New EEG Analysis Technique for Alzheimer's and Mild Cognitive Impairment
Author Information
Author(s): Massimo Buscema, Massimiliano Capriotti, Francesca Bergami, Claudio Babiloni, Paolo Rossini, Enzo Grossi
Primary Institution: Semeion Research Centre of Sciences of Communication
Hypothesis
The new artificial organism TWIST could produce a better classification between Alzheimer's disease and mild cognitive impairment compared to previous methods.
Conclusion
The study confirms that a correct automatic classification of MCI and AD subjects can be achieved by extracting spatial information from EEG data using ANNs.
Supporting Evidence
- The new method achieved a classification accuracy of 94.10%, which is better than previous studies.
- The study included a large sample size of 295 participants.
- Results suggest that EEG-derived spatial features can help in diagnosing Alzheimer's disease.
Takeaway
Scientists found a new way to use brain wave data to tell the difference between people with mild memory problems and those with Alzheimer's disease, and it works really well.
Methodology
EEG data were recorded from 180 AD patients and 115 MCI subjects, processed using a novel ANN methodology to classify the subjects.
Limitations
The study's classification accuracy of 94.10% still leaves a 6% misclassification rate, which is insufficient for diagnostic use.
Participant Demographics
180 AD patients (50 males, 130 females; mean age 77) and 115 MCI subjects (49 males, 66 females; mean age 76).
Statistical Information
P-Value
p<0.05
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website