Çok kanallı CNN mimarisi ile X-Ray görüntülerinden COVID-19 tanısı

Abstract

Proposed multi-channel CNN deep learning architecture with channel selection formula  A new method for diagnosing Covid-19  High performance valuesDeep learning has been widely used in a variety of applications to solve a scope of complex problems that require extremely high accuracy and precision, especially in the medical field. In this study, the Covid-19 is diagnosed automatically using a proposed multichannel CNN method. Patients and healthy individuals' Lung X-Ray images data sets were obtained from three separate online databases. Simple recurrent networks (SRN) architecture was also applied for the same problem to compare the results and demonstrate the efficiency of the proposed method. The study proposes a new CNN-based method for early detection of Covid-19, which is a major risk to human life worldwide. Differently from the studies in the literature, the multi-channel CNN architecture with five convolution channels is proposed and the channel selection formulas are presented. It is used for selecting the most distinctive feature filters among the results produced by these channels. The architecture consists of the following components (Figure A). Figure A. Proposed model architecturePurpose: This study aims to diagnose the Covid-19 with deep learning methods to design an assistive technology system that can be used by doctors and health employee Theory and Methods: The proposed multi-channel CNN, and simple recurrent networks (SRN) models were trained on the three different dataset, which was obtained from the Kaggle data repository. The first dataset has 50 images, the second dataset has 1125 images and the third dataset has 5856 images. Their performances were compared with each other. Results:It was observed that the proposed multi-channel CNN model is better than other models in diagnosing Covid-19 with %99,41-%97,75-%96,74 accuracy, %99,554-%98,221-%96,923 F1-score and 0,89-1,13-1,92 RMSE values. The training process took 981546 seconds and the proposed model can also be used in real-time systems. Conclusion:In this study; A new method based on CNN is proposed for the early detection of Covid-19 from lung X-Ray images, which is a major risk to human life worldwide. Differently the studies in the literature, the multichannel CNN architecture with five convolution channels is proposed and the channel selection formulas is presented which are used for selecting the most distinctive feature filters among the results produced by these channels.

Shoulder Implant Manufacturer Detection by Using Deep Learning: Proposed Channel Selection Layer

“…In this section, the channel selection formula is introduced in the proposed model for selecting the most discriminatory feature filters. Yılmaz [27], diagnosed COVID-19 from lung X-ray images using a different deep learning network architecture with a feature selection layer which was successfully applied. In addition to implementing the 3-channel architecture and using different activation functions, the scores on all feature maps were summed to calculate the probability of each class, as distinct from this study.…”