Using Machine Learning to Understand Alzheimer’s Disease
Even just within the span of the last two decades, technological innovations have helped throttle medical research on unprecedented levels. These innovations have directly corresponded to decreases in death rates, as reported by the Center for Disease Control. Diseases with historically high death rates such as stroke, heart disease, HIV, and cancer have dropped anywhere between a whopping 22%-66% in as much time. However, two diseases have actually seen a spike in their death rates — that of Parkinson’s and Alzheimer’s diseases. The mortality rate for these two neurodegenerative diseases has increased by an enormous 48% for Parkinson’s and an alarming 84% for Alzheimer’s. Though these diseases remain devastating to patients and their families alike, one innovative company sees new hope from an unlikely source — Artificial Intelligence.
Understanding Aging’s Role in Alzheimer’s
Before heading into how AI is helping medical researchers, it’s essential to understand the factors that impact the progression of Alzheimer’s. According to representatives from clinical research company Alkahest, aging plays a pivotal role in the advancement of neurodegenerative diseases such as Alzheimer’s. In order to discover why this is so, research has been performed to isolate the body’s plasma as an extremely communicative aspect of the body’s aging process. Plasma contains many proteins that all carry varying signals and other information throughout the body. Because of this discovery, Alkahest decided to target plasma’s role in the progression of Alzheimer’s disease.
During Alkahest’s clinic research, specialists sampled blood from patients for analysis. Isolating differing cells would separate the plasma from the blood. Once the plasma was further paired down, there existed about 10,000-20,000 proteins that contained immense data that researchers wanted to be able to process. Among these proteins were such called chronokines — proteins whose levels are age-sensitive, along with enacting age-specific implications. The levels of adverse chronokines have been shown to rise with age, while useful chronokines simultaneously take a sharp nosedive.
Research with mice has shown that if older mice are injected with the plasma of younger mice, this has been shown to decrease the progression of some age-related disease. When the plasma of younger participants has been introduced to the systems of older participants, significant benefits have been observed. Though this being the case, simple plasma transfusions are not the end goal of Alkahest. If the cause of these rate fluctuations can be determined and possibly reversed, Alkahest researchers believe this can transform chronokines into a solution for slowing and decreasing age-related diseases such as Alzheimer’s. The issue going forward is analyzing the data contained within the plasma proteins. This challenge requires the use of complex data analysis.
Medical Research with Machine Learning
If medical researchers were uniquely tasked with analyzing what makes chronokines perpetuate the adverse effects of aging on a granular level, the task would be tremendously time, personnel, and cost-intensive. There is also the issue of considering all data variables — including patient health background, genetics, and the like. The sheer volume of data that would need to be reviewed as well as compared is beyond the ability to do so manually. This is where machine learning and artificial intelligence become integral to the research.
Machine learning is a subset of artificial intelligence that builds analytical models to spot patterns and “learn” from complex sets of data. In this instance, machine learning works by correlating information together to identify relationships in the data. This is possible by running the data through advanced algorithms to organize the data, process it, and identify relevant correlations.
Promising First Steps
Alkahest’s utilization of machine learning and other artificial intelligence technologies have given the field of Alzheimer’s research a major leg up. By teaming up with a variety of plasma therapy companies as well as IT companies large and small, they have been able to expand their aim of targets from the usual single target to three simultaneous therapeutic targets. While age and genetics are the two main contributors to the likelihood of developing Alzheimer’s, machine learning has allowed the company to extend the range of data processing exponentially. This research boost thanks to artificial intelligence is already helping researchers discover new data correlations. These discoveries may shed additional light on the complexity of Alzheimer’s and help researchers develop new preventative measures for those at risk and treatments for those presently afflicted.
SENTIR Research Labs believes in the power of AI and Machine Learning to help innovators conquer our most confounding medical dilemmas. You’re invited to learn more about SENTIR Research Lab’s mission of supporting researchers to bring exciting new solutions to fruition.
Learn more about SENTIR Research Labs.
Associated research:
