Alfonso Valencia and Maria A. Blasco. / A. Tabernero. CNIO
Alfonso Valencia and Nuria Oliver took part in the public event organised by the National Cancer Research Centre (CNIO) at CaixaForum Madrid on World Cancer Research Day.
Valencia leads the Life Sciences Department of the Barcelona Supercomputing Center - Centro Nacional de Supercomputación (BSC-CNS). Oliver, a pioneer in artificial intelligence (AI) research, is the director of Fundación ELLIS Alicante.
“We are experiencing a revolution, and the impact on research is clearly visible,” said Valencia. “We have more and better quality data than ever before, and the ability to analyze it.”
With the director of CNIO, Maria A. Blasco, they explained how AI is already changing the diagnosis and treatment of cancer and other diseases.
“In ten years, it is likely that artificial intelligence will be integrated into clinical decision support systems, which would evaluate patient data in real time and allow treatments to be adjusted dynamically,” said Alfonso Valencia, pioneer of computational biology in Spain and a world-renowned expert in this field. In the future that he envisages, diseases could be prevented with the help of smart systems that provide individualised and continuous monitoring.
For Maria A. Blasco, scientific director of CNIO, artificial intelligence “is a tool that will provide great support for research and will accelerate it, helping to diagnose and treat cancer.” At CNIO, AI is already a reality in areas such as: genome and big data analysis; imaging analysis; prediction of protein structures; and the discovery of anti-tumour drugs.
At the event held at CaixaForum with the support of ‘la Caixa’ Foundation, Blasco announced the creation of a new Artificial Intelligence programme at CNIO, with new research groups and the ability to provide cross-cutting support to all research at the centre. This has been possible thanks to 4.6 million in funding provided by NextGenerationEU European funds for the promotion of digital talent (managed by Red.es, part of the Spanish Department for Digital Transformation and Public Service).
“CNIO’s scientific priorities have always evolved in parallel with changes in scientific paradigms and new technologies. Artificial intelligence is one of these new paradigms,” said Blasco.
One concrete example of the application of AI at CNIO will be the analysis of more than 12,000 genomes of people in Spain, which will be sequenced thanks to CNIO’s new advanced equipment. This action is part of the Genome of Europe project, involving the Carlos III Health Institute (ISCIII), CNAG (National Centre for Genome Research) and CIBER (Centre for Networked Biomedical Research).
(Amigos/as CNIO)
Advanced simulation of tumour behaviour
Alfonso Valencia spoke about a future in which AI helps to anticipate diseases and, when they appear, to diagnose them with maximum precision and define the best treatment, a future that is already being built. Working at the Barcelona Supercomputing Center – Centro Nacional de Supercomputación (BSC-CNS), Valencia aims to develop ‘virtual copies’ of patients, technically known as digital twins. “The problem is that making copies of reality is very complex,” he said.
To simulate a whole person would require enormous amounts of the kind of information that defines their biology – data on their genes, their proteins, how their proteins interact with each other and with the environment… – and also their behaviour. With the digital model that would be built with that data, we could do a number of things such as try out possible treatments for their disease and even predict metastases and possible relapses.
Are we 100 years away from that? 50? “We don’t know,” says Valencia, “but the question I ask myself is how we can speed up the process. We don’t have a model of a complete patient yet, but we do have it for some parts: the heart, for example, or a tumour, how it evolves.”
It is not yet a model that reproduces a cancer “exactly”, but for now it is “good enough to generate hypotheses that can be validated experimentally,” he continued.
The model they have developed at the BSC “was impossible just five years ago, because we didn’t have the software, we didn’t have the data and we didn’t have the ability to calculate,” said Valencia. Still, “we are just at the beginning.” For now, the models that simulate cells are different from those that simulate the organs, and those that simulate the behaviour of the whole organism. The challenge is to be able to integrate the three types of models.
Challenges: ensuring access to data and its interoperability
There are still many challenges ahead. The main one, says Valencia, “is access to data from the health system to use them for research, along with the difficulty of making them coherent and interoperable. Issues about the regulation and evaluation of AI-based methods as well as the training of health professionals in these new technologies also need to be addressed.”
The ethical aspects of AI pose a challenge in themselves. Nuria Oliver spoke about them, noting that “any approach to AI must consider five fundamental pillars: technological, regulatory, ethical, educational and economic.”
Her recipe, and that of the foundation she directs, ELLIS Alicante Foundation, for AI to become “the best thing that happens to humanity” is based on the acronym FATEN: F for fairness; A for autonomy or accountability, i.e., “clarity regarding accountability for the consequences of algorithmic decisions”; and augmentation or the increase of human intelligence, “so that artificial intelligence systems are used to augment or complement human intelligence, not to replace it.”
“T for trust, and for transparency; E for education and equity; and N for No harm that is, to minimise the negative impact that may arise from the use of algorithmic decisions,” Oliver explained.
“It is important to apply a principle of prudence, to guarantee the security, reliability an reproducibility of the systems, always preserving the privacy of people”, Oliver said. “Only when we respect these requirements will we be able to advance and achieve socially sustainable artificial intelligence, by and for people.”
Screening of the short film Siempre Positivo
The event included a panel discussion, in which the speakers were joined by Miguel Calero, Deputy Director General of Applied Services, Training and Research at the Carlos III Health Institute (ISCIII), and Ignacio Astilleros, executive of the technology company Openchip and member of the philanthropic initiative Friends of CNIO.
After the round table discussion, there was a screening of the short film Siempre Positivo (Morena Films and Mediabrands Content Studio), in which coach Louis van Gaal talks about his experience as a cancer patient and about science as a solution.
This documentary is part of the ‘Siempre+Positivo’ initiative aimed at raising funds for cancer research at CNIO, through a clothing collection designed especially for this cause by El Ganso.
The event was presented and moderated by science journalist Pampa García Molina, director of the Science Media Centre Spain.