Share Email Print

Proceedings Paper

A multitask deep learning method in simultaneously predicting occult invasive disease in ductal carcinoma in-situ and segmenting microcalcifications in mammography
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

We proposed a two-branch multitask learning convolutional neural network to solve two different but related tasks at the same time. Our main task is to predict occult invasive disease in biopsy proven Ductal Carcinoma in-situ (DCIS), with an auxiliary task of segmenting microcalcifications (MCs). In this study, we collected digital mammography from 604 patients, 400 of which were DCIS. The model used patches with size of 512×512 extracted within a radiologist masked ROIs as input, with outputs including noisy MC segmentations obtained from our previous algorithms, and classification labels from final diagnosis at patients’ definite surgery. We utilized a deep multitask model by combining both Unet segmentation networks and prediction classification networks, by sharing first several convolutional layers. The model achieved a patch-based ROC-AUC of 0.69, with a case-based ROC-AUC of 0.61. Segmentation results achieved a dice coefficient of 0.49.

Paper Details

Date Published: 23 March 2020
PDF: 6 pages
Proc. SPIE 11314, Medical Imaging 2020: Computer-Aided Diagnosis, 1131405 (23 March 2020); doi: 10.1117/12.2549669
Show Author Affiliations
Rui Hou, Duke Univ. (United States)
Lars J. Grimm, Duke Univ. (United States)
Maciej A. Mazurowski, Duke Univ. (United States)
Jeffrey R. Marks, Duke Univ. School of Medicine (United States)
Lorraine M. King, Duke Univ. School of Medicine (United States)
Carlo C. Maley, Arizona State Univ. (United States)
E. Shelley Hwang, Duke Univ. (United States)
Joseph Y. Lo, Duke Univ. (United States)

Published in SPIE Proceedings Vol. 11314:
Medical Imaging 2020: Computer-Aided Diagnosis
Horst K. Hahn; Maciej A. Mazurowski, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?