Evaluating COVID-19 treatments

Evaluating COVID-19 treatments

Can Machine Learning evaluate the efficiency of different COVID-19 treatments?

As the COVID-19 epidemic is expanding to many countries, COVID-19 poses a critical menace to global health. Hence, it is essential to uncover effective medications against COVID-19. Time is of the essence in this fight against the expansion of the disease. Researchers and medical doctors are actively operating clinical experiments on numerous drugs and publishing their outcomes continuously. For practical reasons, these trials include small groups of patients, typically, dozens to hundreds, hardly thousands. Some researchers concentrate their research on a variety of existing molecules, such as hydroxychloroquine and azithromycin, while others consider herbal plants, such as artemisia (Madagascar), on the other. 

These various remedies have been available for a long time. Machine learning can help in understanding what lies beneath the appearances in the data results collected. These results may come from new clinical trials aimed at COVID-19 or from previous results obtained for other diseases. 

Problems to solve

  • Can Machine Learning help support the results of different clinical studies on hydroxychloroquine?
  • Is it possible to estimate whether a particular drug, repeatedly used before the outbreak of COVID-19 shields from SARS-CoV-2?
  • Can Machine Learning aid in weighing the efficiency of several drugs used against COVID-19?
  • Benefits of TADA in determining
    the efficiency of COVID-19 treatments

    Mathematical techniques are regularly applied to clinical results to estimate the effectiveness of the drug tested. Nonetheless, Machine Learning is a smart alternative to scrutinize the outcomes of the clinical trials. TADA, the Small Data Machine Learning platform by MyDataModels, is ideally suited to investigate such data. It can recognize correlations among patients’ features and actual cure. It can also detect the significant criteria leading to a patient recovery impartially. One of the influencing factors might be to have been treated with hydroxychloroquine, or not. TADA can be used to determine these factors. And there is no need to be a data scientist to use TADA. The platform is easy to use, and anyone can operate it.

    It is often tricky to consolidate the results of several clinical research studies. TADA can investigate the results of similar studies as a global one. It can automatically put forth which patterns are prevalent toward disease recovery. Determining factors might be treatment by artemisia or clinical research, which issued the results. In other words, it can show, objectively, potential clinical biases, or treatment performance. 

    Last but not least, TADA makes it reasonable to investigate the impacts of COVID-19 on subjects who were previously treated by one of these drugs before the outburst of the COVID-19 pandemic. For example, a team of Chinese researchers worked on hydroxychloroquine for COVID-19 patients. During a follow-up survey, they discovered that none of their 80 other subjects who received long-term oral hydroxychloroquine (for lupus) had been confirmed to have SARS-CoV-2 virus or seemed to have associated symptoms. Besides, they remarked that, among the 178 patients diagnosed with COVID-19 pneumonia in their hospital, none were receiving hydroxychloroquine treatment before admission. TADA can help distinguish whether these kinds of results are biased (a small number of specimens) or whether they can be generalized (when merged with other analogous results). TADA accomplishes this in a few instants only.

    MyDataModels brings a self-service solution to researchers who want to swiftly identify a drug’s performance during clinical trials using their Small Data.

    TADA brings new possibilities for evaluating COVID-19 treatments

    Rheumatism and malaria are treated with chloroquine and its derivatives. Its further clinical applications, notably the antiviral activity, are increasingly investigated. Notwithstanding the standard small number of cases involved in human clinical trials, TADA can produce insightful meaning to the results obtained. Background treatments are urgently required to prevent hypoxemic respiratory failure and deaths from coronavirus infection 2019 (COVID-19).

    Machine Learning provides an objective means to evaluate the efficiency of COVID-19 treatments

    However, even though the volume of patients tested with potential treatments is modest, and there is an urgency to find a cure, any new drug proclaimed is not the coming of the Messiah. It is critical to remain rational in evaluating the different treatments, whether they regard existing drugs or a newly designed one. Machine Learning and, in particular, TADA is an excellent means for assessing the efficiency of COVID-19 treatments rationally. 







    Gordon C, Amissah-Arthur MB, Gayed M, et al. The British Society for Rheumatology guideline for the management of systemic lupus erythematosus in adults. Rheumatology (Oxford). 2018;57(1): e1-e45. DOI: 10.1093/rheumatology/kex286.