Automatic Diagnosis of COVID-19 Disease from Lung Images Using Hybrid Model of Pre-trained Deep Neural Networks
نام عام مواد
Dissertation
نام نخستين پديدآور
Isam Ali Nasir Al-Kinani
وضعیت نشر و پخش و غیره
نام ناشر، پخش کننده و غيره
Electrical and Computer Engineering
تاریخ نشرو بخش و غیره
1401
مشخصات ظاهری
نام خاص و کميت اثر
70p.
ساير جزييات
cd
یادداشتهای مربوط به پایان نامه ها
جزئيات پايان نامه و نوع درجه آن
M.A.
نظم درجات
Computer Engineering
زمان اعطا مدرک
1401/11/19
یادداشتهای مربوط به خلاصه یا چکیده
متن يادداشت
The COVID-19 pandemic, which has killed more than 5 million people globally since 2019, is one of the most lethal pandemics to have hit the planet. Early diagnosis and utilization of medical treatment are the best ways to manage this condition. Although there are several ways to diagnose the disease, one of the most effective methods of COVID-19 diagnosis is through a computer by using lung X-ray pictures. To identify this ailment, we integrated machine learning, deep learning, and meta-heuristic algorithms in this thesis. In this study, we extracted features from photos, using two pre-trained deep networks, AlexNet and ResNet-50. The grasshopper optimization algorithm (GOA) is also employed in the suggested method's continuation to choose the best features, and then those features are lastly entered into an SVM support vector machine classifier to detect corona illness. Based on the outcomes of the simulations, our suggested method has an average accuracy of 99.37% for detecting corona illness.
عنوانهای گونه گون دیگر
عنوان گونه گون
تشخیص خودکار COVID-19 از تصاویر ریه با استفاده از یک مدل ترکیبی از شبکه های عصبی عمیق از پیش آموزش داده شده
اصطلاحهای موضوعی کنترل نشده
اصطلاح موضوعی
COVID-19; Classification with SVM; Feature extraction deep learning; Grasshopper optimization algorithm; Early diagnosis COVID-19.
اصطلاح موضوعی
کووید 19؛ طبقه بندی با SVM. یادگیری عمیق استخراج ویژگی. الگوریتم بهینه سازی ملخ، تشخیص زودهنگام COVID-19
نام شخص به منزله سر شناسه - (مسئولیت معنوی درجه اول )