Multi-detection and Segmentation of Breast Lesions Based on Mask RCNN-FPN
Authors:
- 16 December 2020
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 2698-2704
Abstract
The presence of different malicious regions on a single breast reveal some necessary information for breast cancer early detection. In current computer-aided diagnosis models, different lesions contained in a single mammogram are not detected and segmented individually. Therefore, the multidetection and segmentation of the breast lesions can help the radiologists for an accurate diagnosis. This study aims to develop a model based on regional learning technique and RoI-based Convolutional neural network (CNN), which is known as Masked Regional Convolutional Neural Network embedded with Feature Pyramid Network. By using Mask RCNN-FPN, we can handle multi-detection, instance segmentation, and classification simultaneously. FPN extracts semantic features at different resolution scales and it can exhibit lesions at multiple scales. The training and testing of the model are performed on the DDSM and Inbreast respectively. In comparison, this model achieved mean average precision 0.84 for multi-detection and segmentation and 91% overall accuracy performance over SegNet and U-Net CNN encoder and decoder segmentation architecture.Keywords
Funding Information
- Shanghai Municipal Commission of Economy and Informatization
This publication has 31 references indexed in Scilit:
- Deep Learning and Structured Prediction for the Segmentation of Mass in Mammograms2015
- Convolutional feature masking for joint object and stuff segmentation2015
- Hypercolumns for object segmentation and fine-grained localization2015
- Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation2014
- Screening Mammography Benefit ControversiesRadiologic Clinics of North America, 2014
- Quantifying the Benefits and Harms of Screening MammographyJAMA Internal Medicine, 2014
- Selective Search for Object RecognitionInternational Journal of Computer Vision, 2013
- Computer‐aided classification of breast masses using speckle features of automated breast ultrasound imagesMedical Physics, 2012
- Speckle reduction imaging of breast ultrasound does not improve the diagnostic performance of morphology‐based CAD SystemJournal of Clinical Ultrasound, 2011
- Computer‐aided detection of breast masses on mammograms: Dual system approach with two‐view analysisMedical Physics, 2009