Grayscale Ultrasound Radiomic Features and Shear-Wave Elastography Radiomic Features in Benign and Malignant Breast Masses

Abstract

Purpose To identify and compare diagnostic performance of radiomic features between grayscale ultrasound (US) and shear-wave elastography (SWE) in breast masses.

Materials and Methods We retrospectively collected 328 pathologically confirmed breast masses in 296 women who underwent grayscale US and SWE before biopsy or surgery. A representative SWE image of the mass displayed with a grayscale image in split-screen mode was selected. An ROI was delineated around the mass boundary on the grayscale image and copied and pasted to the SWE image by a dedicated breast radiologist for lesion segmentation. A total of 730 candidate radiomic features including first-order statistics and textural and wavelet features were extracted from each image. LASSO regression was used for data dimension reduction and feature selection. Univariate and multivariate logistic regression was performed to identify independent radiomic features, differentiating between benign and malignant masses with calculation of the AUC.

Results Of 328 breast masses, 205 (62.5 %) were benign and 123 (37.5 %) were malignant. Following radiomic feature selection, 22 features from grayscale and 6 features from SWE remained. On univariate analysis, all 6 SWE radiomic features (P < 0.0001) and 21 of 22 grayscale radiomic features (P < 0.03) were significantly different between benign and malignant masses. After multivariate analysis, three grayscale radiomic features and two SWE radiomic features were independently associated with malignant breast masses. The AUC was 0.929 for grayscale US and 0.992 for SWE (P < 0.001).

Conclusion US radiomic features may have the potential to improve diagnostic performance for breast masses, but further investigation of independent and larger datasets is needed.

Zusammenfassung

Ziel Ermittlung und Vergleich der diagnostischen Leistung von radiomischen Merkmalen zwischen B-Bild-Ultraschall (US) und Scherwellen-Elastografie (SWE) bei Raumforderungen der Brust.

Material und Methoden Wir haben retrospektiv 328 pathologisch bestätigte Raumforderungen der Brust bei 296 Frauen erhoben, die sich vor einer Biopsie oder Operation einer US- und SWE-Untersuchung unterzogen hatten. Ein repräsentatives SWE-Bild der Raumforderung, das zusammen mit dem B-Bild im Split-Screen-Modus angezeigt wird, wurde ausgewählt. Eine ROI wurde durch einen spezialisierten Brustradiologen um die Grenze der Raumforderung im B-Bild eingezeichnet, kopiert und in das SWE-Bild eingefügt, um die Läsion zu segmentieren. Aus jedem Bild wurden insgesamt 730 mögliche radiomische Merkmale, einschließlich Statistiken erster Ordnung, sowie Struktur- und Wavelet-Merkmale extrahiert. Die LASSO-Regression wurde zur Reduzierung der Datengröße und zur Merkmalsauswahl verwendet. Eine univariate und multivariate logistische Regression erfolgte, um unabhängige radiomische Merkmale zu identifizieren, die bei der AUC-Berechnung zwischen gut- und bösartigen Raumforderungen unterscheiden.

Ergebnisse Von 328 Raumforderungen waren 205 (62,5 %) gutartig und 123 (37,5 %) maligne. Nach der Selektion der radiomischen Merkmale blieben 22 Merkmale aus dem B-Bild und 6 aus der SWE übrig. Bei der univariaten Analyse unterschieden alle 6 SWE-radiomischen Merkmale (p < 0,0001) und 21 von 22 B-Bild-radiomischen Merkmalen (p < 0,03) signifikant zwischen benignen und malignen Raumforderungen. Nach der multivariaten Analyse wurden 3 B-Bild- und 2 SWE-radiomische Merkmale unabhängig voneinander mit malignen Raumforderungen der Brust assoziiert. Die AUC betrug 0,929 für B-Bild und 0,992 für SWE (p < 0,001).

Schlussfolgerung Radiomische US-Merkmale können die Diagnoseleistung bei Raumforderungen der Brust verbessern, jedoch ist die weitere Untersuchung unabhängiger und größerer Datensätze erforderlich.

