About
30
Publications
21,205
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
676
Citations
Introduction
Saqib Qamar is working as a Postdoc in Robotics, Perception, and Learning (RPL) at KTH Royal Institute of Technology in Sweden. Before joining KTH, he worked as a Postdoctoral Fellow at Umea University. He has also worked as an Assistant Professor in the Department of Computer Applications at MITS, Andhra Pradesh. Saqib Qamar received his PhD degree from Huazhong University of Science and Technology (HUST) in China. His research focuses on Compute Vision and Medical Image Analysis.
Current institution
Publications
Publications (30)
Glioma is one of the most widespread and intense forms of primary brain tumors. Accurate subcortical brain seg-mentation is essential in the evaluation of gliomas which helps to monitor the growth of gliomas and assists in the assessment of medication effects. Manual segmentation is needed a lot of human resources on Magnetic Resonance Imaging (MRI...
Automatic accurate segmentation of medical images has significant role in computer-aided diagnosis and disease treatment. The segmentation of cerebrospinal fluid (CSF), gray matter (GM), and white matter (WM) tissues plays an important role in infant brain structure for studying early brain development. However, this task is very challenging due to...
Magnetic Resonance Imaging (MRI) is dominant modality for infant brain analysis. Segmenting the whole infant MRI brain into number of tissues such as Cerebrospinal fluid (CSF), White matter (WM), and Gray Matter (GM) are highly desirable in the clinical environment. However, traditional methods tend to be degrading due to low contrast between GM an...
A residual 3D U-Net enables multi-scaling with the concatenation of feature maps from different scales. The connectors between different sub-networks assists in the concatenation of feature maps. A multi-path architecture enables the fusion of feature maps from different scales. In this paper, the combination of two different architectures is propo...
Purpose
Wood comprises different cell types, such as fibers, tracheids and vessels, defining its properties. Studying cells’ shape, size, and arrangement in microscopy images is crucial for understanding wood characteristics. Typically, this involves macerating (soaking) samples in a solution to separate cells, then spreading them on slides for ima...
Analyzing microscopy images of large growing cell samples using traditional methods is a complex and time-consuming process. In this work, we have developed an attention-driven UNet-enhanced model using deep learning techniques to efficiently quantify the position, area, and circularity of bacterial spores and vegetative cells from images containin...
Purpose: Wood comprises different cell types, such as fibers and vessels, defining its properties. Studying their shape, size, and arrangement in microscopic images is crucial for understanding wood samples. Typically, this involves macerating (soaking) samples in a solution to separate cells, then spreading them on slides for imaging with a micros...
We present a new approach to segment and classify bacterial spore layers from Transmission Electron Microscopy (TEM) images using a hybrid Convolutional Neural Network (CNN) and Random Forest (RF) classifier algorithm. This approach utilizes deep learning, with the CNN extracting features from images, and the RF classifier using those features for...
We present a new approach to segment and classify bacterial spore layers from Transmission Electron Microscopy (TEM) images using a hybrid Convolutional Neural Network (CNN) and Random Forest (RF) classifier algorithm. This approach utilizes deep learning, with the CNN extracting features from images, and the RF classifier using those features for...
CT vertebral segmentation plays an essential role in various clinical applications, such as computer-assisted surgical interventions, assessment of spinal abnormalities, and vertebral compression fractures. Automatic CT vertebral segmentation is challenging due to the overlapping shadows of thoracoabdominal structures such as the lungs, bony struct...
A deep convolutional neural network (CNN) achieves remarkable performance for medical image analysis. UNet is the primary source in the performance of 3D CNN architectures for medical imaging tasks, including brain tumor segmentation. The skip connection in the UNet architecture concatenates multi-scale features from image data. The multi-scaled fe...
UNet and its variations achieve state-of-the-art performances in medical image segmentation. In end-to-end learning, the training with high-resolution medical images achieves higher accuracy for medical image segmentation. However, the network depth, a massive number of parameters, and low receptive fields are issues in developing deep architecture...
The variations among shapes, sizes, and locations of tumors are obstacles for accurate automatic segmentation. U-Net is a simplified approach for automatic segmentation. Generally, the convolutional or the dilated convolutional layers are used for brain tumor segmentation. However, existing segmentation methods of the significant dilation rates deg...
Deep convolutional neural network (CNN) achieves remarkable performance for medical image analysis. UNet is the primary source in the performance of 3D CNN architectures for medical imaging tasks, including brain tumor segmentation. The skip connection in the UNet architecture concatenates features from both encoder and decoder paths to extract mul...
The brain tumor segmentation task aims to classify tissue into the whole tumor (WT), tumor core (TC) and enhancing tumor (ET) classes using multimodel MRI images. Quantitative analysis of brain tumors is critical for clinical decision making. While manual segmentation is tedious, time-consuming, and subjective, this task is at the same time very ch...
Background: Melanoma is one kind of dangerous cancer that has been increasing rapidly in the world. Initial diagnosis is essential to survival, but often the disease is diagnosed in the fatal stage. The rapid growth of skin cancers raises a huge demand for accurate automatic skin lesion segmentation. While deep learning techniques, i.e. convolution...
