At first glance, medicine and astrophysics have no obvious links. However, in several situations, the technologies developed for the second have benefited the first. For example, breast cancer detection algorithms have been improved thanks to the photonic algorithms used by certain telescopes. And recently, the two fields have again joined forces, this time in the field of Alzheimer's disease.
A deep learning algorithm (or deep learning) statistics used in astrophysics to combine and analyze interactions between data collected by telescopes, was able to do the same work from patient records to diagnose Alzheimer's disease early with greater efficiency than from a general practitioner (GP).
The crucial early diagnosis of dementia in general practice could therefore improve thanks to this computer model, designed in collaboration between the University of Brighton, Sussex Medical School (BSMS) and astrophysicists at the University of Sussex. Currently, only two-thirds of people with dementia in the UK receive a formal diagnosis, and many receive it late in the disease process.
The team, led by Elizabeth Ford, used data from general practitioner (GP) records to create a list of 70 indicators linked to the onset of dementia, recorded over the five years prior to diagnosis.
Working with astrophysicists, they then tried several types of machine learning models to identify patterns of clinical information in patient records before a dementia diagnosis. The best model was able to identify 70% of dementia cases before MG, but also generated a number of false positives.
Ford explains, “Patients appear to exhibit a wide range of indicators before being diagnosed with dementia. It can be very difficult for GPs to connect all these indicators and make the link to dementia, but with a computer program we can potentially do it, and more efficiently. Early diagnosis could make a significant difference in the care dementia patients receive .
Related:How can astronomy help fight cancer?
“These findings are exciting, but they prompt the need for discussion with GPs and patients about the place this type of technology should have in the clinic. As technology develops, we need to have broader conversations about whether we're happy with computers that assess our risk for these kinds of life-changing diseases, like dementia adds Ford.
Astrophysicist Seb Oliver says:“It has been fantastic working on this project with Ford and his team. It's always amazing how statistical methods like AI and deep learning can be used to extract useful insights from data, whether it's images from space telescopes or patient records. Of course, statistics are only part of the understanding and it is really exciting to work in new areas to try to understand the different challenges these present. .
