Mushfiqul Islam - Academia.edu (original) (raw)

Uploads

Papers by Mushfiqul Islam

Mathematical models of cancer growth have been subject of research for many years. At least six d... more Mathematical models of cancer growth have been subject of research for many years. At least six different mathematical models and their countless variations and combinations have been published till date in scientific literature that reasonably explains epidemiological prediction of multi-step carcinogenesis. Each one deals with a particular set of problems at a given organizational level ranging from populations to genes. None of the articles have incorporated all the types of cancers. Any of the models adopted in those articles so far does not account for both epidemiological and molecular levels of carcinogenesis. In other words, those models are used in ‘specialized’ ways to focus on specific attributes of cancer.Therefore, our work aims at the derivation of a mathematical model consisting of fewer than five equations that reasonably explains epidemiological prediction of multi-step carcinogenesis. We have come up with a mathematically rigorous system to derive those equations t...

2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

In this work we propose 3D-FFS, a novel approach to make sensor fusion based 3D object detection ... more In this work we propose 3D-FFS, a novel approach to make sensor fusion based 3D object detection networks significantly faster using a class of computationally inexpensive heuristics. Existing sensor fusion based networks generate 3D region proposals by leveraging inferences from 2D object detectors. However, as images have no depth information, these networks rely on extracting semantic features of points from the entire scene to locate the object. By leveraging aggregated intrinsic properties (e.g. point density) of point cloud data, 3D-FFS can substantially constrain the 3D search space and thereby significantly reduce training time, inference time and memory consumption without sacrificing accuracy. To demonstrate the efficacy of 3D-FFS, we have integrated it with Frustum ConvNet (F-ConvNet), a prominent sensor fusion based 3D object detection model. We assess the performance of 3D-FFS on the KITTI dataset. Compared to F-ConvNet, we achieve improvements in training and inference times by up to 62.80% and 58.96%, respectively, while reducing the memory usage by up to 58.53%. Additionally, we achieve 0.36%, 0.59% and 2.19% improvements in accuracy for the Car, Pedestrian and Cyclist classes, respectively. 3D-FFS shows a lot of promise in domains with limited computing power, such as autonomous vehicles, drones and robotics where LiDAR-Camera based sensor fusion perception systems are widely used.

2019 8th International Conference System Modeling and Advancement in Research Trends (SMART), 2019

In Bangladesh huge amount of agricultural products are destroying by the pests every year due to ... more In Bangladesh huge amount of agricultural products are destroying by the pests every year due to lack of poor knowledge about pest detection. As we know that manually identification is difficult for a farmer. So, classic pest detection and identification can ensure excellent productivity. This would be a fulfil research in the technical area of computer vision. The dataset is typically random cropping of square size images together with grayscale color and brightness shifts are used here. Here Convolutional Neural Network (CNN) will be used to do the image recognition and the algorithm will provide an optimal architecture for image recognition. The big idea behind CNNs is that a local understanding of an image is good enough. The research contains the proportions of validation accuracy of 93.46%. This approach resulted in the agriculture sector that will help a farmer to recognize the insect from harvest. The computer vision and object recognition can be used with image processing to create an interactive and enlarge user experience of the real world. This research aims to demonstrate the possibility and test the performance of the project which only focuses on insect detection in crop plants that recognize the pest which can help a farmer to get immediate solution of harvest problem.

A candidate gene approach was used to identify single nucleotide polymorphisms (SNPs) and their a... more A candidate gene approach was used to identify single nucleotide polymorphisms (SNPs) and their associations with body fat deposition and carcass merit traits in beef cattle. In total, 37 SNPs from 9 candidate genes have been genotyped on 463 hybrid, 206 Angus and 187 Charolais steers for association analyses with 10 different fat deposition and carcass merit traits. In single SNP analyses, 28 SNPs of 9 genes have been found significantly (P<0.05) associated with different traits in the cattle populations. Gene-specific linkage disequilibrium assessment of SNPs revealed the existence of haplotype blocks within 4 genes. Haplotype analyses have identified 31 haplotypes of 6 genes having significant associations (P<0.05) with different fat deposition and carcass merit traits in the cattle populations. These findings will provide insight into the genetic mechanism regulating body fat deposition in beef cattle and will assist the beef industry to improve beef quality through marker assisted selection.

