SCIENTIFIC RESEARCH TRY TO KEEP UP WITH TECHNOLOGICAL DEVELOPMENT. THE INTRODUCTION OF THE USE OF ARTIFICIAL INTELLIGENCE HAS ALSO EXPANDED INTO THE MEDICAL FIELD. AND, MORE GENERALLY, IN THE SCIENTIFIC ONE, ALLOWING YOU TO MAKE FURTHER STEPS FORWARD IN RESEARCH. NOW, WITH THE USE OF ARTIFICIAL INTELLIGENCE IT IS POSSIBLE TO ALSO IDENTIFY TOXIC SUBSTANCES
Technology e science they have always tried to go hand in hand and continue to do so even in the face of the latest technological developments. Thanks to a recent study, it was possible to exploit theArtificial intelligence to identify toxic substances, based exclusively on their molecular structure. A further step forward made by research in Swedenwas done thanks to the collaboration of Chalmers University of Technology and of theUniversity of Gothenburg.
Artificial Intelligence has undoubtedly changed our approach, even in everyday life. Just think of the assiduous use that many are now unable to do without. In this regard, the news dated May 21st that the European Council approved a law with the aim of harmonizing the rules on Artificial Intelligence, the so-called Artificial Intelligence Act.
But if in some cases this scientific and technological progress can almost be frightening, on the other hand it has also brought about an impact efficient contribution in the field of science and research.
The study conducted in Sweden
The chemicals are pervasive in modern society. For this reason, tests that can evaluate the potential danger of specific products for the environment and human health are becoming increasingly necessary.
This sometimes also involves long animal testing to certify its safety. Consequently, it highlights the need and urgency to evaluate alternative methods for such purposes. This was the starting point that led to the cooperation of Chalmers University of Technology and the University of Gothenburg, Sweden. In order to research a new method of identifying toxic substances.
Scientists have exploited the effectiveness of transformers (i.e. models like Chat Gpt) to extract information from chemical structures, allowing forecasts of toxicity and any further dangers.
Thanks to deep neural networks, in fact, these AI models analyze vast sets of data derived from laboratory tests. In this way, we try to improve the precision of the evaluation of the chemical toxicityeven for previously untested compounds.
Mikael Gustavsson e Erik Kristianssonwho contributed to and actively carried out scientific research activities, underlined the singular ability of this method predict toxicity of a substance based only on its chemical composition, refined over time through extensive data analysis.
But, above all, both highlighted the importance of this project, considering the high amount of toxic substances in circulation today, some of which are not yet completely identified.
An example of this is i Pfas (“perfluorinated alkylated substances” – perfluorinated alkylated substances), a group of alarming substances recently discovered in groundwater and drinking water.
Will Artificial Intelligence replace laboratory experiments?
There is still a modicum of skepticism in thinking that theArtificial intelligence can replace, as in this case, laboratory experiments. Often, in fact, some tools have not proven to be so effective as to be perfectly substitutable for laboratory tests. However, new studies conducted by Swedish researchers have shown that overcome conventional tools both in terms of precision both of versatility.
The study has, in fact, highlighted the potential of Artificial Intelligence in revolutionizing the toxicity assessments. In this way, we can try to imagine a future in which AI systems could increasingly replace laboratory experiments.
Likewise, they would draw themselves advantages also in safety and economic terms, reducing costs and accelerating the development of safer chemicals. Not only will it be diminished and almost completely avoided animal testingbut the identification of ecological alternatives to harmful substances will also be facilitated, further mitigating the negative effects of chemical pollution on both human health and ecosystems.
So, yes, AI could replace laboratory experiments, with a view to science and technology continuing their active collaboration in the common goal of protect the environment and the human health.