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

Hamyoon H 1 , Yee Chan W 2 , Mohammadi A 3 , Yusuf Kuzan T 4 , Mirza-Aghazadeh-Attari M 5 , Leong WL 6 Show all authors , Murzoglu Altintoprak K 4 , Vijayananthan A 6 , Rahmat K 6 , Ab Mumin N 7 , Sam Leong S 8 , Ejtehadifar S 3 , Faeghi F 9 , Abolghasemi J 10 , Ciaccio EJ 11 , Rajendra Acharya U 12 , Abbasian Ardakani A 13

Affiliations 

  • 1 Urology Research Center, Tehran University of Medical Sciences, Tehran, Iran
  • 2 Gleneagles Hospital Kuala Lumpur, Imaging Department, Jalan Ampang, Kampung Berembang, 50450 Kuala Lumpur, Malaysia
  • 3 Department of Radiology, Faculty of Medicine, Urmia University of Medical Science, Urmia, Iran
  • 4 Department of Radiology, Sancaktepe Sehit Prof. Dr. Ilhan Varank Training and Research Hospital, Istanbul, Turkey
  • 5 Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, MD, USA
  • 6 Department of Biomedical Imaging, Universiti Malaya Research Imaging Centre, Faculty of Medicine, Universiti Malaya, 50603 Malaysia
  • 7 Department of Biomedical Imaging, Universiti Malaya Research Imaging Centre, Faculty of Medicine, Universiti Malaya, 50603 Malaysia; Department of Radiology, Faculty of Medicine, Universiti Teknologi MARA, 47000 Selangor, Malaysia
  • 8 Centre of Medical Imaging, Faculty of Health Sciences, Universiti Teknologi MARA Selangor, Puncak Alam Campus, 42300 Bandar Puncak Alam, Selangor, Malaysia
  • 9 Department of Radiology Technology, School of Allied Medical Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran
  • 10 Department of Biostatistics, School of Public Health, Iran University of Medical Sciences, Tehran, Iran
  • 11 Department of Medicine, Columbia University, New York, NY 10032, USA
  • 12 Ngee Ann Polytechnic, Department of Electronics and Computer Engineering, 599489 Singapore, Singapore; Department of Biomedical Engineering, School of Science and Technology, SUSS University, Singapore; Department of Biomedical Informatics and Medical Engineering, Asia University, Taichung, Taiwan
  • 13 Department of Radiology Technology, School of Allied Medical Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran. Electronic address: Ardakani@sbmu.ac.ir
Eur J Radiol, 2022 Dec;157:110591.
PMID: 36356463 DOI: 10.1016/j.ejrad.2022.110591

Abstract

PURPOSE: To develop and validate a machine learning (ML) model for the classification of breast lesions on ultrasound images.

METHOD: 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.

RESULTS: 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).

CONCLUSIONS: These results support the possible role of morphometric features in enhancing the already well-excepted classification schemes.

* Title and MeSH Headings from MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine.

MeSH terms
Similar publications