Article Highlight | 17-May-2022
Better face masks with optimum filtration efficacy could be created with assistance from machine learning predictions which decide the best materials mixture, with important time and value financial savings.
Agency for Science, Technology and Research (A*STAR), Singapore
picture: With assistance from machine learning predictions to decide the best materials mixture, higher face masks with optimum filtration efficacy could be created with important time and value financial savings.
view extra
Credit: A*STAR
Researchers from the Agency for Science, Technology and Research (A*STAR) have efficiently used machine learning in a research to enhance the filtration effectiveness of Egyptian Cotton (EC) face masks.
The research was printed in Materials Today Advances in December 2021. The researchers from A*STAR’s Institute of Materials Research and Engineering (IMRE) and Institute for Infocomm Research (I2R) used machine learning algorithms (Lasso and XGBoost machine learning fashions) to decide essentially the most excellent mixture of material layers utilizing the information of single-layer materials properties. Characteristics of EC materials with differing thread counts have been analysed, with the materials getting used to create triple-layered masks with completely different layer mixtures and stacking orders. Their filtration efficiencies have been then measured and evaluated primarily based on differential strain (ΔP), particle filtration effectivity (PFE) and bacterial filtration effectivity (BFE).
Study outcomes confirmed that filtration effectivity is mostly higher with materials which can be thicker and have smaller pore sizes, due to the improved packing density and subsequent seize efficiency. They discovered that essentially the most optimised ranges of PFE (45.4 %) and BFE (98.1 %) for cotton fabric-based masks have been achieved from stacking EC materials within the order of thread rely 100-300-100.
Through the research, AI know-how was discovered to be correct and helpful in guiding future design and development of masks. With important value and time financial savings, it’s a welcome departure from the present masks improvement course of, which is lengthier as repeated experimentation is completed utilizing human instinct for materials choice.“Our findings present that machine learning is in a position to assist design high-performance face masks in a sensible and environment friendly means, and information our discovery of novel linkages in supplies science. I consider that we are able to use an analogous machine learning-based prediction method for different materials design functions to advance clever and sustainable manufacturing in Singapore”, mentioned Dr Kai Dan, Senior Scientist from the division of Soft Materials at A*STAR’s Institute of Materials Research and Engineering (IMRE).
About the Agency for Science, Technology and Research (A*STAR)
The Agency for Science, Technology and Research (A*STAR) is Singapore’s lead public sector R&D company. Through open innovation, we collaborate with our companions in each the private and non-private sectors to profit the economic system and society. As a Science and Technology Organisation, A*STAR bridges the hole between academia and trade. Our analysis creates financial progress and jobs for Singapore, and enhances lives by enhancing societal outcomes in healthcare, city residing, and sustainability. A*STAR performs a key function in nurturing scientific expertise and leaders for the broader analysis neighborhood and trade. A*STAR’s R&D actions span biomedical sciences to bodily sciences and engineering, with analysis entities primarily positioned in Biopolis and Fusionopolis. For ongoing information, go to www.a-star.edu.sg.
Disclaimer: AAAS and EurekAlert! should not accountable for the accuracy of reports releases posted to EurekAlert! by contributing establishments or for the usage of any data via the EurekAlert system.
https://www.eurekalert.org/news-releases/952941
