There are many degenerative diseases such as ALS and Alzeihmer’s disease that have a progression of symptoms. In many cases, this rate of progression is different in each individual. Scientists at MIT wanted to find a way to use machine learning and other mathematical tools in order to identify the patterns of these diseases. In addition, scientists wanted to explore if these patterns were shared among individuals. Using machine learning, scientists were able to identify many patterns of progression, many of them being non-linear. Scientists also worked with healthcare professionals using patient data in order to identify and connect patterns of progression. Through this, they were able to identify 4 main patterns of disease progression. In addition, they were also able to interpolate data from less saturated data sets. The most important result is that these patterns are reproducible in other studies and simply do not apply to just one population set. For future studies, scientists aim to use these patterns to further understand these diseases and identify any subtypes.
This study is very interesting as most researchers examine the disease first and then move onto patterns and treatments. However, this study takes the opposite approach. This research is also very important because for many diseases like ALS, doctors may be able to understand the onset of symptoms, however understanding the progression of the disease is equally, if not more important. Once the pattern of progression is identified, healthcare professionals can understand the disease better and will offer more treatment options if they have knowledge about the disease’s trajectory. All in all, this article was very insightful as it offered a different approach to research and healthcare using new technology such as machine learning.