AI & Health Winter School: abstracts

Tuesday 12th March

Annette ten Teije (VU, Computer Science, Learning and Reasoning group)
Design patterns for Hybrid Learning and Reasoning in medical AI
 
In recent years there has been a strong increase in systems that combine data-driven machine learning with knowledge-driven reasoning. We have proposed a set of compositional design patterns to describe a wide variety of systems that combine statistical techniques from machine learning with symbolic techniques from knowledge representation. These design patterns help to understand the rapidly growing literature for such neuro-symbolic systems. They can also be used as guidelines for the design of such systems. In this talk we will introduce a notation for these design patterns, and illustrate these patterns with neuro-symbolic systems in the medical domain. We will dive into a guideline-informed reinforcement learning method for mechanical ventilation in critical care along the lines of these patterns.
Sjors in ‘t Veld & Martijn Schut (Amsterdam UMC, Translational AI Lab)

Leveraging blood derived omics for ​AI enhanced cancer detection

The integration of omics technologies and artificial intelligence (AI) has revolutionized cancer detection. This lecture explores cancer biomarkers’ historical evolution, emphasizing omics’ pivotal role in advancing detection techniques. Two scientific papers discussing CancerSEEK and thromboSeq respectively, exemplify the power of omics in identifying novel biomarkers for cancer diagnosis. Additionally, the lecture delves into the implementation of AI-enhanced cancer detection tests, showcasing how AI models process vast omics datasets to improve diagnostic accuracy. Through case studies of CancerSEEK and thromboSeq and discussions of AI algorithms, attendees gain insights into the potential of synergistic approaches to transform cancer diagnosis and personalized medicine. The lecture concludes by highlighting future directions and practical strategies for integrating omics and AI into clinical practice, underscoring the importance of ongoing research and innovation in this critical field.
Martijn Schut (Amsterdam UMC, Translational AI Lab)

The Next Frontier: Medical Foundation Models

This talk explores innovative AI paradigms reshaping healthcare provision. We will delve into emerging frameworks that blend medical expertise with technological advancements, aiming to optimize patient care and operational efficiency. From integrated data-driven platforms to community-centric approaches, the discussion navigates diverse models driving transformative change in healthcare delivery. Emphasizing collaboration between practitioners, researchers, and industry stakeholders, it highlights the pivotal role of interdisciplinary synergy in fostering sustainable medical foundations. 
Vincent Francois-Lavet (VU, Computer Science, Quantitative Data Analytics group)

Reinforcement Learning: Applications to Healthcare Solutions

This presentation provides an overview of reinforcement learning (RL) with a focus on its relevance and applications within the healthcare domain. Beginning with a foundational introduction to RL principles, the discussion gradually transitions towards practical applications in healthcare settings. Through a concrete example in the context of treatment planning for cancer, this presentation illustrates how RL techniques can be utilized to address challenges in treatment optimization.

Wednesday 13th March

Emma Beauxis-Aussalet (VU, Computer Science, User Centric Data Science group)
Towards responsible and explainable AI

AI can have critical impacts on people or society, for instance by creating bias and discrimination. These impacts are difficult to assess and communicate to practitioners or end-users, and this gap worsens the ethical issues. This lecture will discuss the means to inform AI practitioners and end-users about the risks of error and bias. First, we will  investigate this issue through the lens of a seemingly simple task: measuring AI errors. We will highlight key challenges with collecting reliable test data, with ensuring that vulnerable populations are well-represented, and with assessing the test data as well as the test results. Then we will overview the ways to explain AI results, and to understand why certain results are obtained. We will outline the state-of-the-art, and draw critical views on its limitations. Finally, we will argue that designing responsible AI systems also requires designing the human organisations in which such AI systems take place.

