Recommended picks

FREMONT, CA: Machine learning is increasing in all industries as technology advances. When it comes to healthcare, it has had a significant influence on the system. ML in healthcare may perform miracles by processing massive datasets beyond human capabilities, converting data analysis into clinical insights that can help physicians provide care, resulting in better results. Let's take a look at three recent machine learning advancements in the healthcare industry.

Predicting COVID-19 severity

The National Covid-19 Cohort Collaborative (N3C) is a centralized electronic health record currently houses a Covid-19 cohort. Using this information, researchers discovered that treatment with blood thinners might lower the risk of mortality in Covid-19 users. In addition, according to the researchers from the Universities of Colorado, Rochester Medical Center, Michigan, and John Hopkins, this data can assist the rigorous proof-based creation of predictive and diagnostic tools to improve clinical treatment and policies. Data from 34 medical facilities get utilized in the study, including 174,568 people who tested positive for Covid-19 and 1,133,848 who tested negative between January 2020 and December 2020.

Stroke recovery prediction

A group of international scientists led by EPFL has created a system that can analyze and predict the prognosis of stroke patients by combining information from the brain's connectome with machine learning.

Scientists examined the connectomes of 92 individuals 14 days after a stroke for this study. Then, they monitored improvements up to three months later by using a standard scale to assess motor deficits. It allowed them to keep track of the patients' progress while they recovered.

Detecting Alzheimer’s during phone conversations

Alzheimer's is a disease for which there is currently no cure. The severe illness affects millions of individuals globally. As a result, scientists have begun to investigate novel methods of predicting who would be afflicted by the disease. A previous study has shown that one of the early symptoms of Alzheimer's disease is speaking more slowly than usual. Another analysis got conducted to identify similar speech difficulties. The team in Japan has been utilizing the Telephone Interview for Cognitive Status (TICS-J) exam, in which phone conversations are recorded and evaluated. Because manually listening to phone calls is time-consuming, researchers created a machine-learning system to listen to and assess phone conversations. They propose that these algorithms got utilized to give a low-cost and easily accessible type of early Alzheimer's testing.