Development and Assessing a Competitive Deep Learning Framework forBrain Tumor classification Based on MRI Images

  • سال انتشار: 1402
  • محل انتشار: نهمین کنفرانس بین المللی دانش و فناوری مهندسی مکانیک,برق و کامپیوتر ایران
  • کد COI اختصاصی: DMECONF09_096
  • زبان مقاله: انگلیسی
  • تعداد مشاهده: 46
دانلود فایل این مقاله

نویسندگان

Barat Barati

Department of Radiology technology, Shoushtar School of Medical Sciences, Shoushtar, Iran

Fariba farhadi Birgani

Department of Basic Sciences, Shoushtar Faculty of Medical Sciences, Shoushtar, Iran

Tahereh Navidifar

Department of Basic Sciences, Shoushtar Faculty of Medical Sciences, Shoushtar, Iran

seyed ali Mousavi

Department of Health, Shoushtar Faculty of Medical Science, Shoushtar, Iran

Karim Khoshgard

Department of Medical Physics, School of Medicine, Kermanshah University of Medical Sciences,Kermanshah, Iran

چکیده

The objective of this study is to introduce a model with the same number of layers as AlexNet andZFNet models with less training data in order to be competitive in the development of automatedmedical image analysis and brain tumor classification using a set of brain MRI images. A total of۳۰۰۰ brain magnetic resonance (MR) images, both with and without tumor presence, were sourcedfrom the Kaggle site. This dataset was divided into ۲۴۰۰ images for training and ۶۰۰ for testing.The model, implemented using the Python programming language with TensorFlow and Keraslibraries, competes against AlexNet, ZFNet, and VGG۱۶. Performance was evaluated on the basisof several experimental criteria, including accuracy, recall, F۱ score, precision, area under thereceiver operating characteristic curve (AUC), and training time. A comparative analysis amongthe models was conducted based on these metrics. The proposed model demonstrated competitiveperformance, ranking after AlexNet and ZFNet but outperforming VGG۱۶. This study contributedto the evolution of efficient deep learning models for medical image analysis, particularly for braintumor classification

کلیدواژه ها

brain, tumor, deep learning, algorithm, classification

مقالات مرتبط جدید

اطلاعات بیشتر در مورد COI

COI مخفف عبارت CIVILICA Object Identifier به معنی شناسه سیویلیکا برای اسناد است. COI کدی است که مطابق محل انتشار، به مقالات کنفرانسها و ژورنالهای داخل کشور به هنگام نمایه سازی بر روی پایگاه استنادی سیویلیکا اختصاص می یابد.

کد COI به مفهوم کد ملی اسناد نمایه شده در سیویلیکا است و کدی یکتا و ثابت است و به همین دلیل همواره قابلیت استناد و پیگیری دارد.