their paper on “Off-body channel measurements at 2.4 GHz and 868 MHz in an indoor environment” at the 9th International Conference on Body Area Networks (BodyNets 2014, London, 29th September – 1st October).
The work was a collaboration with the Telecommunications Research Lab of Toshiba Research Europe Ltd (Bristol, UK).
SPHERE's Data Fusion and Data Mining team presenting at conferences
SPHERE's work package 5, Data Fusion and Data Mining, will be presenting papers at two different conferences this September. Niall Twomey and Peter Flach will be presenting 'Context Modulation of Sensor Data Applied to Activity Recognition in Smart Homes' at this year's European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD) conference in Nancy, France. Tom Diethe will be presenting at the Large-scale Online Learning and Decision Making Workshop in Windsor, UK.
The British Machine Vision Conference 2014 Nottingham
SPHERE's Adeline Paiement and Lili Tao will be at the British Machine Vision Association (BMVA) conference 1st -5th September at Nottingham University. Adeline and Lili will be presenting their method for movement quality assessment. 15th International Symposium on Electrets August 2014
The 15th International Symposium on Electrets (ISE) was held John Hopkins University in Baltimore, Maryland, USA in August. It was attended by SPHERE's Dr Zhenhua Luo. He presented a study entitled "Energy Harvesting Study on Single and Multilayer Ferroelectret Foam under Compressive Force". SPHERE at the Festival of Nature June 2014
SPHERE attended the Festival of Nature on Bristol's Harbourside in June 2014. Over 100,000 people visit the festival over the three days it runs, and many of them talked to us about SPHERE. The responses we got were interesting and varied. On the school days, children were overwhelmingly positive, thought, such as in the comments above there was some useful feedback. Comments included: "It would be okay if we have control of the sensor" "I think it's a good idea for old people" "Sensors would be useful becuase you might be ill but you don't know about it"
What do you think?