Establishing conclusive cortical activity dynamics from neuroimages and high dimensional neuronal data post-processing, such as scalp-EEG/ERP or its localized source data, is always challenging. For addressing this, we introduce LEDA, localized-EEG dynamics analyzer, offering our novel techniques, namely (1)the localized source activity to duration (LSAD) ratio that elegantly combines voxel activation with temporal information enabling a spatiotemporal condition comparison, (2)cortical region’s activation-duration summary generation, (3)anatomical dominance calculator, and (4)voxel activations’ GIF animation extraction feature. They effectively explore and quantify physiological spatiotemporal dynamics and provide insights more intuitive than EEG/ERP biomarkers. To demonstrate, we chose to study sensory gating deficits in schizophrenia for which the P50 wave biomarker is widely discussed. EEG response (64 channels@1 kHz) to auditory paired-click paradigm from 9 patients and 9 controls was processed to ERP. Source localization was achieved by using sLORETA, a widely-used source reconstruction algorithm. LSAD ratio of patients significantly differed from that of controls for the auditory cortical regions (p value < 0.05). It revealed that inhibitory alpha activity pattern was prominent in parietal, temporal and occipital lobe for controls and not patients. Conversely, the frontal lobe, specifically the middle frontal gyrus, was temporally the most active analogously across all subjects. LEDA also extracted that swift alpha activity exists between 100 and 300 ms time period. These derived results match with existing fMRI studies, speak beyond P50 wave suppression biomarker and emphasize that our methods successfully uncovered meaningful brain mechanisms at high spatiotemporal resolution. They are provided as an open-source MATLAB based toolbox for applicability to other neuronal pathological investigations.
Read full abstract