Key words

breast - breast neoplasm - ultrasound - elastography - clinical informatics

• References

• 1 Evans A, Whelehan P, Thomson K. et al. Differentiating benign from malignant solid breast masses: value of shear wave elastography according to lesion stiffness combined with greyscale ultrasound according to BI-RADS classification. Br J Cancer 2012; 107: 224-229
• 2 Wojcinski S, Farrokh A, Weber S. et al. Multicenter study of ultrasound real-time tissue elastography in 779 cases for the assessment of breast lesions: improved diagnostic performance by combining the BI-RADS®-US classification system with sonoelastography. Ultraschall in Med 2010; 31: 484-491
• 3 Farrokh A, Wojcinski S, Degenhardt F. Diagnostic value of strain ratio measurement in the differentiation of malignant and benign breast lesions. Ultraschall in Med 2011; 32: 400-405
• 4 Youk JH, Gweon HM, Son EJ. Shear-wave elastography in breast ultrasonography: the state of the art. Ultrasonography 2017; 36: 300-309
• 5 Zhang Q, Xiao Y, Chen S. et al. Quantification of elastic heterogeneity using contourlet-based texture analysis in shear-wave elastography for breast tumor classification. Ultrasound Med Biol 2015; 41: 588-600
• 6 Xiao Y, Zeng J, Niu L. et al. Computer-aided diagnosis based on quantitative elastographic features with supersonic shear wave imaging. Ultrasound Med Biol 2014; 40: 275-286
• 7 Gillies RJ, Kinahan PE, Hricak H. Radiomics: Images Are More than Pictures, They Are Data. Radiology 2016; 278: 563-577
• 8 Guo Y, Hu Y, Qiao M. et al. Radiomics Analysis on Ultrasound for Prediction of Biologic Behavior in Breast Invasive Ductal Carcinoma. Clin Breast Cancer 2018; 18: e335-e344
• 9 Moon WK, Huang YS, Lee YW. et al. Computer-aided tumor diagnosis using shear wave breast elastography. Ultrasonics 2017; 78: 125-133
• 10 Bhatia KS, Lam AC, Pang SW. et al. Feasibility Study of Texture Analysis Using Ultrasound Shear Wave Elastography to Predict Malignancy in Thyroid Nodules. Ultrasound Med Biol 2016; 42: 1671-1680
• 11 Aerts HJ, Velazquez ER, Leijenaar RT. et al. Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach. Nat Commun 2014; 5: 4006
• 12 Youk JH, Jung I, Yoon JH. et al. Comparison of Inter-Observer Variability and Diagnostic Performance of the Fifth Edition of BI-RADS for Breast Ultrasound of Static versus Video Images. Ultrasound Med Biol 2016; 42: 2083-2088
• 13 Mendelson EB, Böhm-Vélez M, Berg WA. et al. ACR BI-RADS® Ultrasound. ACR BI-RADS® Atlas, Breast Imaging Reporting and Data System. Reston, VA: American College of Radiology; 2013
• 14 Lee SH, Cho N, Chang JM. et al. Two-view versus single-view shear-wave elastography: comparison of observer performance in differentiating benign from malignant breast masses. Radiology 2014; 270: 344-353
• 15 Cosgrove DO, Berg WA, Dore CJ. et al. Shear wave elastography for breast masses is highly reproducible. Eur Radiol 2012; 22: 1023-1032
• 16 Xiao Y, Yu Y, Niu L. et al. Quantitative evaluation of peripheral tissue elasticity for ultrasound-detected breast lesions. Clin Radiol 2016; 71: 896-904
• 17 Wang M, Yang Z, Liu C. et al. Differential Diagnosis of Breast Category 3 and 4 Nodules Through BI-RADS Classification in Conjunction with Shear Wave Elastography. Ultrasound Med Biol 2017; 43: 601-606
• 18 Skerl K, Vinnicombe S, Giannotti E. et al. Influence of region of interest size and ultrasound lesion size on the performance of 2D shear wave elastography (SWE) in solid breast masses. Clin Radiol 2015; 70: 1421-1427
• 19 Blank MAB, Antaki JF. Breast Lesion Elastography Region of Interest Selection and Quantitative Heterogeneity: A Systematic Review and Meta-Analysis. Ultrasound Med Biol 2017; 43: 387-397
• 20 Youk JH, Son EJ, Han K. et al. Performance of shear-wave elastography for breast masses using different region-of-interest (ROI) settings. Acta Radiol 2018; 59: 789-797
• 21 Acharya UR, Ng WL, Rahmat K. et al. Shear wave elastography for characterization of breast lesions: Shearlet transform and local binary pattern histogram techniques. Comput Biol Med 2017; 91: 13-20
• 22 Lo CM, Lai YC, Chou YH. et al. Quantitative breast lesion classification based on multichannel distributions in shear-wave imaging. Comput Methods Programs Biomed 2015; 122: 354-361
• 23 Acharya UR, Ng WL, Rahmat K. et al. Data mining framework for breast lesion classification in shear wave ultrasound: a hybrid feature paradigm. Biomed Signal Process Control 2017; 33: 400-410
• 24 Yu Y, Xiao Y, Cheng J. et al. Breast lesion classification based on supersonic shear-wave elastography and automated lesion segmentation from B-mode ultrasound images. Comput Biol Med 2018; 93: 31-46
• 25 Parekh V, Jacobs MA. Radiomics: a new application from established techniques. Expert Rev Precis Med Drug Dev 2016; 1: 207-226
• 26 Srinivasan GN, Shobha G. Statistical texture analysis proceedings of world academy of science. Eng Technol 2008; 36: 2070-3740
• 27 Barr RG, Nakashima K, Amy D. et al. WFUMB guidelines and recommendations for clinical use of ultrasound elastography: Part 2: breast. Ultrasound Med Biol 2015; 41: 1148-1160
• 28 Athanasiou A, Tardivon A, Tanter M. et al. Breast lesions: quantitative elastography with supersonic shear imaging – preliminary results. Radiology 2010; 256: 297-303
• 29 Evans A, Whelehan P, Thomson K. et al. Quantitative shear wave ultrasound elastography: initial experience in solid breast masses. Breast Cancer Res 2010; 12: R104
• 30 Au FW, Ghai S, Moshonov H. et al. Diagnostic performance of quantitative shear wave elastography in the evaluation of solid breast masses: determination of the most discriminatory parameter. Am J Roentgenol 2014; 203: W328-W336