Back

Countering Cross-Individual Variance in Event Related Potentials with Functional Profiling

Alsufyani, A.; Zoumpoulaki, A.; Filetti, M.; Janssen, D. P.; Bowman, H.

2020-04-12 bioengineering
10.1101/455030 bioRxiv
Show abstract

In this study, a simple method called the Weight Template (WT) is proposed for classifying brain responses of individuals into deceiving and non-deceiving. The performance of our method was evaluated on a number of identity deception data sets and on artificial EEG data sets. A comparison was made with a standard method used to measure P3 presence, called Peak-to-Peak. In the real experimental data, the WT showed higher performance in terms of sensitivity and specificity. In the artificial EEG data, in ERPs with low Signal-to-Noise Ratio (SNR), the WT was more resistant to noise and provided more accurate measures. HighlightsO_LIWe propose a method for classifying ERPs of the P3 Concealed Information Tests. C_LIO_LIThe new method is based on computing a weighted template (WT) for each individual. C_LIO_LIThe WT is used to characterize the shape of each individuals P3. C_LIO_LIThe performance of the WT was compared with the standard Peak-to-Peak method. C_LIO_LIThe WT showed higher performance in terms of sensitivity and specificity. C_LI

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.