Literature DB >> 34374941

Automatic breast mass detection in mammograms using density of wavelet coefficients and a patch-based CNN.

Behrouz NiroomandFam1, Alireza Nikravanshalmani2, Madjid Khalilian1.   

Abstract

PURPOSE: For the purpose of accurate and efficient mass detection in full-field digital mammograms, we propose a method for automated mass detection that consists of two stages: suspicious region localization and false-positive (FP) reduction, by classifying these regions into mass and non-mass regions (normal tissues).
METHODS: In the first stage, the density of the wavelet coefficients based on Quincunx Lifting Scheme (DWC-QLS) is used to find suspicious regions (regions of interest, ROIs) in full mammograms. In the second stage, a patch-based CNN classifier is developed as an FP reduction to classify the suspicious regions. The main aim of this stage is to reduce the false-positive suspicious regions while keeping the true-positive suspicious regions. To further improve the performance of the FP reduction, the effectiveness of different transfer learning strategies is further explored and the best fine-tuning strategy in training InceptionV3 model is determined experimentally.
RESULTS: The experimental results show that the proposed method can achieve an overall performance of 0.98 TPR @1.43 FPI on the INbreast database. In addition, the suggested segmentation method detects the mass location with 100% sensitivity and average of 5.4 false positives per image.
CONCLUSIONS: Based on the obtained results, the introduced method was able to successfully detect and classify suspicious regions in digital mammograms and provide better TPR and FPI results in comparison with other state-of-the-art method.
© 2021. CARS.

Entities:  

Keywords:  Breast masses; CAD; CNNs; InceptionV3; Transfer learning

Year:  2021        PMID: 34374941     DOI: 10.1007/s11548-021-02443-9

Source DB:  PubMed          Journal:  Int J Comput Assist Radiol Surg        ISSN: 1861-6410            Impact factor:   2.924


  6 in total

1.  INbreast: toward a full-field digital mammographic database.

Authors:  Inês C Moreira; Igor Amaral; Inês Domingues; António Cardoso; Maria João Cardoso; Jaime S Cardoso
Journal:  Acad Radiol       Date:  2011-11-10       Impact factor: 3.173

2.  A deep learning approach for the analysis of masses in mammograms with minimal user intervention.

Authors:  Neeraj Dhungel; Gustavo Carneiro; Andrew P Bradley
Journal:  Med Image Anal       Date:  2017-01-28       Impact factor: 8.545

3.  Automated Analysis of Unregistered Multi-View Mammograms With Deep Learning.

Authors:  Gustavo Carneiro; Jacinto Nascimento; Andrew P Bradley
Journal:  IEEE Trans Med Imaging       Date:  2017-09-12       Impact factor: 10.048

4.  Multicontext multitask learning networks for mass detection in mammogram.

Authors:  Rongbo Shen; Ke Zhou; Kezhou Yan; Kuan Tian; Jun Zhang
Journal:  Med Phys       Date:  2020-03-05       Impact factor: 4.071

Review 5.  Positive psychological functioning in breast cancer: An integrative review.

Authors:  Anna Casellas-Grau; Jaume Vives; Antoni Font; Cristian Ochoa
Journal:  Breast       Date:  2016-04-23       Impact factor: 4.380

6.  Deep Convolutional Neural Networks for breast cancer screening.

Authors:  Hiba Chougrad; Hamid Zouaki; Omar Alheyane
Journal:  Comput Methods Programs Biomed       Date:  2018-01-11       Impact factor: 5.428

  6 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.