Brenda Penninx, George Aalbers (Amsterdam UMC, Psychiatry) & Sjors van de Ven (VU, Biological Psychology)

AI for daily-life stress research

This tri-partite lecture starts with a presentation by prof. dr. Brenda Penninx who will introduce the interdisciplinary Stress-in-Action (SiA, www.stress-in-action.nl) consortium (50+ members, 7 universities), its rich multimodal data (e.g., genetics, actigraphy, digital phenotyping, clinical diagnoses) and future data collection, the questions that motivate this data collection, as well as the potential added value of artificial intelligence (AI) for answering them. Brenda’s general discussion is illustrated by postdoc George Aalbers and PhD researcher Sjors van de Ven, who will each discuss their ongoing research to show how they apply AI techniques to make sense of temporally dense, high-dimensional data and generate clinically and theoretically useful insights. George will show an analysis that applies explainable AI to digital phenotyping data for depression/anxiety disorder recognition. Sjors will present a study in which he implemented machine-learning algorithms to address methodological challenges in detecting physiological stress responses in daily life.
Marike van der Leeden (Amsterdam UMC, Rehabilitation Medicine)

Automated recognition of (recovery in) functioning of COVID-19 patients with natural language processing

COVID-19 may result in impairments in functioning which may require rehabilitation. However, assessment of functioning based on the International Classification of Functioning, Disability and Health (ICF) is not easily obtained.  We tested the feasibility, reliability and internal validity of automated assessment of functioning of key ICF categories and their levels of hospitalised COVID-19 patients. The assessment is performed by applying state-of-the-art natural language processing (NLP) methods on unstructured text in electronic health records (EHR) from a large academic hospital.
In this talk, I will discuss the methods used to develop these NLP algorithms and what the next steps will be to use the algorithms in rehabilitation care and research. I will also share lessons learned to carry out such a project, in terms of technology, data and human factors.

Thursday 14th March

Mohammad Rezazade Mehrizi (VU, Centre for Digital Innovation) & Willem Grootjans (Leiden UMC)

End-user adoption of AI in clinical practice: A socio-technical perspective

Like many other technologies, AI solutions witness a complex journey until they are actually used by the end users in organisations. Especially in the medical context, this journey is unpredictable and involves socio-technical factors beyond only the intentions and decisions of individual users. Metaphorically, the introduction of AI into clinical practice is like planting a seed into a jungle (not even a protected green house!). The eventual adoption and use of such a technology requires a wide range of conditions coming together to naturally nurture an effective implementation and use. In this session, we unpack such a process through a socio-technical change perspective. We learn how we can design, act, and assess the adoption of AI solutions through systematic and dynamic arrangements among actors, technologies, structures, and tasks. In the second part, we collectively discuss a real-life use-case of adopting an AI solution in radiology at the LUMC. The aim is to apply our learning into a practical example and learn how to from a critical and practical view on the adoption of AI solutions in clinical practice.
Ilaria Tiddi (VU, Computer Science, Knowledge in AI group)

Knowledge Engineering for Hybrid Intelligence 

Hybrid Intelligence (HI) is a rapidly growing field aiming at creating collaborative systems where humans and intelligent machines cooperate in mixed teams towards shared goals. In this talk, we will discuss how ontologies and knowledge engineering methods can be used to design, compare and improve Hybrid Intelligence applications.
Matthan Caan (Amsterdam UMC, Biomedical Engineering & Physics)

What no eye has seen: how AI is revolutionizing medical imaging

For long time, human sense was the only diagnostic measure available in healthcare. Today, along the chain of clinical decision making and treatment planning and monitoring, medical imaging plays a pivotal role. When should these images be acquired and how should they be evaluated? How can we make Magnetic Resonance Imaging faster through the use of AI? This is for instance relevant in the context of stroke treatment. We will explore the integrated view on data analysis through so-called end-to-end deep learning for accurate delineation of white matter lesions in the brain. Major depression treatment outcome prediction on multimodal data can be realized and potentially provide a first treatment decision making tool to psychiatrists.
Organized challenges on a specific medical imaging task are an attractive and energizing way of competing and collaborating, with high impact. This holds for research and education.
Finally, we will reflect on involving and informing the general public about our work by presenting in the press and regular media.

Friday 15th March

Willem Jespers (Leiden University, Leiden Academic Centre for Drug Research)

Computational Methods in Drug Discovery: AI and Physics Based Approaches

Pharmaceutical science is changing; while perhaps not a paradigm shift, the influence and catalytic effect of data science on drug discovery cannot be denied. Scientific data is becoming public and even open access. Moreover, better computing capabilities and more data make it easier to use prior data to improve ongoing work.  Chemical Biology explores biology via chemical tools. In practice this means that the molecular interaction space of protein targets is probed. Computational Chemical Biology is the computational addition to these goals and is located in between the fields of medicinal chemistry, cheminformatics, bioinformatics, and computational biology. In this lecture, bio and cheminformatic approaches coupled to structure-based methods using crystal structures are discussed.