For many years now proponents of AI have described a future in which all neurodegenerative diseases are a thing of the past. Alzheimer’s is often the one most talked about. The disease has no cure as of this time and causes a devastating loss of memory, confusion, and all around cognitive decline. Hundreds of clinical trials have been conducted since 2002 to find a cure. So far efforts have been unsuccessful.

The disease has proven to be brutal not just for sufferers but for their loved ones as well. IBM Australia has recently published a paper talking about how machine learning can be used to predict the severity of the disease and slow down its progression. IBM believes the biggest chance they have is to detect the disease earlier when there is still a chance to slow down its progression.


The basic idea of what these amalyoid-beta plaques do to your otherwise healthy neurons.


Decades of research has suggested that a biological marker associated with the disease changes long before any cognitive impairment is experienced. This marker is a peptide called amalyoid-beta. Examining the concentration of this peptide in an individuals spinal fluid provides an indication of risk before any issues occur. However, accessing a patients spinal fluid is highly invasive and just not practical. Over the past few years there has been a strong effort across the board to develop a less invasive test.


The paper published by IBM Australia claims to have used machine learning to identify a set of proteins in blood that can predict the concentration of amlyoid-beta in spinal fluid. They have created models that they believe could help clinicians predict the risk with an accuracy of up to 77 percent. The test is still in its early phases but there is a lot of potential to improve the selection of individuals for drug trials. Individuals with mild cognitive impairment who were predicted to have an abnormal concentration of amyloid-beta in their spinal fluid were found to be 2.5 times more likely to develop Alzheimer’s in this study.

If the patients in these drug trials demonstrate an increases risk for alzheimer’s decades into the future it would allow for more accurate tests on prevention of the disease to be conducted.

Shortly after this paper was published Ben Goudey, Staff Researcher of the Genomics Research Team at IBM wrote these comments in a short summary of the findings:

“Neurodegenerative diseases such as Parkinson’s, Alzheimer’s and Huntington’s are affecting millions of people around the world. While these mysterious and crippling diseases do not yet have a cure, the answer to slowing their growth may lie in prevention. At IBM Research, our mission is to use AI and technology to understand how to help clinicians better detect and ultimately prevent these diseases in their early stages.”

The full paper is published in the science journal Nature here.



Alzheimer’s mainly affects those 65 and older.


Machine Learning in general has its skeptics but no matter how you look at it we are making great strides. Machine learning will probably never lead to conscious AI but it can help prevent billions of people from developing one of these crippling diseases. At some point even a bullet proof cure will be developed utilizing this technology. It is only a matter of time.

Leave a Reply

Your email address will not be published. Required fields are marked *