Dharmendra Mishra
Development
Delhi, India
Skills
Data Science
About
Dharmendra Mishra's skills align with IT R&D Professionals (Information and Communication Technology). Dharmendra also has skills associated with Childcare Employees (Healthcare). Dharmendra Mishra appears to be an entry-level candidate, with 23 months of experience.
View more
Work Experience
Project Engineer - Data Scientist
CDAC Delhi
May 2022 - Present
- Working on creating a SSVEP based speller for children with special needs in collaboration with AIIMS Delhi. Processing real-time stream data from EEG device using Canonical correlation analysis(CCA) to identify the class of stimuli the subject is focussing on based on its flashing frequency. The brain generates distinct eeg waves for stimuli flashing with distinct frequencies. In speller, different visual stimuli are made to flicker at different frequencies. After a significant number of trials i.e the subject focussing on a specific stimuli flashing for about thirty seconds the eeg buffer data is sent for preprocessing i.e filtration and baseline correction. The SSVEP responses in the EEG signal are expected to have peaks at the frequencies of the presented stimuli. After preprocessing the signal, canonical-correlation is applied to identify these peaks and determine the frequencies associated with the attended stimuli. Tools: Python, Scipy, MNE & Psychopy. Developing a solution for P300-based guilt detection. P300 is associated with memory processes and is thought to reflect the allocation of attention to stimuli with high cognitive relevance. Guilt detection assumes that a person familiar with a crime will exhibit a different P300 response when exposed to crime-relevant details compared to an innocent person. Subjects are presented with a series of stimuli, such as words, images, or other sensory inputs. Among these stimuli, there are occasional deviant stimuli that are of particular interest in the context of the investigation (e.g., details related to a crime). EEG (Electroencephalogram) signals are recorded during the task. The recorded signals are analyzed to identify the P300 component, particularly its amplitude and latency, in response to different stimuli and response conditions. Tools: Python, Matplotlib.
Education
CDAC
Post Graduate Diploma in Artificial Intelligence
January 2021 - January 2022
