Research Assistant at Computer Vision and Multimedia Laboratory,
Masters Student in Computer Science,
Illinois Institute of Technology, Chicago
I, Nikhil Sharma, currently enrolled in the Computer Science department at the Illinois Institute of Technology in Chicago, is an accomplished graduate student. I'm working with Dr. Yan Yan in his Computer Vision and Multimedia Laboratory on topic such as multi-modality based deep learning frameworks, diffusion models based projects while in collaboration with teams from universities such as Stanford, Michigan State, UT Austin.
Recently worked as an Artificial Intellegnce and Computer Vision Research Engineer at Matdun Labs India Private Limited, where I contibuted in developing and desgining various automated modules for the product "The Nock".
Apart from working in a commerical industy, during my academic journey, he engaged in multiple research endeavors at esteemed Indian institutions including the Indian Institute of Technology in Delhi and the National Institute of Technical Teachers' Training & Research where he has worked on applications of AI in various sector such as diagnosing Cancers and SARS-CoV-2.
My academic background culminated in a B.Tech degree in Electronics and Communication Engineering from the Jaypee Institute of Information Technology. Nikhil Sharma's contributions extend beyond his education; he has co-authored approximately 17 research papers featured in various international journals and conferences, showcasing his dedication to advancing knowledge in his field.
Moreover, I am actively engaged as a reviewer for over 6 prominent international journals, underscoring his commitment to the scholarly community. His research pursuits span a wide spectrum, encompassing Artificial Intelligence, Computer Vision, Computational Intelligence, Computational Biology, Machine Learning, and Pattern Recognition. In addition to this he has also delivered valauble knowlege as a guest lecturer in Mahindra University, India.
A comprehensive genomic analysis was conducted on 566 SARS-CoV-2 genomes originating from India. This investigation unveiled genetic mutations spanning substitutions, deletions, insertions, and Single Nucleotide Polymorphisms (SNPs). The viral genomes were procured from the GISAID database. Within the spectrum of genetic alterations, 57 out of 64 SNPs were pinpointed within the coding regions of the SARS-CoV-2 genomes. The application of phylogenetic analysis accentuates the notable diversity existing among distinct strains of the virus.
The aim was to define the initial T1 values of tumors, quantify alterations following the chemotherapy regimen, and assess its viability as an indicator for evaluating treatment response in individuals with osteosarcoma. As a result of the study it was noticed that The mean and skewness of T1 values in osteosarcoma are potential non-invasive imaging markers for chemotherapy response assessment.
In this study, we conducted a comprehensive analysis of the SARS-CoV-2 genome to uncover its genetic diversity and ensure accurate diagnostic detection. Our approach involved analyzing 10,664 complete or nearly complete SARS-CoV-2 genomes from 73 countries, globally and at the country level. Utilizing multiple sequence alignment and a reference sequence from NCBI, we identified mutation points as substitutions, deletions, insertions, and single nucleotide polymorphisms (SNPs) on a global scale, yielding 7,209 substitutions, 11,700 deletions, 119 insertions, and 53 SNPs..... (read more)
Incorporating CRISPR-Cas technology for precise target site localization. Employing the k-mer approach to discern Protospacer Adjacent Motifs (PAMs). Utilizing identified PAMs to locate corresponding palindromic sequences within reference virus sequences. Creating primers that complement these palindromes for accurate virus identification. Evaluating the population coverage of palindrome-PAM combinations across virus sequences.
In this study 38 potential SARS-CoV-2 regions are identified consisting of 249 human miRNAs. Among these 38 potential regions, some top regions belong to nucleocapsid, RdRp, helicase, and ORF8. To understand the biological significance of these potential regions, the targets of the human miRNAs are considered for KEGG pathways and protein–protein and drug–protein interaction analysis as the human miRNAs are similar to the potential regions of SARS-CoV-2.
A novel U-Net architecture based on deep learning principles is introduced as a framework for automated segmentation of various suspicious regions within CT scans of Covid-19 patients. Leveraging the benefits of Dense Residual Connections, the model captures comprehensive global hierarchical features across all convolutional layers. This approach strikes an improved balance between efficiency and effectiveness within the U-Net framework... (read more)