The Politecnico di Milano has obtained from the European Commission important funding for two research projects: one for the fight against breast cancer and the other for the fight against climate change..
This takes the form of two ERC Advanced Grants 2021, funding awarded by the European Research Council to researchers well-established in their field, in order to carry out innovative and high-risk projects. The selection for this type of funding is very competitive: this year, out of 1735 projects submitted, only 14.6% obtained the funds. With these two projects, the Politecnico di Milano has been awarded a total of 86 European Individual Grants ( ERC and Marie Curie).
Daniele Ielmini, professor at the Department of Electronics, Information and Bioengineering, will lead ANIMATE (ANalogue In-Memory computing with Advanced device TEchnology), a project that aims to develop a new computing concept to reduce energy consumption in machine learning. We don’t think about it when we use a computer, but the energy cost of the actions we perform on the internet, starting with the everyday things, is very high. Data centres, which currently meet most of the world's AI needs, now consume about 1% of global energy demand, with growth expected to reach up to 7% by 2030.
Apparently simple operations, such as searching for a consumer product or service (for example when we book holidays or choose a film on a streaming site) are based on data-intensive algorithms and have an significant impact on the production of greenhouse gases: it has been estimated that training a conventional neural network for artificial intelligence produces the same amount of carbon dioxide as 5 cars in their full cycle of use.
Professor Ielmini’s preliminary ANIMATE research has shown that computational energy requirements can be reduced by closed-loop in-memory computing (CL-IMC), which can solve linear algebra problems in a single computational step. In CL-IMC, the time to solve a given problem does not increase in proportion to the size of the problem, unlike other computing concepts, such as digital and quantum computers. Thanks to the reduction in calculation time, CL-IMC requires 5,000 times less energy than digital computers with the same precision in terms of number of bits. Ielmini's project will develop the device and circuit technology, system architectures and set of applications to fully validate the CL-IMC concept.
Professor Ielmini is an expert in artificial intelligence and supercomputing: we recently talked about this in relation to another research project, intended to develop a new type of circuit for cryptography based on the concept of physical unclonable function.
Manuela Raimondi, Professor in the ‘Giulio Natta’ Department of Chemistry, Materials and Chemical Engineering, combines mechanobiology, bioengineering, oncology, genetics, microtechnology, biophysics and pharmacology to develop a new method for the treatment of breast cancer.
In this type of cancer, the aggressiveness is related to the fibrotic stiffening of the tumour tissue. Fibrosis progressively prevents drugs from reaching the tumour cells, due to the formation of a matrix with mechanical properties that stabilise the tumour's vascular network. Professor Raimondi's BEACONSANDEGG research Professor Raimondi is developing a platform capable of recapitulating tumour fibrosis by exploiting the vascularisation of a living organism.
The project is called BEACONSANDEGG (Mechanobiology of cancer progression): it will model microtumours at various levels of fibrosis, bypassing the fact that some characteristics of the tumour are not reproducible in vitro. Human breast cancer cells adhered to 3D polymeric microplates will be used. The microtumours will be implanted in vivo in the respiratory membrane of embryonated avian eggs in order to elicit a fibrotic foreign body reaction in the microtumours. The geometry of the 3D microsupports will be manipulated to condition the infiltration of the microtumours by the vessels and cells of the embryo. This study model will be validated with anticancer drugs whose clinical outcome is known to depend on the level of tumour fibrosis.
It will also provide a standardisable and ethical platform to promote the clinical translation of new therapeutic products in oncology. This is a key issue for Professor Raimondi: some of the research and modelling tools she has developed over the last ten years have precisely this goal: to reduce drastically or replace the pre-clinical experimental phases in vivo, for example, with the use of 3D supports for cell cultures and microfluidic chambers for tissue and organoid culture (we talked about this in Issue 6 of MAP).
