Biology subjects

Keihani, A.

Publications and source records attributed to Keihani, A..

3 recordsLinked to original sources

bioRxiv2025

Recognizing EEG responses to active TMS vs. sham stimulations in different TMS-EEG datasets: a machine learning approach

BackgroundTranscranial Magnetic Stimulation (TMS) with simultaneous Electroencephalogram (TMS-EEG) allows assessing the neurophysiological properties of cortical neurons. However, TMS-evoked EEG potentials (TEPs) can be affected by components unrelated to TMS direct neuronal activation. Accurate, automatic tools are therefore needed to establish the quality of TEPs. ObjectiveTo assess the discriminability of EEG responses to TMS vs. EEG responses to sham stimulations using sequence-to-sequence machine learning (ML). MethodsTwo indipendent TMS-EEG datasets including TMS and several sham stimulation conditions were obtained from the left motor area of healthy volunteers (N=33 across datasets). A Bi-directional Long Short-Term Memory (BiLSTM) ML network was used to label each time point of the EEG signals as pertaining to TMS or sham conditions. Main outcome measures included accuracy at single-trial level and after averaging five to twenty trials. ResultsFor TMS conditions, post-stimulus vs. baseline/pre-stimulus EEG comparisons yielded moderate (60%-75%) single-trial accuracy and high-accuracy (>75%) for 20 trials across datasets, while for sham conditions post- vs. baseline/pre-stimulus EEG comparisons yielded lower accuracy rates than for TMS conditions, except for unmasked auditory stimulation. Furthermore, baseline/pre-stimulus TMS vs. baseline/pre-stimulus sham EEG comparisons showed chance-level accuracy, whereas post-stimulus TMS vs. post-stimulus sham EEG comparisons had moderate (single trial) to high (20 trial) accuracy, except for TMS with and without the click noise masking. Single-subject findings were comparable to group-level results across datasets. ConclusionsTEPs after active TMS are discernible from various sham stimulations even after a handful of trials and at the single-subject level using a BiLSTM ML approach.

neuroscience↗

bioRxiv2024

Targeting complement C3a receptor resolves mitochondrial hyperfusion and subretinal microglial activation in progranulin-deficient frontotemporal dementia

Mutations in progranulin (GRN) cause frontotemporal dementia (GRN-FTD) due to deficiency of the pleiotropic protein progranulin. GRN-FTD exhibits diverse pathologies including lysosome dysfunction, lipofuscinosis, microgliosis, and neuroinflammation. Yet, how progranulin loss causes disease remains unresolved. Here, we report that non-invasive retinal imaging of GRN-FTD patients revealed deficits in photoreceptors and the retinal pigment epithelium (RPE) that correlate with cognitive decline. Likewise, Grn-/- mice exhibit early RPE dysfunction, microglial activation, and subsequent photoreceptor loss. Super-resolution live imaging and transcriptomic analyses identified RPE mitochondria as an early driver of retinal dysfunction. Loss of mitochondrial fission protein 1 (MTFP1) in Grn-/- RPE causes mitochondrial hyperfusion and bioenergetic defects, leading to NF-kB-mediated activation of complement C3a-C3a receptor signaling, which drives further mitochondrial hyperfusion and retinal inflammation. C3aR antagonism restores RPE mitochondrial integrity and limits subretinal microglial activation. Our study identifies a previously unrecognized mechanism by which progranulin modulates mitochondrial integrity and complement-mediated neuroinflammation.

neuroscience↗

bioRxiv2019

Introducing Rhythmic Sinusoidal Amplitude-Modulated Auditory Stimuli with Multiple Message Frequency Coding for Fatigue Reduction in Normal Subjects: An EEG Study

Many of the brain-computer interface (BCI) systems depend on the users voluntary eye movements. However, voluntary eye movement is impaired in people with some neurological disorders. Since their auditory system is intact, auditory paradigms are getting more patronage from researchers. However, lack of appropriate signal-to-noise ratio in auditory BCI necessitates using long signal processing windows to achieve acceptable classification accuracy at the expense of losing information transfer rate. Because users eagerly listen to their interesting stimuli, the corresponding classification accuracy can be enhanced without lengthening of the signal processing windows. In this study, six sinusoidal amplitude-modulated auditory stimuli with multiple message frequency coding have been proposed to evaluate two hypotheses: 1) these novel stimuli provide high classification accuracies (greater than 70%), 2) the novel rhythmic stimuli set reduces the subjects fatigue compared to its simple counterpart. We recorded EEG from nineteen normal subjects (twelve female). Five-fold cross-validated naive Bayes classifier classified EEG signals with respect to power spectral density at message frequencies, Pearsons correlation coefficient between the responses and stimuli envelopes, canonical correlation coefficient between the responses and stimuli envelopes. Our results show that each stimuli set elicited highly discriminative responses according to all the features. Moreover, compared to the simple stimuli set, listening to the rhythmic stimuli set caused significantly lower subjects fatigue. Thus, it is worthwhile to test these novel stimuli in a BCI experiment to enhance the number of commands and reduce the subjects fatigue. Significance StatementAuditory BCI users eagerly listen to the stimuli they are interested in. Thus, response classification accuracy may be enhanced without the need for trial lengthening. Since humans enjoy listening to rhythmic sounds, this study was carried out for introducing novel rhythmic sinusoidal amplitude-modulated auditory stimuli with multiple message frequency coding. Our results show that each stimuli set evoked reliably discriminative responses according to all the features, and rhythmic stimuli set caused significantly lower fatigue in subjects. Thus, it is worthwhile to test these novel stimuli in a BCI study to increase the number of commands (by NN permutations of just N message frequencies) and reduce the subjects fatigue.

neuroscience↗