Classification of malignant and benign breast cancer using artificial neural network based on mammography image feature

سال انتشار: 1394
نوع سند: مقاله کنفرانسی
زبان: انگلیسی
مشاهده: 383

نسخه کامل این مقاله ارائه نشده است و در دسترس نمی باشد

این مقاله در بخشهای موضوعی زیر دسته بندی شده است:

استخراج به نرم افزارهای پژوهشی:

لینک ثابت به این مقاله:

شناسه ملی سند علمی:

ICBCMED12_165

تاریخ نمایه سازی: 2 تیر 1397

چکیده مقاله:

Introduction & Aim: There are different biomarkers to detect and classify breast cancer, but they may be obtained invasively. Image biomarkers are newly advanced markers which extracted from mammography images and can be used to diagnosis, prognosis and classify breast cancer. The aim of current work was to test these biomarkers using artificial neural network (ANN) for classification of malignant and benign breast cancer Methods: This retrospective study was done on mammography image dataset including 408 malignant and 435 benign breast cancer patients (843 patients). Different image features such as fractal dimension, shape density, margin, mass, and convexity were extracted mammography and were used as input for ANN machine learning approach. Two ANN classifier including multilayer-perceptron (MLP) and radial basis function (RBF) were used to classification of benign and malignant breast cancer. Sixty six percent of data was recruited as test and remained was used as train. Results: Our results for accuracy, area under the curve, recall and precision for MLP were 80.42, 85.6, 80.4 and 80.5 and for RBF were 74.56, 82.7, 74.6 and 77.5 respectively. Conclusion: Image classification using image features based on artificial intelligence is advanced technique which demonstrated high precision and yielded good accuracy. Using these methods, breast cancer malignancy can be detected more accurate and reduce invasive method such as biopsies

نویسندگان

Isaac Shiri

Department of Medical Physics, Iran University of Medical Sciences, Tehran, Iran

Hamid Abdollahi

Department of Medical Physics, Iran University of Medical Sciences, Tehran, Iran

Seied Rabi Mahdavi

Department of Medical Physics, Iran University of Medical Sciences, Tehran, Iran