ARTIFICIAL INTELLIGENCE DR LOUISE WEBB, DIRECTOR, DRLC LTD; CARL-MAGNUS VON BEHR, MD,
INNEX.AI, UK
AI to manage compliant engineering systems
AI is becoming a well-established tool in the field of computer systems and solutions. If harnessed correctly, the ability of a system to ‘learn’ over time can be a positive asset in engineering. Authorising engineers DRLC Ltd are collaborating with AI start- up
Innex.ai – based out of the University of Cambridge – to create a new system for healthcare engineers. Dr Louise Webb, director at DRLC Ltd – with contributor Carl- Magnus von Behr, MD,
innex.ai – tell us more.
“What we want is a machine that can learn from experience.” Alan Turing, February 1947, in Lecture to the London Mathematical Society.
The development of AI has mirrored the rise in the amount of data stored electronically. When IT was in its infancy, a megabyte – that is, 1000 kilobytes or a million bytes – seemed like a large amount of data. In 2020, it was estimated that there were 64 zettabytes of data in storage, equating to as much information as there are grains of sand on all the beaches in the world. In mathematical terms, this is 1,000,000,000,000,000 megabytes. The rate of growth of data is increasing exponentially, and when the data above was collated, it was estimated that 90 per cent of this data was created in the last two years. At the source of all AI systems is high
quality data. Recent advancements in natural language processing and text mining have significantly increased the availability and quality of textual data. Data mining processes structured data,
for example, found in healthcare engineering guides and applies modelling functions to produce results. Text mining is a similar function; however, it starts with unstructured data which must be organised and structured before it undergoes data modelling and analytics. The exciting process which can be
added to data and text mining and which has caused an explosion in the use of AI is machine learning (ML). This process mimics human learning in its approach, which uses historical data to achieve a desired outcome. A simplistic description of the AI learning phase could be described as follows:
Dr Louise Webb
Louise is a director of DRLC Ltd, with her partner of over 40 years David Livingstone. DRLC work with healthcare
providers supplying authorising engineers in many areas including pressure systems, water safety and medical
gases. Louise started her career as a medical student at Edinburgh University. She then went on to gain a MSc in
Computer Studies. As a senior project manager at BT she project managed the first Google servers to be installed in the UK
Carl-Magnus Von Behr
Carl-Magnus von Behr is an industrial engineer specialising in interdisciplinary problem-solving in healthcare facilities management. Witnessing the silo working and challenges faced by NHS EFM departments during the COVID-19
pandemic inspired Carl to dedicate four years to PhD research focused on improving knowledge sharing among NHS Trusts. Drawing on the insights from his research, he co-founded
innex.ai, with the goal of driving innovation and operational efficiency across the NHS.
IFHE DIGEST 2025
Learning Phase l Provide a sample of historical data – for example, a description of Planned Preventative Maintenance (PPM) for Healthcare Air Handling Units (AHUs).
l Process the data using a software process (algorithm) to establish or learn the key patterns and trends.
l Output of the above is a model or set of rules.
Once the learning phase has been carried out, the next step is to use the model that has been created to look at new data and reach some intelligent solutions.
Prediction Phase l Load the set of rules from the learning phase.
l Use the new model with new data – for example, an updated version of PPMs.
l Use the AI system to predict the likelihood of an outcome – for example, the answer to a question about frequency of filter checking in a critical AHU.
How AI systems can benefit healthcare engineering To date, AI systems have been seized upon by profit-making organisations, for example, online retailers, to enable them to push advertisements to their customers based on data that the customer has either overtly or unintentionally made available to the retailer. Examples of overt data capture are using the clients past history of purchasing from that organisation. More covert data capture which also feeds AI engines includes clients’ postings on social media platforms and cookie settings from websites. Using AI for healthcare engineering
moves away from this ‘for profit’ model into the realms of AI for positive social change.
Carl-Magnus von Behr explains what 63
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92 |
Page 93 |
Page 94 |
Page 95 |
Page 96
