May 14, 2020

expert reaction to paper using machine learning tools to identify three biomarkers associated with COVID-19 death

Research, published in Nature Machine Intelligence, reports on machine learning tools used to identify three biomarkers associated with COVID-19 death.

Prof Duc Pham, Chance Professor of Engineering, University of Birmingham, and Fellow of the Royal Academy of Engineering, said:

“This is a very good and straightforward application of pattern classification.  The usefulness of the work would depend on whether any relevant biomarkers were left out of the experiments.  To give confidence in the results, it would also be necessary to increase the sizes of the test and validation data sets.  There should not be difficulties obtaining more data, given the unfortunately large numbers of COVID-19 casualties.”

‘An interpretable mortality prediction model for COVID-19 patients’ by Li Yan et al. was published in Nature Machine Intelligence on Thursday 14 May 2020.

DOI: 10.1038/s42256-020-0180-7

https://www.nature.com/articles/s42256-020-0180-7

All our previous output on this subject can be seen at this weblink:

www.sciencemediacentre.org/tag/covid-19

Declared interests

Prof Duc Pham: “I have no interest to declare.”