Speech Analysis in Progressive Aphasia and Related Disorders
Author Information
Author(s): Henderson Shalom K, Ramanan Siddharth, Patterson Karalyn E, Garrard Peter, Patel Nikil, Peterson Katie A, Halai Ajay, Cappa Stefano F, Rowe James B, Lambon Ralph Matthew A
Primary Institution: Medical Research Council Cognition and Brain Sciences Unit, University of Cambridge
Hypothesis
Can a simple checklist of words improve the diagnosis of progressive aphasia and related disorders?
Conclusion
A 15-word checklist can effectively differentiate between patients with progressive aphasia and healthy controls.
Supporting Evidence
- The total language output was significantly reduced in patients with nfvPPA, PSP, and CBS compared to svPPA and controls.
- Patients with svPPA and lvPPA used a greater proportion of words that were more frequent and semantically diverse.
- The 15-word checklist showed excellent accuracy for within-sample k-fold validation.
- Out-of-sample test accuracy was 91% for patients versus controls and 74% for differentiating patient groups.
- Supplementing the checklist with cognitive measures improved diagnostic accuracy for certain patient groups.
Takeaway
Doctors can use a short list of words to help tell if someone has a speech problem related to brain issues.
Methodology
Connected speech samples were analyzed from patients with different types of aphasia and related disorders using a picture description task.
Potential Biases
Potential biases in participant selection and the reliance on clinical diagnoses.
Limitations
The study only presents clinical diagnoses, not pathological ones, and the sample size for some groups was small.
Participant Demographics
Participants included 24 healthy controls and patients with various types of primary progressive aphasia and related disorders, primarily white.
Statistical Information
P-Value
p<0.001
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website