Submitted by Pooja Pandey on Thu, 22/09/2022 - 13:51
Founded by Dr. Will Handley, Prof. Mike Hobson and Professor Anthony Lasenby from Cambridge AstroPhysics, PolyChord is a next generation data technology.
A spin-off from the University of Cambridge, the core PolyChord technology is uniquely able to cope with complex, messy, noisy and multivariate data as it was invented to analyse the microwave radiation present after the Big Bang, effectively tackling a 10 billion year dataset. This capability makes Polychord technology powerful and agile enough for all kinds of different industry sectors. Further, the PolyChord company works in close collaboration with its commercial partners to bring together specialised data science expertise with deep domain knowledge. As the company has evolved, the collaborations have helped define a truly effective approach.
The technology has the ability to fully explore a complex, challenging data landscape. Developed at the astrophysics research group of the Cavendish Laboratory, this cutting-edge tool for solving big complex data problems can be applied in a variety of areas. Sectors where PolyChord has managed to make a serious impact range from predictive maintenance in light and mainstream rail networks, bridge and road infrastructure to more long-range work in food processing and storage and advanced drug discovery.
Predictive maintenance in Light and Mainstream rail network One of the key applications is in analysing and making predictive maintenance judgements in rail infrastructure. This has been made possible by using PolyChord’s ability to gain more useful information from the complex data landscape created by a set of simple low-cost sensors, to extract signal from noise. Working in a group of established organisations within the sector, the end product is PolyTrack, which is revolutionary in using 4G and 5G to provide real-time decision-making tools. These genuinely actionable insights make predictive maintenance a reality for the tram companies rather than just an idea. In the future, PolyChord hopes to apply the approach into Light Rail and on to mainstream Network Rail. Recent progress with LNER is extremely encouraging. |
Enabling next-generation drug discovery PolyChord is making this possible by supporting a leading biotech company which operates a first-in-class drug discovery programme. The protein folding initiative from the Astrophysics community at the Cavendish – Polyfold – is making this possible; where the way in which proteins fold and unfold is being recognised as a data problem in a high dimensional space. The Polyfold initiative uses PolyChord’s ability to fully explore complex data landscapes to better understand the way a protein folds. This has the ability to support the development of new drug discovery programmes to address a range of unmet diseases that advanced drug discovery companies are currently researching and developing. |
Offering more advanced model selection metrics than the W.H.O in food processing and pathogen prediction PolyChord is able to account for more variables and to model atypical as well as normal growth in food processing and pathogen prediction. This ability supports a more advanced model selection metrics than the W.H.O standard. Using lab data and real-world scenarios, the focus for PolyChord is now shifting to saving energy in large scale milk and food processing by applying learning from FoodScan. It also has applications in companies growing algae on a large scale for commercial purposes. |
Recent commercial focusing on sensor investigation As PolyChord has grown with £1.3m government grant funding, the commercial focus is to target specific sectors in the short term in order to maximise the impact with a team of 11 highly-qualified people. The focus now is on “making sense of sensors”: for instance, the PolyTrack product is currently being extended by the organisation that collaborated in its development, West Midlands Metro. Development is ongoing with mainstream rail operator LNER as well. PolyStructure takes a similar approach with sensor investigation of bridges run by National Highways, with some early interest from Transport for London. There is also serious interest in the MIDAS product developed with the MOD which has the unique abilities to reduce the amount spent on expensive hardware in telecom network deployment and maximise coverage at the same time. |
PolyChord technology has managed to make an impact by solving industry problems in the real world. Other data science technologies which attempt to do optimisation or manage trade-offs fail where there are complex constraints. This is where PolyChord offers telecom companies significant advantage.
PolyChord continues to solve problems and gain forward momentum, particularly where there are sensors delivering data which can be turned into actionable intelligence for companies and organisations.
Reference - https://polychord.io
Image(s) Credit - Pixabay