Research Project Conducted by ACSLab Members
Decoding Brain Network Dysfunction in Alzheimer’s Disease/ Mild Cognitive Impairment
Alzheimer’s disease and mild cognitive impairment are characterized by altered brain network activity due to the pathology. Most of the existing detection techniques can detect the pathology at a mature stage rather than starting stage. Therefore, a stable biomarker of the pathology detected at an early stage would greatly facilitate the early diagnosis of the disease and, thus, its treatment. To the aim, we use in vivo recordings of local field potentials from genetically modified animal models at different pathological stages, and apply frequency based sophisticated analysis techniques to detect network dysfunctions. We have been examining the possibility of using network activity as a biomarker for early diagnosis of the pathology.
Deep Learning for Biological Data Analysis
Recent research in Deep learning, Reinforcement learning, and their combination promise to revolutionize Artificial Intelligence. And, rapid advancement in the hardware based technologies over past decades opened up new possibilities for Biological and Life scientists to gather multimodal data from various application domains (e.g., Omics, Bioimaging, Medical Imaging, and [Brain/Body]-Machine Interfaces). Novel data intensive machine learning techniques are required to decipher these data. Increasing computational power, faster data storage devices, and declining computing costs allowed scientists to apply these techniques on such enormous and complex datasets which otherwise would not have been possible. We therefore, have been applying DL based methods to analyze data coming from different biological sources.
Distributed Bio-Signal Analysis
To keep pace with the rapid growth of neuronal probe technology scientists require tools capable of helping them in inferring meaningful conclusions from the mountain of data. With time and increasing capability of the signal acquisition systems performing the required analyses are becoming difficult day by day for a normal desktop computer. That is why we are exploiting the power of distributed systems to perform the job.
Neuronal Simulation for Stimulus Optimization
I work on neuronal simulations in terms of stimulation optimization. Neuronal stimuli are widely adopted to evoke brain responses. Specially in case of certain diseases, e.g., parkinsons, deep brain stimulation is a widely accepted therapy. However, the protocol for stimulation is heuristic and requires periodic tuning after a certain time.
The goal of the stimulus optimization is to determine the optimized stimuli required to evoke neuronal response taking into account the existing stimulation parameters.
Neuronal Simulation for Stimulus Optimization
Motor movement is what keeps us able. Some diseases or trauma may disrupt our motor movement making us paralyzed / disabled. To design appropriate therapy and realistic assistive devices (e.g., artificial neuromuscular junction) we need to understand the mechanism of the transmission process. We have developed a realistic model for neuromusclular junction mimicking the neurobiological phenomena taking place at the junction during the synaptic transmission process.
Extracellular Neuronal Signal Processing and Analysis Toolboxes
QSpike Tools: The Queued Spike Tools (QSpike Tools) is an effective web-based, client-server, software workflow for parallel processing and analysis of extracellular multi-unit activity, acquired by a standard commercial Multi Electrode Array (MEA) platform.
This extensible workflow may be utilized in analyzing any extracellular neural signals.
SigMate: is a Matlab based comprehensive toolbox developed for processing and analyzing of extracellular neuronal signals.
A three-layered architecture was adhered during its development and currently includes:
o Signal visualization
o File operations
o Stimulus artifact removal
o Noise characterization
o Spike detection, sorting and spike train analysis
o Current source density calculator
o Latency estimator
o Contour based single local field potential classifier
o EEG analysis
Project Title: Design and Implementation of Flood Warning System for Bangladesh Using WSN, Financed by Jahangirnagar University,,July 2015 to June 2016).
Project Title: Design and Implementation of Deserter Warning System for Bangladesh Using WSN, Financed by Ministry of Science and Technology, Government of the People’s Republic of Bangladesh, (BDT 200K), July 2014 to June 2015
Project Title: Design and Implementation of Grid Connected Solar Water Pumping System for Bangladesh. Financed by Ministry of Education, Government of the People’s Republic of Bangladesh (BDT 1800K), July 2013 to June 2015
Project Title: Design and Implementation of Solar Power Boat, Financed by Jahangirnagar University, Dhaka, Bangladesh, (BDT 35K), July 2013 to June 2014
Project Title: Solar Powered Wheel-chair for Physically Challenged People, Funded by Jahangirnagar University, (BDT 35K), June 2013 to July 2014
Project Title: Design and Implementation of Solar Power Tricycle Driving Circuit, Funding:Ministry of Science and Technology (BDT 70K), July 2012 to June 2013
Project Title: Financial Feasibility and Performance Evaluation of Solar Powered Rickshaw, Funded by University Grant Commission (UGC) of Bangladesh, (BDT 150K), June 2013 to July 2014