ARTIFICIAL INTELLIGENCE


originally motivated his research: “When I volunteered at Addenbrooke’s Hospital during the COVID-19 pandemic, I was struck by the lack of communication and information sharing across the NHS. Facing the overwhelming challenge of managing a surge in patients requiring oxygen, I saw firsthand the gaps in connecting engineers and healthcare professionals from different hospitals. This experience motivated me to pursue my PhD research, exploring ways to enhance knowledge sharing in this fragmented system. Inspired by my colleague Jan, who was developing AI chatbots, I discovered the transformative potential of generative AI. Together, we founded innex.ai, aiming to bridge these information gaps and better support the healthcare community.”


Innex.ai solution for healthcare engineering The innex.ai system uses data sources from a variety of engineering repositories, including NHS England and the Department of Health Guidance [which includes Health Technical Memorandum (HTMs)], legislation, individual healthcare trusts policies, and industry journals e.g. IFHE Digest and the Health Estate Journal (HEJ). This is the initial data set up and innex.ai plan to expand their data set to include international standards and journals. Innex.ai uses Large Language Models (LLM) and generative AI to learn the patterns and structure of their input ‘training data’, which then allows new data to be created that has similar characteristics. Carl carried out research in the estates


and facilities departments within the NHS in England in 2023. He asked the question: “How much time do you spend every


Table 1.


Example 1: What is the benchmark value for fossil thermal energy consumption per m2


in general office spaces? (HTM).


The answer from innex.ai, tookseconds to arrive, was as follows:


The benchmark value for fossil thermal which energy consumption per m2 office spaces is 120 kWh/m2


in general .


The system also shows the HTM 07-02 A Page: 32, Publication Date: 25/03/2015 Reference source:


Example 2: What per cent in pathogen reduction does treating a zone with UV achieve, according to a recent research study?


The AI system returned the following answer:


According to a recent research study, treating a zone


with UV achieves a reduction of pathogenic microorganisms by a minimum of 3 log steps, which is equivalent to at least 99.9 per cent reduction of the pathogenic population 1.


Reference source: 64 NETB 2023_01B.pdf Page: 10


Health Technical Memorandum 07-02: EnCO2de 2015 – making energy work in healthcare


traditional keyword search, the word ‘bank’ might retrieve results about rivers and financial institutions, our semantic search understands the context of your query, ensuring that if you’re asking about ‘bank regulations’, it fetches information related to financial laws and not riverbanks.


“When a user asks a question, our 1960 telephone exchange nuclear bunker.


week searching for technical information?” His research covered the following NHS England staff: technical and engineering experts, project and facilities managers, and the heads and directors of estates and facilities. His result was:


l “more than 11 hours are spent weekly searching for information”


The system has a natural language interface which allows the user to ask questions and also to indicate, if they want, where they want the result to come from.


Some examples of how innex.ai can


solve an example healthcare estates query are shown in Table 1: The first step is to enter the following


question into the innex.ai platform. In the first case, the questioner has specified that they want the result to be from the NHS England Health Technical Memorandum Carl explains how the system is


answering the questions: “Our system uses hybrid search approach with both filters and semantic search rather than traditional keyword search. While in


system first rephrases it to enhance clarity and context. For example, if a user asks about ‘HVAC efficiency standards’, our system might rephrase this query to ‘current regulatory compliance requirements for HVAC systems’ energy efficiency’. This refined query is then converted into a vector and matched against our database to retrieve the 3-5 most relevant document sections. This method ensures our system delivers contextually appropriate responses and avoids generating inaccurate information, maintaining reliability by citing all sources.”


Functionalities of the AI system What innex.ai can do:


Retrieve relevant information This refers to the system’s ability to access, search, and extract relevant data or knowledge based on the user’s input. This can include facts, figures, explanations, or summaries from internal knowledge databases. When you ask a question or request


information, the AI parses your query, identifies the key components, and matches them against its knowledge base to find the most relevant answers.


Generate coherent and contextually appropriate responses The AI’s ability to produce responses that are not only accurate but also logically structured and appropriate for the context of the conversation. This involves understanding the


nuances of human language, including grammar, tone, and intent, and generating responses that align with the ongoing conversation. Coherence ensures that the response makes sense as part of the dialogue, while contextual appropriateness means the answer fits the specific situation or question asked, avoiding irrelevant or confusing information.


Understand natural, conversational language The system’s capacity to comprehend and process language as humans naturally speak it, including idioms, slang, and various dialects. Unlike rigid, keyword-based systems,


advanced AI models are trained on large datasets of conversational text, allowing them to understand complex language structures, infer meaning from context,


IFHE DIGEST 2025


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