A deep feature fusion methodology for breast cancer diagnosis demonstrated on three imaging modality datasets

Background Deep learning methods for radiomics/computer‐aided diagnosis (CADx) are often prohibited by small datasets, long computation time, and the need for extensive image preprocessing. Aims We aim to develop a breast CADx methodology that addresses the aforementioned issues by exploiting the ef...

Full description

Saved in:
Bibliographic Details
Published in:Medical physics (Lancaster) Vol. 44; no. 10; pp. 5162 - 5171
Main Authors: Antropova, Natalia, Huynh, Benjamin Q., Giger, Maryellen L.
Format: Journal Article
Language:English
Published: United States 01.10.2017
Subjects:
breast cancer
Breast Neoplasms - diagnostic imaging
deep learning
Diagnosis, Computer-Assisted - methods
feature extraction
Humans
Image Processing, Computer-Assisted - methods
Neural Networks (Computer)
Retrospective Studies
ROC Curve
deep learning
feature extraction
breast cancer
ISSN:0094-2405, 2473-4209, 2473-4209
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Background Deep learning methods for radiomics/computer‐aided diagnosis (CADx) are often prohibited by small datasets, long computation time, and the need for extensive image preprocessing. Aims We aim to develop a breast CADx methodology that addresses the aforementioned issues by exploiting the efficiency of pre‐trained convolutional neural networks (CNNs) and using pre‐existing handcrafted CADx features. Materials & Methods We present a methodology that extracts and pools low‐ to mid‐level features using a pretrained CNN and fuses them with handcrafted radiomic features computed using conventional CADx methods. Our methodology is tested on three different clinical imaging modalities (dynamic contrast enhanced‐MRI [690 cases], full‐field digital mammography [245 cases], and ultrasound [1125 cases]). Results From ROC analysis, our fusion‐based method demonstrates, on all three imaging modalities, statistically significant improvements in terms of AUC as compared to previous breast cancer CADx methods in the task of distinguishing between malignant and benign lesions. (DCE‐MRI [AUC = 0.89 (se = 0.01)], FFDM [AUC = 0.86 (se = 0.01)], and ultrasound [AUC = 0.90 (se = 0.01)]). Discussion/Conclusion We proposed a novel breast CADx methodology that can be used to more effectively characterize breast lesions in comparison to existing methods. Furthermore, our proposed methodology is computationally efficient and circumvents the need for image preprocessing.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
Authors contributed equally
ISSN:0094-2405
2473-4209
2473-4209
DOI:10.1002/mp.12453