TY - JOUR
T1 - End-to-End Calcification Distribution Pattern Recognition for Mammograms
T2 - An Interpretable Approach with GNN
AU - Yao, Melissa Min Szu
AU - Du, Hao
AU - Hartman, Mikael
AU - Chan, Wing P.
AU - Feng, Mengling
N1 - Funding Information:
This research was funded by Ministry of Education under the Higher Education Sprout Project in Taiwan, grant numbers DP2-109-21121-01-A-03, DP2-110-21121-01-A-03, and DP2-111-21121-01-A-11, the National Research Foundation Singapore under its AI Singapore program, grant number AISG-GC-2019-001-2A, and an NMRC Health Service Research Grant, grant number MOH-000030-00.
Funding Information:
Funding: This research was funded by Ministry of Education under the HigherEducation Sprout Project in Taiwan, grant numbers DP2-109-21121-01-A-03, DP2-110-21121-01-A-03, and DP2-111-21121-01-A-11,theNationalResearchFoundationSingaporeunderitsAISingaporeprogram,grant numberAISG-GC-2019-001-2A,andanNMRCHealthServiceResearchGrant,grantnumberMOH-000030-00.
Publisher Copyright:
© 2022 by the authors. Licensee MDPI, Basel, Switzerland.
PY - 2022/6
Y1 - 2022/6
N2 - Purpose: We aimed to develop a novel interpretable artificial intelligence (AI) model algorithm focusing on automatic detection and classification of various patterns of calcification distribution in mammographic images using a unique graph convolution approach. Materials and methods: Images from 292 patients, which showed calcifications according to the mammographic reports and diagnosed breast cancers, were collected. The calcification distributions were classified as diffuse, segmental, regional, grouped, or linear. Excluded were mammograms with (1) breast cancer with multiple lexicons such as mass, asymmetry, or architectural distortion without calcifi-cations; (2) hidden calcifications that were difficult to mark; or (3) incomplete medical records. Re-sults: A graph-convolutional-network-based model was developed. A total of 581 mammographic images from 292 cases of breast cancer were divided based on the calcification distribution pattern: diffuse (n = 67), regional (n = 115), group (n = 337), linear (n = 8), or segmental (n = 54). The classification performances were measured using metrics including precision, recall, F1 score, accuracy, and multi-class area under the receiver operating characteristic curve. The proposed model achieved a precision of 0.522 ± 0.028, sensitivity of 0.643 ± 0.017, specificity of 0.847 ± 0.009, F1 score of 0.559 ± 0.018, accuracy of 64.325 ± 1.694%, and area under the curve of 0.745 ± 0.030; thus, the method was found to be superior compared to all baseline models. The predicted linear and diffuse classifications were highly similar to the ground truth, and the predicted grouped and regional classifications were also superior compared to baseline models. The prediction results are interpretable using visualization methods to highlight the important calcification nodes in graphs. Conclu-sions: The proposed deep neural network framework is an AI solution that automatically detects and classifies calcification distribution patterns on mammographic images highly suspected of showing breast cancers. Further study of the AI model in an actual clinical setting and additional data collection will improve its performance.
AB - Purpose: We aimed to develop a novel interpretable artificial intelligence (AI) model algorithm focusing on automatic detection and classification of various patterns of calcification distribution in mammographic images using a unique graph convolution approach. Materials and methods: Images from 292 patients, which showed calcifications according to the mammographic reports and diagnosed breast cancers, were collected. The calcification distributions were classified as diffuse, segmental, regional, grouped, or linear. Excluded were mammograms with (1) breast cancer with multiple lexicons such as mass, asymmetry, or architectural distortion without calcifi-cations; (2) hidden calcifications that were difficult to mark; or (3) incomplete medical records. Re-sults: A graph-convolutional-network-based model was developed. A total of 581 mammographic images from 292 cases of breast cancer were divided based on the calcification distribution pattern: diffuse (n = 67), regional (n = 115), group (n = 337), linear (n = 8), or segmental (n = 54). The classification performances were measured using metrics including precision, recall, F1 score, accuracy, and multi-class area under the receiver operating characteristic curve. The proposed model achieved a precision of 0.522 ± 0.028, sensitivity of 0.643 ± 0.017, specificity of 0.847 ± 0.009, F1 score of 0.559 ± 0.018, accuracy of 64.325 ± 1.694%, and area under the curve of 0.745 ± 0.030; thus, the method was found to be superior compared to all baseline models. The predicted linear and diffuse classifications were highly similar to the ground truth, and the predicted grouped and regional classifications were also superior compared to baseline models. The prediction results are interpretable using visualization methods to highlight the important calcification nodes in graphs. Conclu-sions: The proposed deep neural network framework is an AI solution that automatically detects and classifies calcification distribution patterns on mammographic images highly suspected of showing breast cancers. Further study of the AI model in an actual clinical setting and additional data collection will improve its performance.
KW - artificial intelligence
KW - calcifications
KW - deep learning
KW - graph convolution network
KW - mammography
UR - http://www.scopus.com/inward/record.url?scp=85131717784&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85131717784&partnerID=8YFLogxK
U2 - 10.3390/diagnostics12061376
DO - 10.3390/diagnostics12061376
M3 - Article
AN - SCOPUS:85131717784
SN - 2075-4418
VL - 12
JO - Diagnostics
JF - Diagnostics
IS - 6
M1 - 1376
ER -