Artificial intelligence to save the day? How smart computers help us understand HD. – HDBuzz

Through Dr Rachel Harding Edited by Dr Sarah Hernandez

Scientists have developed a new model that maps the different stages of HD (HD) in detail. Using artificial intelligence approaches, the researchers were able to filter information from large data sets collected during the observational studies contributed by patients with Huntington’s disease. A team of researchers from IBM and the CHDI Foundation have published a new model of HD progression in the journal Movement Disorders, which they hope will improve the way HD clinical trials are designed in the future.

HD is caused by an extension of the huntingtin gene that leads to the production of an expanded form of the huntingtin protein† Studies of lab models of HD and people who carry the HD gene show that having the enlarged gene and making the enlarged form of the protein creates a cascade of problems. Starting with small molecular changes, people with HD will eventually experience a range of different symptoms related to thinking, movement and mood that get worse over time.

Scientists are in the dark about how best to categorize the different stages of HD, but this new study using artificial intelligence hopes to shed some light on this issue.
Scientists are in the dark about how best to categorize the different stages of HD, but this new study using artificial intelligence hopes to shed some light on this issue.

Image credit: Ars Electronica / Robert Bauernhansl

HD symptoms usually appear between the ages of 30 and 50, but when this happens, a number of factors influence it. We’ve long known that people with larger extensions in their Huntingtin gene tend to get symptoms earlier, healthy lifestyle choices such as a balanced diet and regular exercise can delay the onset of symptoms, and other so-called genetic “modifiers” can affect it too. how early can the disease strike a gene carrier.

However, there’s still a lot we don’t understand about how HD progresses over time and how symptoms worsen. To tackle this problem, scientists from around the world have developed numerous observational trials and natural history studies that track the patient’s symptoms, biomarkers, and other measurements over time. These include PREDICT-HD, REGISTRY, TRACK-HD, and Enroll-HD. Together, these studies have resulted in very large datasets comprising more than 2000 different measurements from 25,000 participants. This is tons of really useful data, all made possible by the dedication of HD families to participate in these studies.

By examining all these data sets at once, scientists can discover new patterns and draw new conclusions, but doing this kind of analysis manually is extremely laborious and challenging. This is where the smart computer scientists come in! Scientists can use cool new methods to make the computers look at all the data at the same time using special types of programs often referred to as artificial intelligence or AI.

Together, these studies have resulted in very large datasets comprising more than 2000 different measurements from 25,000 participants. This is tons of really useful data, all made possible by the dedication of HD families to participate in these studies.

A commonly used AI approach is called machine learning. This type of AI software gets better at making predictions of certain outcomes by building models based on training datasets that it uses to “learn” without being explicitly programmed to do so. Machine learning is a field of its own in biomedical research, but it also has many different applications for things like email filtering and speech recognition.

IBM and CHDI researchers used machine learning approaches to build and test a new model to understand how HD progresses and categorize different stages of disease. The model was then tested against a number of different measures commonly collected and collected in HD research that monitors disease progression, including the Unified Huntington’s Disease Rating Scale (UHDRStotal functional capacityTFC), and the CAG age product, also known as the CAP score.

The new model defines 9 states of HD, all specified by different measures that assess movement, thinking and daily functioning. These states range from the early stages of the disease before the motor symptoms begin, all the way to the late stages of disease with the most severe symptoms. The model was able to predict how likely participants in the studies were to transition between states and how long participants would spend in the different stages of HD. While other studies have found that the entire course of disease takes place over a period of about 40 years, this is the first time researchers have predicted the expected amount of time HD patients will spend in each of the 9 states described in the new model. .

Artificial intelligence is used in many different ways to solve problems in fields such as medicine, business, communication and transportation.
Artificial intelligence is used in many different ways to solve problems in fields such as medicine, business, communication and transportation.

Image credit: Image via www.vpnsrus.com

Having this handy new 9-state model of HD progression could help scientists and clinicians learn more about the different stages of HD and the time frames it takes people with HD to move from one state to another. to go. With this information in hand, the researchers at IBM and CHDI believe it could help select the most appropriate participants for certain HD clinical trials, identify robust biomarkers for tracking disease progression, and also help design better clinical trials.

This is an exciting step forward for HD research and we look forward to learning more about other AI applications in HD research as new approaches are designed and this exciting field of science matures.

Leave a Comment

Your email address will not be published.