📅 Wednesday, 7th October, 2026
⏰ 11:00 h – 12:00 h
- Format: Online (live)
- Duration: 60 minutes
- Level: Basic to Intermediate
- Language: English
Artificial intelligence is increasingly supporting healthcare organisations in areas such as diagnostics, medical imaging, patient monitoring, risk prediction, treatment optimisation, and operational planning. As AI becomes more deeply embedded in clinical workflows, healthcare providers face a common challenge: ensuring that AI systems are not only accurate, but also understandable, reliable, and usable in real-world healthcare environments.
Bridging the Gap Between AI Models and Clinical Trust
In the second session of our Explainable AI (XAI) webinar series (you can rewatch the first session, XAI for decision makers here), we will explore how XAI can help bridge the gap between technical model performance and practical clinical adoption.
This session focuses on how to design and communicate AI explanations that support healthcare professionals in understanding, validating, and appropriately using AI-powered insights.
These capabilities are also increasingly relevant to regulatory readiness in Europe, as the EU AI Act places requirements relating to transparency, output interpretation, and effective human oversight on certain high-risk healthcare AI systems. While this is not a regulatory session, the webinar will briefly explore how practical, clinician-centred approaches to XAI can support these expectations.
How can clinicians trust an AI recommendation if they do not understand how it was reached? The reality is that technical accuracy alone rarely drives adoption in healthcare. Even highly performant models may remain unused if healthcare professionals cannot assess the reasoning behind predictions, understand the contributing factors, or determine when an output should be challenged.
Why Explainable AI matters in Healthcare
Healthcare decisions directly affect patient outcomes. As a result, clinicians must be able to evaluate and understand the recommendations provided by AI systems before incorporating them into their workflows.
One of the most common barriers to AI adoption in healthcare is not model performance, but lack of trust.
Questions frequently raised by clinicians include:
- Why was this patient classified as high risk?
- Which factors contributed most to this prediction?
- Can this recommendation be clinically justified?
- Does the explanation align with established medical knowledge?
- When should the model’s output be challenged or reviewed?
Without meaningful explanations, AI systems can be perceived as black boxes that are difficult to validate in real-world clinical settings.
Explainable AI helps address these concerns by making model reasoning more transparent, allowing clinicians to better understand predictions and integrate AI insights into their decision-making process.
Healthcare organisations are increasingly looking beyond model accuracy when evaluating AI solutions. To create meaningful value in clinical settings, AI systems must also be:
- Transparent enough for practitioners to understand the factors behind predictions and recommendations
- Reliable and robust across diverse patient populations and clinical scenarios
- Aligned with clinical reasoning and established healthcare practices
- Capable of supporting collaboration between data scientists, healthcare professionals, and decision makers
- Trusted by end users who are ultimately responsible for patient care
While lack of trust is often cited as a barrier to AI adoption, it is rarely the only challenge. Effective AI deployment also requires clear communication, interpretability, and a shared understanding of model behaviour across both technical and clinical teams.
Explainable AI provides the tools and methodologies needed to address these challenges and transform AI from a technical output into a practical decision-support capability.
Making Explainability understandable to clinicians
Producing explanations is only the first step.
Many XAI techniques generate outputs that are meaningful to data scientists but difficult for healthcare professionals to interpret within the context of patient care. For AI to become truly useful in healthcare, explanations must be translated into forms that align with clinical thinking and decision-making processes.
This session will explore how do we translate AI explanations into clinically meaningful insights?:
- What clinicians actually need from AI explanations
- How different healthcare stakeholders interpret model outputs
- Techniques for presenting model reasoning in clinically meaningful ways
- How XAI can strengthen collaboration between healthcare experts and AI development teams
The objective is not simply to explain the model, but to ensure that explanations support informed clinical judgement. Many organizations successfully implement sophisticated XAI techniques, yet the resulting visualizations, metrics, and technical outputs are often designed for data scientists rather than healthcare professionals.
Participants will learn how to move from technical explanations to explanations that support clinical judgment and decision-making.
What you will learn
During this session, participants will gain practical insights into:
- Why trust remains one of the biggest barriers to healthcare AI adoption
- The role of Explainable AI in supporting clinical decision-making
- Translating technical explanations into clinician-friendly insights, the difference between explaining a model and communicating an explanation effectively
- Designing transparent explanations for different healthcare stakeholders that fit within clinical workflows
- How XAI can support transparency and human oversight for high-risk healthcare AI under the EU AI Act
- Best practices for human-centred approaches to AI adoption in healthcare
The focus is firmly on creating AI systems that healthcare professionals can understand, evaluate, and confidently use in practice.
By the end of this session, participants will understand how Explainable AI can help create healthcare AI systems that are:
- Transparent – clinicians can understand the factors driving predictions
- Interpretable – explanations are presented in clinically meaningful ways
- Trustworthy – healthcare professionals can confidently evaluate and use AI outputs
- Adoptable – AI solutions fit naturally into clinical workflows
The emphasis will be on making explainability useful, understandable, and actionable for healthcare professionals.
At the end of the session, our Chief Data Scientist, Burcu Kolbay, will host a live Q&A.
Special guest speaker: joining our Chief Data Scientist will be Dr. Josep Vidal, a member of the multidisciplinary and multi-territorial AI projects working group at the Institut Català de la Salut (ICS). Dr. Vidal will share practical perspectives on the adoption of AI in healthcare environments, the importance of explainability for clinical decision-making, and the challenges of translating AI insights into information that healthcare professionals can confidently use in their daily practice.
Who should attend
This masterclass is designed for professionals involved in the development, deployment, evaluation, or adoption of AI in healthcare, including:
- Clinical Informatics Specialists
- Digital Health Leaders
- Medical Device and HealthTech Professionals
- Physicians and Clinical Researchers
- Hospital Innovation Teams
- Healthcare Decision Makers
No advanced technical background is required. The session is designed to foster dialogue between technical and clinical communities.
Event Details
📅 Wednesday, 7th October, 2026
⏰ 11:00 h – 12:00 h
- Format: Online (live)
- Duration: 60 minutes
- Level: Basic to Intermediate
- Language: English
Join this second session of our Explainable AI webinar series to explore how healthcare organisations can develop AI solutions that clinicians trust, decision makers understand, and governance and regulatory stakeholders can confidently evaluate.