Artificial Intelligence
I. Zabbah; K. Layeghi; Reza Ebrahimpour
Abstract
Background and Objectives: COVID-19 disease still has a devastating effect on society health. The use of X-ray images is one of the most important methods of diagnosing the disease. One of the challenges specialists are faced is no diagnosing in time. Using Deep learning can reduce the diagnostic error ...
Read More
Background and Objectives: COVID-19 disease still has a devastating effect on society health. The use of X-ray images is one of the most important methods of diagnosing the disease. One of the challenges specialists are faced is no diagnosing in time. Using Deep learning can reduce the diagnostic error of COVID-19 and help specialists in this field. Methods: The aim of this study is to provide a method based on a combination of deep learning(s) in parallel so that it can lead to more accurate results in COVID-19 disease by gathering opinions. In this research, 4 pre-trained (fine-tuned) deep model have been used. The dataset of this study is X-ray images from Github containing 1125 samples in 3 classes include normal, COVID-19 and pneumonia contaminated.Results: In all networks, 70% of the samples were used for training and 30% for testing. To ensure accuracy, the K-fold method was used in the training process. After modeling and comparing the generated models and recording the results, the accuracy of diagnosis of COVID-19 disease showed 84.3% and 87.2% when learners were not combined and experts were combined respectively. Conclusion: The use of machine learning techniques can lead to the early diagnosis of COVID-19 and help physicians to accelerate the healing process. This study shows that a combination of deep experts leads to improved diagnosis accuracy.
Image Processing
A. Mohammadi Anbaran; P. Torkzadeh; R. Ebrahimpour; N. Bagheri
Abstract
Background and Objectives: Programmable logic devices, such as Field Programmable Gate Arrays, are well-suited for implementing biologically-inspired visual processing algorithms and among those algorithms is HMAX model. This model mimics the feedforward path of object recognition in the visual cortex. ...
Read More
Background and Objectives: Programmable logic devices, such as Field Programmable Gate Arrays, are well-suited for implementing biologically-inspired visual processing algorithms and among those algorithms is HMAX model. This model mimics the feedforward path of object recognition in the visual cortex. Methods: HMAX includes several layers and its most computation intensive stage could be the S1 layer which applies 64 2D Gabor filters with various scales and orientations on the input image. A Gabor filter is the product of a Gaussian window and a sinusoid function. Using the separability property in the Gabor filter in the 0° and 90° directions and assuming the isotropic filter in the 45° and 135° directions, a 2D Gabor filter converts to two more efficient 1D filters.Results: The current paper presents a novel hardware architecture for the S1 layer of the HMAX model, in which a 1D Gabor filter is utilized twice to create a 2D filter. Using the even or odd symmetry properties in the Gabor filter coefficients reduce the required number of multipliers by about 50%. The normalization value in every input image location is also calculated simultaneously. The implementation of this architecture on the Xilinx Virtex-6 family shows a 2.83ms delay for a 128×128 pixel input image that is a 1.86X-speedup relative to the last best implementation.Conclusion: In this study, a hardware architecture is proposed to realize the S1 layer of the HMAX model. Using the property of separability and symmetry in filter coefficients saves significant resources, especially in DSP48 blocks.