The brain tumor segmentation task aims to classify tissue into the whole tumor (WT), tumor core (TC), and enhancing tumor (ET) classes using multimodel MRI images. Quantitative analysis of brain tumors is critical for clinical decision making. While manual segmentation is tedious, time-consuming and subjective, this task is at the same time very ch...
Deep convolutional neural network (CNN) achieves remarkable performance for medical image analysis. UNet is the primary source in the performance of 3D CNN architectures for medical imaging tasks, including brain tumor segmentation. The skip connection in the UNet architecture concatenates features from both encoder and decoder paths to extract mul...
The accurate automatic segmentation of brain tumors enhances the probability of survival rate. Convolutional Neural Network (CNN) is a popular automatic approach for image evaluations. CNN provides excellent results against classical machine learning algorithms. In this paper, we present a unique approach to incorporate contexual information from m...
Deep learning algorithms have achieved remarkable results in natural language processing (NLP) and computer vision. Especially, deep learning methods such as convolution and recurrent neural networks have shown remarkable performance in text analytic task. Moreover, from the attention mechanism perspective convolutional neural network (CNN) is appl...
In the medical domain, 3D convolutional neural network has accomplished excellent results in analysing brain MRI. Inspired by the dense connection, dilation, types of feature fusion methods and availability of essential standards of information presented in multi-modalities and hierarchical knowledge transfers, we present a unique solution for auto...
The exact position of dead tissue in ischemic stroke lesions plays an important role to cure the life-threatening condition. However, this issue stays difficult caused by variation of ischemic strokes such as shape and location. Fully convolutional neural networks (FCN) have great potential for semantic image segmentation. Recently, UNet based CNN...
Convolutional neural networks (CNNs) are important methods in deep learning. They have presented up-to-date performance in different challenging areas, such as natural language processing and computer vision. These powerful and efficient neural networks implement training slowly under a massive number of network training parameters. The primary cha...
Objective of this article is to investigate the success (by determining citizen’s intention to use and satisfaction) of Land Record Information Systems (LRMIS) from perspective of Pakistan. The success of this egovernment information system is investigated using an incorporated IS success model. Model formulated here features constructs such as Sys...
Data mining is gaining popularity in disparate research fields due to its boundless applications and approaches to mine the data in an appropriate manner. Owing to the changes, the current world acquiring, it is one of the optimal approach for approximating the nearby future consequences. Along with advanced researches in healthcare monstrous of da...
Questions
Questions (11)
I have been using Mask-RCNN for work. I have to do custom object detection. For this, I have labeled all my images using polygon in the given image. I have 50 images and hence I have 50 annotations.json files. But according to the information given on the Mask RCNN Github Repo, we need only one annotation JSON file.
So my question is:
How to change the export_boxes and load_mask functions given in the code to accommodate my problem? If so, how do I do that? or should I merge all the JSON files into one? If I go ahead with the merging, would that have the correct formatting?
Pls help me out.
regards,
Saqib
Dear Researchers,
One of the big barrier in Medical Imaging is unavailability of annotated dataset . This question is still answerable, that whether we will be able to get enough training data without effecting the performance of deep learning algorithms.
Result get better with "More data + Bigger Model + More Computation" as described by Jeff Dean . What is the recent deep architecture (full automatic +semi automatic) with data enhanced technique to handle the issues of shortage data?? What factors we should consider in analyzing result of particular architecture such CNN, GAN etc...?? Your responses are highly appreciable..
Hi all,
I am working on Medical Imaging, I used Multi-scaled approach by using DenseNet, ResNet, Inception module etc.. to gain more contextual information from images (BRATS, iSeg, ISLES). Please let me know, what factors more affects to exploit image information and what would be recent approach for skin lesion type dataset? Please let me know about recent useful CNN architecture for Skin dataset and factors to make changes in them? your suggestion will be highly appreciable..
Thanks
Hi all,
I am working on Medical Image Segmentation, recently I used UNet and some variant form of UNet but with same skip connections from the encoder path to a decoder path. Please let me know, any recent application for skip connections to gain more information in the features concatenation from an encoder to decoder path?
Hi all,
This is a question only for knowledge purpose, I know someone introduced hybrid form of Dice and Cross Entropy to handle data imbalance problem in Colorectal Cancer Segmentation. Let me know about any recent loss function approach in ISLES, BRATS dataset?
Thanks
Hi,
As I read DWI modality data then its shape shows four dimensions (192, 192, 19, 80), and other modalities dimensions (192, 192, 19) even with label. Is DWI fourth dimension '80' represents "number of channels". Please let me know, what is the difference between DWI and other modalities of ISLES dataset ?
Your answer is highly appreciable.
Hello everybody,
Let me know about multiscale based deep learning method through Caffe. If anybody known about papers on the base of mulscaling, please share it....
What are the current research areas and challenges in brain image segmentation?
Does anyone have the IET computer vision latex template?
Please provide the link or Pdf for it..
I don`t know how to make changes in weight update method in Caffe solver. I want to apply my own formula at weight update place in Caffe. I have attached Formula doc and link where we have to modify.What should i do for the modification to it...
this is the link of solver file in Caffe
