Deep Learning for Fully Automatic Tumor Segmentation on Serially Acquired Dynamic Contrast-Enhanced MRI Images of Triple-Negative Breast Cancer-Reference-Cited by-同舟云学术

Deep Learning for Fully Automatic Tumor Segmentation on Serially Acquired Dynamic Contrast-Enhanced MRI Images of Triple-Negative Breast Cancer

Published:2023-10-02 Issue:19 Volume:15 Page:4829
ISSN:2072-6694
Container-title:Cancers
language:en
Short-container-title:Cancers

Author:

Xu Zhan¹^ORCID,Rauch David E.¹,Mohamed Rania M.²^ORCID,Pashapoor Sanaz²,Zhou Zijian¹,Panthi Bikash¹,Son Jong Bum¹,Hwang Ken-Pin¹^ORCID,Musall Benjamin C.¹,Adrada Beatriz E.²,Candelaria Rosalind P.²,Leung Jessica W. T.²,Le-Petross Huong T. C.²,Lane Deanna L.²,Perez Frances²,White Jason³,Clayborn Alyson³,Reed Brandy⁴,Chen Huiqin⁵,Sun Jia⁵,Wei Peng⁵^ORCID,Thompson Alastair⁶,Korkut Anil⁷,Huo Lei⁸,Hunt Kelly K.⁹,Litton Jennifer K.³,Valero Vicente³,Tripathy Debu³^ORCID,Yang Wei²,Yam Clinton³,Ma Jingfei¹

Affiliation:

1. Department of Imaging Physics, The University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA

2. Department of Breast Imaging, The University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA

3. Department of Breast Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA

4. Department of Clinical Research Imaging, The University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA

5. Department of Biostatistics, The University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA

6. Section of Breast Surgery, Baylor College of Medicine, Houston, TX 77030, USA

7. Department of Bioinformatics & Computational Biology, The University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA

8. Department of Pathology, The University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA

9. Department of Breast Surgical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX 77030, USA

Abstract

Accurate tumor segmentation is required for quantitative image analyses, which are increasingly used for evaluation of tumors. We developed a fully automated and high-performance segmentation model of triple-negative breast cancer using a self-configurable deep learning framework and a large set of dynamic contrast-enhanced MRI images acquired serially over the patients’ treatment course. Among all models, the top-performing one that was trained with the images across different time points of a treatment course yielded a Dice similarity coefficient of 93% and a sensitivity of 96% on baseline images. The top-performing model also produced accurate tumor size measurements, which is valuable for practical clinical applications.

Funder

NIH/NCI

Publisher

MDPI AG

Subject

Cancer Research,Oncology

Link

https://www.mdpi.com/2072-6694/15/19/4829/pdf

Reference55 articles.

1. Wu, Q., Siddharth, S., and Sharma, D. (2021). Triple Negative Breast Cancer: A Mountain Yet to Be Scaled Despite the Triumphs. Cancers, 13.

2. Triple-Negative Breast Cancer: Clinical Features and Patterns of Recurrence;Dent;Clin. Cancer Res.,2007

3. Automatic Detection and Segmentation of Breast Cancer on MRI Using Mask R-CNN Trained on Non–Fat-Sat Images and Tested on Fat-Sat Images;Zhang;Acad. Radiol.,2022

4. Multi-input deep learning architecture for predicting breast tumor response to chemotherapy using quantitative MR images;Drisis;Int. J. Comput. Assist. Radiol. Surg.,2020

5. Prior to Initiation of Chemotherapy, Can We Predict Breast Tumor Response? Deep Learning Convolutional Neural Networks Approach Using a Breast MRI Tumor Dataset;Ha;J. Digit. Imaging,2018