Second The European Commission’s Joint Research Centre (JRC), around 25% of the world’s annual cancer cases occur in Europe and half of all European citizens will develop cancer at some point in their lives. Artificial intelligence (AI) and big data analytics could help improve the situation. Scientists and politicians discussed the role of AI in cancer research during an event entitled ‘Equity, excellence and innovation — modern cancer care for all«, jointly organised by the Swedish Presidency of the Council of the European Union (EU) and the European Commission on 1 February.
Launched in January 2023, the European Initiative for Imaging in Cancer, a flagship of European Beating Cancer Plan, aims to make the most of digital technologies such as intelligence artificial and high-performance computing to make available and link resources and databases across the EU to a cancer imaging infrastructure that is easily accessible to doctors, researchers and innovators.The initiative will ensure full compliance with EU values and standards on ethics, reliability and safety. This would not only enable faster and more accurate clinical, diagnostic and therapeutic decision-making, but would also foster the development of innovative tools for personalised medicine, says Marco Marsella, Head of the European Commission’s eHealth, Wellness and Ageing Unit. «In the next 3 to 4 years we will be successful with this initiative and contribute to the European Plan to Fight Cancer,» he said.
In March 2021, JAMA Open Network published a study on a new AI-supported point-of-care digital microscopy diagnostic platform, created by researchers from Sweden, Finland and Kenya to facilitate cervical cancer screening in a rural clinic in Kenya, where access to pathologists is limited. The platform digitized Pap tests with a handheld full-slide microscope scanner, analyzed the slides using a cloud-based deep learning system, created high-quality images for various diagnoses, and enabled remote consultation at the point of care.The platform successfully identified atypical cervical smears with high sensitivity. The technology «can also be used in low-resource settings in the European region,» says one of the study’s authors, Nina Linder of the Finnish Institute of Molecular Medicine at the University of Helsinki. The author believes that, at the current state of its development, AI is not able to solve the most complex cases or to elaborate revolutionary medical insights from data but could automate routine tasks and recognize samples to offer data-driven prediction of results.
Delving into how AI can accelerate cancer research, scientists described an ambitious project that involves integrating single-cell multiomics, high-content imaging and personalised disease models with AI and machine learning. AI would be used to generate and analyze large-scale, multidimensional datasets that provide information on cell development and differentiation to elucidate the cellular and molecular mechanisms underlying cancer onset and progression.
Nikolaus Rajewsky of Germany’s Max Delbrück Center for Molecular Medicine explains that machine learning can be used to exploit data from hospitals, biobanks, cohorts and hematoxylin-eosin stains.It also supports the importance of connecting existing and emerging European cancer centres to share data, establish quality controls and include genetic diversity. «There is much more genetic diversity in Europe than we may know,» he says. «A good example is Portugal, where the genetic variability is really very different from that found in Scandinavia, Germany, France or Italy.» Recognizing this diversity is essential to «analyze data in a truly human-oriented way.»
Despite the many benefits of AI in cancer care, some challenges need to be addressed, including the lack of structured health data. However, the EU’s response is European Health Data Space, a legislative proposal to provide a more digitised and connected healthcare ecosystem between Member States that is coherent, reliable and secure. Another major concern is AI’s potential to further perpetuate existing health inequalities and disparities.For example, if AI tools and algorithms are not developed and tested on different populations, they may not be effective or accurate for some subgroups. The World Health Organization guidelines on AI ethics and governance for health should be followed to support patient consent, transparency and data protection.
The path may be long and bumpy, but these initiatives, once implemented, could put the EU at the forefront of AI and automated delivery in cancer care.
Access to the site is restricted and reserved for healthcare professionals
You have reached the maximum number of visits
Source — https://www.univadis.it/viewarticle/sweden-2023-eu-l-intelligenza-artificiale-aumenta-la-presa-2023a10002k0