FDG-PET model showed moderate accuracy for Alzheimer’s disease-related TDP-43 pathology
Original source
FDG-PET imaging to identify brain regions associated with Alzheimer's disease-related TDP-43 proteinopathy: A predictive model using penalized logistic regression analysis. (opens in a new tab)Compass summarised this from the study's abstract.
Study details
- Studied in
- Human
Related topics
A study used fluorodeoxyglucose positron emission tomography (FDG-PET) scans and a statistical model to predict brain regions associated with Alzheimer’s disease-related TDP-43 pathology. The model achieved 68% accuracy in 294 participants and 65% accuracy in a 159-person subgroup.
Why this matters
TDP-43 pathology is relevant to neurodegenerative disease research, but this study examined Alzheimer’s disease, not amyotrophic lateral sclerosis or motor neurone disease. Its findings do not establish a diagnostic test for ALS/MND or change treatment for people living with those diseases.
Limitations and context
This was a predictive modelling study using FDG-PET data, with participants divided into training and testing sets. Accuracy was moderate, and the study assessed Alzheimer’s disease-related pathology rather than ALS/MND. The findings would need independent validation and testing in ALS/MND populations before any clinical use could be considered.