2018 4th International Conference on Electrical Engineering and Information & Communication Technology (iCEEiCT), 2018

This paper proposes a method to detect road surface using intensity histogram. The key idea is to... more This paper proposes a method to detect road surface using intensity histogram. The key idea is to determine the intensity range of road surface for three different channels. Then this range of intensity is used to find the surface in an image. The testing image is splitted into RGB channel. If the pixel's intensity is in determined range then it is replaced by zero, otherwise by one. So, a binary image is created for every channel. After some processing these images are merged. The merged image containing large number of zeros is regarded as road surface. This method is computationally efficient and easy to implement. The overall system has approximately 78.6% accuracy with 65% precision which made this applicable for use.

American Journal of Biochemistry and Biotechnology, 2015

Clinically significant 18 Single Nucleotide Polymorphisms (SNPs) from exon regions of Retinoblast... more Clinically significant 18 Single Nucleotide Polymorphisms (SNPs) from exon regions of Retinoblastoma gene (RB1) were analyzed to find out the structural variations in mRNAs. Online bioinformatic tools i.e., Vienna RNA, RNAfold were used for secondary structure analysis of mRNAs. Predicted minimum Free Energy Change (MFE) was calculated for mRNAs structures. It has been observed that the average of predicted MFE value from 13 nonsense mutations was higher (0.76 kcal/mol) in comparison to 5 missense mutations. Presumably, 13 nonsense mutations are responsible for Nonsense Mediated mRNA Decay (NMD), therefore, excluded from haplotype analysis. From the statistical analysis all the thermodynamic data obtained from four SNP haplotypes are significant (p≤0.05), followed by three-SNP haplotype data except Ensemble diversity (p≤0.10). Interestingly, MEF of Centroid Secondary Structure is highly significant (p≤0.01) in all the cases (Two-SNP haplotypes, Three-SNP haplotypes and Four-SNP haplotypes).

Biomedical Research and Therapy, 2014

We have reviewed the COP1 mediated tumor suppressor protein p53 pathway and its oncogenic role. C... more We have reviewed the COP1 mediated tumor suppressor protein p53 pathway and its oncogenic role. COP1 is a negative regulator of p53 and acts as a pivotal controller of p53-Akt death-live switch (Protein kinase B). In presence of p53, COP1 is overexpressed in breast, ovarian, gastric cancers, even without MDM2 (Mouse double minute-2) amplification. Following DNA damage, COP1 is phosphorylated instantly by ATM (Ataxia telangiectasia mutated) and degraded by 14-3-3σ following nuclear export and enhancing ubiquitination. In ATM lacking cell, other kinases, i.e. ATR (ataxia telangiectasia and Rad3-related protein), Jun kinases and DNA-PK (DNA-dependent protein kinase) cause COP1 & CSN3 (COP9 signalosome complex subunit-3) phosphorylation and initiate COP1's down regulation. Although, it has been previously found that co-knockout of MDM2 and COP1 enhance p53's half-life by eight fold, the reason is still unknown. Additionally, while interacting with p53, COP1 upregulate MDM2's E3 ubiquitin ligase, Akt, CSN6 (COP9 signalosome 6) activity and inhibit 14-3-3σ's negative regulation on MDM2 and COP1 itself. Conclusively, there persists an amplification loop among COP1, MDM2, Akt and 14-3-3σ to regulate p53's stability and activity. However, the role of another tumor suppressor PTEN (phosphatase and tensin homologue) is yet to be discovered. This study provides insight on the molecular genetic pathways related to cancer and might be helpful for therapeutic inventions.