With AI taking the world by storm, the healthcare industry will not remain unaffected. Recent advances in AI could bring about substantial improvements in the industry, increasing accessibility and effectiveness.
The global healthcare industry is presently facing challenges AI could aid in overcoming such as increased demand due to ageing populations, chronic disease burden, and a persisting lack of access to quality healthcare. In fact, the world is not on track to achieve the UN’s Sustainable Development Goal of universal health coverage by 2030. This is not solely due to COVID-19 pandemic, as health service coverage improvements have stagnated since 2015 worldwide, with an estimated 4.5 billion people currently without access to essential healthcare services.
AI Applications in Healthcare
Presently, AI is being used in multiple areas in the healthcare industry. Firstly, machine learning (ML) shows great promise in diagnostics and disease classification, particularly in regards to cancer. Multiple studies show ML use leads to increased accuracy in breast cancer diagnosis, with an AI-assisted diagnosis process showed higher sensitivity than radiologists, detecting breast cancer earlier. Similar results have been shown with diabetic retinopathy detection and arrhythmia detection from ECG analysis amongst others. Furthermore, ML has been utilised in microbiology labs to improve efficiency in blood culture analysis, enabling the selection of appropriate antibiotics within 24 to 48 hrs.
As such, AI tools can enhance medical diagnostics by improving accuracy, reducing costs, and saving time compared to traditional methods. They minimise human error and deliver faster, more precise results. In the future, AI could assist clinicians in real time by offering insights and decision support. Using machine learning, AI can detect abnormalities such as tumours, and provide quantitative data to support quicker, more reliable diagnoses.
With regards to disease management, AI tools can assist in the deliverance of personalised medicine; an approach that seeks to take into account an individual’s unique characteristics like genetics, lifestyle, their environment to tailor medical care to the individual thus improving patient outcomes. AI has become a powerful asset in advancing personalised treatment by enabling the analysis of complex datasets, forecasting outcomes, and enhancing the effectiveness of treatment strategies. Notably, ML has been used to accurately predict cancer patients’ response to various chemotherapies via the analysis of their gene expression data (with a demonstrated prediction accuracy of >80%). Another interesting use has been in the modelling of individual patient response to various classes of antidepressants using electronic health record analysis. Of course, further research is required in the use of AI tools for predictive modelling of treatment response to effectively train these tools and improve their reliability prior to real-world clinical use on a widespread basis.
AI tools can be used in a predictive capacity not solely on an individual level, but also on the population level. ML algorithms and other AI technologies can be used to analyse medical data such as lifestyle factors and patient medical history to identify patients which are at an increased risk of developing chronic conditions such as cardiac or endocrine disorders. Thus treatments and prevention strategies can be more targeted, reducing costs and improving health outcomes. Similarly, AI predictive models can be used to identify patients at high risk of hospital readmissions and targeting interventions accordingly.
Even in mental healthcare, AI has shown potential in delivering personalised, accessible care tailored to individual needs. AI tools can supplement the work of therapists, psychiatrists and other mental health professionals; AI-powered applications can aid the early diagnosis and treatment of mental health conditions, monitoring of patient progress and treatment efficacy. It is difficult to imagine AI dominating mental healthcare at the expense of human professionals, and is not desirable – the human element in a positive therapeutic relationship is generally critical to successful treatment. However, such tools show its potential as a means to supplement the existing work done by professionals to increase their effectiveness.
The above is not an exhaustive list of AI tool applications in healthcare serving to illustrate the broad impact this group of technologies can have in transforming healthcare.
Adoption by Healthcare Industry
Despite the above applications, currently the healthcare industry is “below the global average” in its adoption of AI when compared to other industries. The World Economic Forum (WEF) identifies three main obstacles to be overcome that are hindering AI adoption in healthcare:
- The complexity of AI in healthcare discourages policy-makers and business leaders from embracing its implementation; both in understanding the value of AI applications in healthcare and in securing financing.
- A misalignment between technical decisions and strategic objectives; the WEF identified a failure of public leaders to integrate AI into their health vision leading to missed opportunities.
- Low confidence in AI in the context of a fragmented regulatory context; there is an understandable growing concern about the impact of AI as it becomes more and more prevalent. To avoid a breakdown in trust, transparent, effective and accountable regulations and governance are required.
Innovative collaboration between public and private health leaders aimed at building a sustainable ecosystem for AI in healthcare is required to harness the full potential of these new technologies.
EU Regulation
The EU has enacted several pieces of legislation to address the challenges faced in the confluence of AI and healthcare. The European AI Act, which came into effect on 1st August 2024, and aims to enable the responsible development and use of AI throughout the EU. It sets strict requirements for high-risk AI systems, such as those used in healthcare with a focus on transparency, human oversight, and risk mitigation helping to build confidence in these tools. At the same time, the AI Act aims to reduce administrative and financial burdens on AI developers and SMEs by setting clear requirements and obligations with regards to AI applications. The implementation and enforcement of this act is currently under the purview of the newly established EU AI Office. Together with the AI Innovation Package and the Coordinated Plan on AI, the EU is aiming to strengthen investment and innovation in this AI while safeguarding citizens’ rights.
Furthermore, understanding that the development of effective AI tools in healthcare depends on access to high-quality, reliable, diverse data, the EU has established the European Health Data Space (EHDS) in 2025 to secure the secondary use of electronic health data for research, innovation, and policy-making. The EHDS is the first common EU data space for a specific sector and seeks to create a unified framework for the use and sharing of electronic health data throughout the EU, helping to address the fragmented regulatory landscape. The EHDS enables the training and assessment of AI algorithms used in medical devices, diagnostic tools, and digital health applications, supporting the development of innovative healthcare solutions while upholding robust data protection and ethical standards. It also contributes to enhancing patient safety and promoting equity in AI-powered healthcare.
Finally, recognising that there is currently limited market uptake and adoption into clinical practice of AI, the EU has embarked on a series of initiatives targeting at accelerating AI integration in healthcare, under the framework of AICare@EU. Key components of this framework include:
- A study on AI deployment in healthcare, identifying challenges in four main areas: technology/data, legal/regulatory, organisational/business, and social/cultural factors.
- EU4Health (WP2024) Call for Proposals, closed January 2025, to promote the safe and effective use of AI in clinical settings.
- SHAIPED project, launched in March 2025, to develop and deploy AI tools using the EHDS infrastructure.
- AI strategy and health priorities, including the promotion of industrial AI applications and public service improvement, and the development of a Biotech Act to accelerate AI solutions in biotech under EHDS.
International cooperation also plays a vital role, with DG Health and Food Safety partnering with WHO Europe, the OECD, and engaging with G7 and G20 to support AI adoption and policy alignment in healthcare.
Conclusion
AI has the potential to transform healthcare by improving diagnosis accuracy, clinical efficiency, population health management, and personalised care. Applications such as predictive analytics, virtual health, and mental health support show promise, but challenges like bias, lack of personalisation, data privacy, and public trust must be addressed.
Regulatory efforts such as those undertaken by the EU noted above will help to revolutionise the healthcare industry via the further development, uptake, and integration of AI into healthcare, while addressing these concerns. Care needs to be taken to avoid pitfalls in AI, such as via the provision of data diverse enough to avoid bias, and further research and refinement is required to have AI tools reach a sufficient level of reliability and quality for widespread use. Successful integration depends on interdisciplinary collaboration, ethical standards, and continued innovation. With these in place, AI can significantly enhance patient outcomes, system efficiency, and access to personalised, high-quality care.
