TRENDING RESEARCH...

Motor Imagery EEG classification set to renovate existing Brain Computer Interface (BCI)

Motor Disability is a dreadful body impairment astonishing the living community. The supportive roots to be blamed are: The aging population and the growing number of diseases. So, what can be the stroke of luck for the people with heavy muscle impairments? The very big answer is “The Brain Computer Interface (BCI)”. The reason is attributable to the use of electroencephalogram (EEG) signals to control any device, without effecting the physically-challenged to move. Hence, the function of BCI in a nutshell is “classify the EEG signal into device actions, based on human thoughts”. A myriad of research on motor imagery EEG classification has constantly proliferated to accomplish this function. A combined group of three researchers from India and one from Ethiopia have attempted motor imagery EEG classification, via an optimization-enabled deep residual neural network.  Two datasets from Berlin Brain-Computer Interface have been deployed for testing their approach. Although the approach imparted an accuracy of 91.83% for the multi-class dataset, it could indeed mark the start of boundless research on competent motor imagery classification.

Green Supply Chain Management (GSCM) for sustainable and eco-friendly process control in industries

With the alarming rise in population and urbanization, the prevalence of our planet’s resources in the future is highly dubious. To bestow our descendants with an unremitting flow of resources and pollution-free environment, the optimistic approach is “Urbanize through sustainable and eco-friendly procedures”. Green Supply Chain Management (GSCM) can effectively support in this regard, as the excessive resource utilization in industries is cut down by resource reusage, reclamation and recycling of degradable materials. Nevertheless, the GSCM implementation in industries would be trivial, if various governing factors like, the supply of quality goods at minimum cost, the security as well as time-consumption of environmental processes and the waste disposal procedure are incongruously assessed. Knowing its significance, the researchers from Aditya college of engineering and Vignan's Institute of information technology, Andhra Pradesh, India have conducted a survey to unveil the factors needed for the successful GCSM implementation in the leather industries of Northern Tamil Nadu. 

Their analysis on the learned factors using Deep Belief Networks (DBN) can be a major breakthrough for GSCM implementation in industries. “The proposed framework's implementation has implications for increasing organizational efficiency, lowering costs, minimizing waste, and encouraging green culture among personnel. This study will aid in the development of policies and a better knowledge of how to implement green innovation methods'', the researchers say.  

Image credit: www.freepik.com 

Research on VANETs- An initiative to accomplish malicious-free ITS

In recent years, the transportation sector in economically-developing countries demand Intelligent Transportation Systems (ITS) to cope with the increasing traffic. Vehicular Adhoc Networks (VANET) of ITS tend to deliver a secure, reliable and fast traffic management, as it enables message transmission between vehicles and vehicles (V2V) or vehicles and infrastructure (V2I) along wireless communication modes. 

Unfortunately, there is a hindrance that deny the smooth functioning of VANETs - “The malicious intruders”. These intruders purely intend to disrupt the transmission, alter the transmitted contents as they seek and retransmit, questioning security with light to heavy losses. So, what might be the solution? It might be a mere classification of messages as belonging to intruder or not, considering features like, data rate, radio range, speed of the vehicle and many more. Though it looks simple, attack detection and secure routing in VANETs is more intricate because the user is unaware of the message source to perform the classification. Here comes the need of clustering the vehicles in VANETs, since clustering enables organized and secure routing across numerous vehicles with reduced transmission bandwidth. Considering the consequences, Gurjot Kaur and Deepti Kakkar, researchers from Dr. B.R. Ambedkar National Institute of Technology, Jalandhar have put forward an optimization-based Deep Maxout Network (DMN) for VANET-attack classification. The researchers say, “The limitation of this approach is that it failed to consider a data-driven intrusion detection approach to afford automatic security services”, paving way to future research in this network domain.

Accurate sarcoma diagnosis from MRI needs heed

Nowadays, it is not very uncommon that life-threatening diseases pop up and chill out, causing irreparable desolations to life on Earth. Sarcoma is a tumor-type disease, affecting bone and the primary connective tissues of the body. It is not as fatal as Carcinoma that affects internal organs, but its diagnosis is challenging. The American Cancer Society presets a figure of 13190 people to be diagnosed with Soft Tissue Sarcoma (STS) in 2022, estimating 5130 people to be dead. Better diagnosis leads to better sarcoma survival rates. So, which modality is the choice for diagnosis Computed Tomography (CT), (Magnetic Resonance Imaging) MRI or an X-ray? It is MRI that sounds good, since the soft tissues, its tumor and the tumor extent is clear cut than in other modalities. However, the MRI artifacts have led to abundant research on improved diagnosis in the past few years. The authors of “Optimized Convolutional Neural Network for Soft Tissue Sarcoma Diagnosis”, published in “Multimedia Tools and Applications”, have suggested an improved diagnosis of STS using the datasets from “The Cancer Imaging Archive”. The attained sensitivity was 0.85, which was approximately 23-29% better than the prevailing approaches. The authors hope to use deep learning methodologies for detecting the STS subtypes from various modalities and thus, makes an initiative to elude the lethal sarcoma.