By Dr Mike Roberts, Medical Advisor at Camgenium
Medicine is advancing at an extraordinary pace. We can now sequence genomes, deliver precision therapies tailored to individual patients, develop groundbreaking biotechnology and perform surgical procedures that would have seemed impossible just decades ago. Yet despite these remarkable achievements, healthcare systems continue to be challenged by far more basic issues. Medication errors, misread X-rays and test results, and communication failures still cause preventable harm. Human factors such as clinician fatigue, burnout and fragmented care pathways compound these risks. Whilst we can develop cutting-edge interventions, risk mitigation has become an increasingly important focus in healthcare. This is where artificial intelligence (AI) can play a crucial role, leveraging vast datasets to catch the simple errors and identify risks before they harm patients.
In this article, Dr Mike Roberts, Medical Advisor at Camgenium, draws on 15 years of experience in hospital risk management whilst serving in Medical Director roles. During his time in the role, he began using AI tools and quickly recognised their potential to enhance patient safety and clinical outcomes. Their impact led him to work more directly in the field, helping to develop AI-powered analytics tools and increase their adoption across healthcare systems. His experience highlights how AI, when used thoughtfully, can complement rather than replace clinical judgement, a principle central to modern risk management. Dr Roberts advises Camgenium, providing essential clinical input to the company’s work on producing medical device grade algorithms for determining patient risk.
Why Traditional Risk Management Isn’t Enough
Over many years, high-profile cases such as the Bristol Royal Infirmary inquiry into paediatric cardiac deaths and the Mid Staffordshire inquiry have highlighted the need for strong patient safety oversight.
The introduction of clinical governance in the UK has helped formalise processes such as mortality reviews, incident reporting and audits to monitor outcomes. However, these manual systems have inherent limitations:
- They are often subjective and review single cases rather than patterns across many hospitals.
- Clinicians may unintentionally underreport or fail to identify all areas for improvement.
- Manual processes make it difficult to maintain comprehensive, objective datasets.
For example, the National Joint Registry tracks orthopaedic implants and outcomes such as infection rates but still relies on manual entry and can miss key indicators such as pain levels or functional recovery time. AI can automate and broaden this analysis, providing a more complete and objective view of patient outcomes and system performance.
Some clinicians resist AI due to concerns about over-reliance. It’s important to emphasise that AI is a support tool, not a replacement for human judgement. AI can highlight an area that needs review, and clinicians can then investigate further. Used in this way, AI identifies problems earlier and more objectively, saving time, money and stress.
Whilst AI can be transformative, its effectiveness depends on the quality and depth of the data on which it is trained. Poor or biased datasets can unintentionally introduce inequities in care, underscoring the importance of robust data governance and clinical oversight alongside technological innovation. Ensuring transparency and accountability in the way AI systems are developed and deployed will be critical to building trust among clinicians and patients alike.
How AI Is Changing the Risk Landscape
AI is transforming the way in which healthcare systems approach risk reduction. AI imaging tools for CT scans and X-rays can reliably detect abnormalities that radiologists sometimes miss due to fatigue or interruptions. The use of AI improves accuracy and speeds up delivery of diagnosis. In medication management, AI-driven prescribing systems help to prevent dosage errors, missed allergies and drug interactions—all common mistakes that can compromise patient safety.
Beyond these applications, AI can synthesise vast amounts of medical data and research, enabling clinicians to make faster and more accurate decisions. Even highly specialised doctors struggle to keep pace with the rapid expansion of new medical literature, whilst generalists such as those in emergency departments cannot possibly remain current across every specialty. AI bridges these knowledge gaps in real time, offering decision support based on the most up-to-date information available.
Telemedicine and remote AI systems extend expertise to clinicians working in isolated settings, providing round-the-clock decision support without fatigue or distraction. In this way, AI complements human judgement and helps clinicians deliver safer, more reliable care across every setting.
Managing Risk Within a Hospital Setting: A Real-World Example
The UK is experiencing unprecedented patient waiting list times. By implementing AI tools capable of analysing patient waiting list data, the most urgent patient cases can be prioritised. For instance, an outpatient clinic for a rheumatology or respiratory department will likely contain thousands of patients with differing severity of disease and varied comorbidities. If the list were to be reviewed and prioritised by a clinician, it would take hundreds of hours away from their front-line clinical work. AI can deliver the same analysis in minutes.
Waiting list management tools can analyse data for thousands of patients in a way that is unbiased and promotes health equity by ensuring fair prioritisation based on clinical needs and social determinants.
Applying AI for Measurable Impact
The true value of AI emerges when it is seamlessly integrated into clinical workflows. Systems that complement clinicians’ natural decision-making processes, rather than disrupt them, are far more likely to gain adoption and deliver measurable safety improvements. Successful implementation depends not only on the technology itself but also on engaging clinicians early and designing solutions that align with the realities of frontline care.
Camgenium’s AI-powered clinical audit platform demonstrates this potential in practice. It supports hospitals in managing safety retrospectively by assessing performance against expected outcomes based on patient risk profiles. This approach highlights areas where care exceeds or falls short of expectations, allowing for learnings or corrective actions.
Measuring success in this context goes beyond mortality rates. Complications such as infections, readmissions and prolonged hospital stays reveal important insights into both patient recovery and system efficiency. By identifying whether these issues stem from, for example, surgical, anaesthetic or postoperative factors, AI helps ensure that improvement efforts and resources are directed where they will have the greatest impact.
Camgenium’s separate AI surgical risk management platform allows analysis of waiting lists to identify patients at greater risk if treatment is delayed and enables effective prioritisation and resource allocation across the system.
The Future: Resilient, Data-Driven and Individualised Healthcare
AI platforms such as Camgenium’s are continuously evolving and becoming more broadly adopted across hospitals. This allows hospitals to benchmark performance more effectively, provide actionable insights to improve patient care and deliver more individualised care. By developing a more comprehensive understanding of each patient’s individual health needs, AI can predict and prevent complications before they occur.
Collaboration between clinicians, data scientists and technology providers will be key to realising this potential. By combining medical expertise with advanced analytics, healthcare can evolve into a learning system—one that continuously improves from every patient interaction. The real promise of AI lies in making healthcare proactive, helping clinicians prevent harm, extend healthy life years and restore trust in a safer, smarter system.
About Author
Dr Mike Roberts, Camgenium’s Medical Advisor, leads the delivery of Camgenium’s AI-powered analytics tools for healthcare settings globally. Dr Roberts initially trained in Emergency Medicine in the UK and spent his early career working as a doctor in the NHS. He later held positions as Clinical Director of an Emergency Department in the NHS, and was medical director at an acute Trust.
In 2010, Dr Roberts moved to New Zealand for a clinical post in Emergency Medicine before being appointed Chief Medical Officer in January 2012. Dr Roberts has experienced firsthand the positive impact AI tools can make to a hospital. He is now an advocate for the use of AI tools to support clinical governance and improve patient outcomes worldwide.
