Artificial intelligence, BI-RADS evaluation and morphometry: A novel combination to diagnose breast cancer using ultrasonography, results from multi-center cohorts

•ACR BI-RADS is a widely accepted lexicon and reporting scheme used for breast lesions to communicate the degree to which a lesion may be of malignant potential.•We proposed a machine-learning model fed with BI-RADS and morphometric data, and externally validated the model with data from multiple ce...

Full description

Saved in:

Bibliographic Details
Published in:	European journal of radiology Vol. 157; p. 110591
Main Authors:	Hamyoon, Hessam, Yee Chan, Wai, Mohammadi, Afshin, Yusuf Kuzan, Taha, Mirza-Aghazadeh-Attari, Mohammad, Leong, Wai Ling, Murzoglu Altintoprak, Kübra, Vijayananthan, Anushya, Rahmat, Kartini, Ab Mumin, Nazimah, Sam Leong, Sook, Ejtehadifar, Sajjad, Faeghi, Fariborz, Abolghasemi, Jamileh, Ciaccio, Edward J., Rajendra Acharya, U., Abbasian Ardakani, Ali
Format:	Journal Article
Language:	English
Published:	Elsevier B.V 01-12-2022
Subjects:	Artificial intelligence BI-RADS Breast cancer Machine learning Ultrasound BI-RADS Breast cancer Artificial intelligence Ultrasound Machine learning
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	•ACR BI-RADS is a widely accepted lexicon and reporting scheme used for breast lesions to communicate the degree to which a lesion may be of malignant potential.•We proposed a machine-learning model fed with BI-RADS and morphometric data, and externally validated the model with data from multiple centers.•Our model achieved a sensitivity of 93.0%, specificity of 79.5%, accuracy of 86.0%, and an AUC of 0.885 in classifying 563 breast lesions from three centers.•The BI-RADS and morphometric data model outperformed an expert radiologist and radiology resident in all datasets. To develop and validate a machine learning (ML) model for the classification of breast lesions on ultrasound images. In the present study, three separate data cohorts containing 1288 breast lesions from three countries (Malaysia, Iran, and Turkey) were utilized for MLmodel development and external validation. The model was trained on ultrasound images of 725 breast lesions, and validation was done separately on the remaining data. An expert radiologist and a radiology resident classified the lesions based on the BI-RADS lexicon. Thirteen morphometric features were selected from a contour of the lesion and underwent a three-step feature selection process. Five features were chosen to be fed into the model separately and combined with the imaging signs mentioned in the BI-RADS reference guide. A support vector classifier was trained and optimized. The diagnostic profile of the model with various input data was compared to the expert radiologist and radiology resident. The agreement of each approach with histopathologic specimens was also determined. Based on BI-RADS and morphometric features, the model achieved an area under the receiver operating characteristic (ROC) curve (AUC) of 0.885, which is higher than the expert radiologist and radiology resident performances with AUC of 0.814 and 0.632, respectively in all cohorts. DeLong’s test also showed that the AUC of the ML protocol was significantly different from that of the expert radiologist (ΔAUCs = 0.071, 95%CI: (0.056, 0.086), P = 0.005). These results support the possible role of morphometric features in enhancing the already well-excepted classification schemes.
Bibliography:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23
ISSN:	0720-048X 1872-7727
DOI:	10.1016/j.ejrad.2022.110591