Skip to content
akaturk Academic measurement

Article detail · 2022

Automated BI-RADS Classification of Lesions Using Pyramid Triple Deep Feature Generator Technique on Breast Ultrasound Images

Journal

Medical Engineering & Physics

ISSN 1350-4533

The ISSN points to another catalog journal; the name is from the YÖKSİS record.

YÖKSİS OpenAlex SJR Q3 JCR Q3 Citations 27 Top 10% Percentile 92.6% FWCI 3.2
Year
2022
Type
article

Data source split

  • YÖKSİS YÖKSİS article record
  • YÖKSİS venue Medical Engineering & Physics
  • Catalog match (ISSN) Medical Engineering and Physics
  • OpenAlex OpenAlex enrichment (abstract, citations, topics)

Abstract

English (OpenAlex)

Ultrasound (US) is an important imaging modality used to assess breast lesions for malignant features. In the past decade, many machine learning models have been developed for automated discrimination of breast cancer versus normal on US images, but few have classified the images based on the Breast Imaging Reporting and Data System (BI-RADS) classes. This work aimed to develop a model for classifying US breast lesions using a BI-RADS classification framework with a new multi-class US image dataset. We proposed a deep model that combined a novel pyramid triple deep feature generator (PTDFG) with transfer learning based on three pre-trained networks for creating deep features. Bilinear interpolation was applied to decompose the input image into four images of successively smaller dimensions, constituting a four-level pyramid for downstream feature generation with the pre-trained networks. Neighborhood component analysis was applied to the generated features to select each network's 1,000 most informative features, which were fed to support vector machine classifier for automated classification using a ten-fold cross-validation strategy. Our proposed model was validated using a new US image dataset containing 1,038 images divided into eight BI-RADS classes and histopathological results. We defined three classification schemes: Case 1 involved the classification of all images into eight categories; Case 2, classification of breast US images into five BI-RADS classes; and Case 3, classification of BI-RADS 4 lesions into benign versus malignant classes. Our PTDFG-based transfer learning model attained accuracy rates of 79.29%, 80.42%, and 88.67% for Case 1, Case 2, and Case 3, respectively.

Topics

  • AI in cancer detection
  • Radiomics and Machine Learning in Medical Imaging
  • Cervical Cancer and HPV Research

Primary topic AI in cancer detection

Authors

  1. ELA KAPLAN ADIYAMAN ÜNİVERSİTESİ
  2. WAI YEE CHAN
  3. ŞENGÜL DOĞAN
  4. PRABAL DATTA BARUA
  5. HACİ TANER BULUT
  6. TÜRKER TUNCER
  7. MERT ÇİZİK
  8. RU-SAN TAN
  9. U. RAJENDRA ACHARYA