Machine learning for clinical prescription and other medical uses

por: Agencies – 26 February 2021, 07:52 nm

By: Notipress

– With machine learning comense a possible create models that preach the progress of patients in patients

MÉXICO.- A study describes the use of a learning machine to facilitate health analysis. You can use the clinical prescription

Although the use of machine learning in medicine has been extended, the difficulty is to find patterns in health records, as this type of information is not static or is regularly retrieved. Without embarrassment, a Carnegie Mellon University investigator developed a transparent and reproducible machine learning tool to facilitate health information analysis. This tool, called TL-Lite, can be used in clinical prediction, with the aim of predicting trends and results in individual patients.

Following the study published in Proceedings of Machine Learning Research, TL-Lite begins with data base visualizations and term with visual risk evaluations of a temporal model. The TL-Lite objective facilitates predictive prognosis and, according to its author Jeremy Weiss, can be used to predict severe thrombocytopenia during the stages of intensive care unit (UCI). Also to predict the survival of patients admitted to the UCI one day after the onset and to predict the microvascular complications in patients with type 2 diabetes.

The use of machine learning in medicine has seen a recent eye and in the context of the Covid-19 pandemic, it has been used to address problematic distances. A study published in the journal Journal of Medical Internet Research proposes the use of this technology to examine electronic medical records to better predict patients with Covid-19.

The machine learning models, in the context of the health of the health, suele require various dates and the large scale, afirma the study published in the year 2021. It is to be able to replicate its efficiency of the population with what would train. When these models are built inside a hospital, they are always effective for other patient populations, because they are trained with dates that are not representative of the entire population. To make sure it is limiting, investigators should try a learning machine called federated learning. This ground has been used to allow models to learn from many sources, for example, many hospitals, without exposing sensitive patient data.

In the wake of the pandemic, but in the area of ​​medicines and therapies, the learning machine also had to be a great help. A study published in the journal Nature Communications in February proposes the use of machine learning to identify existing medicinal sources that can be used to combat Covid-19 in adult adults. This is to identify with algorithms the intersection of the genes of the pathway and the pulmonary recurrence and the SARS-CoV-2. The next step is to identify existing drugs that act on these genes and perform clinical trials to test the efficacy of the more promising candidates.

The new technologies promise great advances in medicine, through an automated and accelerated analysis of data and the creation of models, mainly mediating machine learning. Although there are several ways to improve, the learning machine can be used as a tool to monitor clinical prognosis and to predict the progression of patients’ illnesses. In a near future, in accordance with its operation, these products will become extensively used in world health systems.

.Source