Skip to main content
Top
Published in:

Open Access 09-05-2024 | Artificial Intelligence | Breast Radiology

Artificial intelligence-based, semi-automated segmentation for the extraction of ultrasound-derived radiomics features in breast cancer: a prospective multicenter study

Authors: Tommaso Vincenzo Bartolotta, Carmelo Militello, Francesco Prinzi, Fabiola Ferraro, Leonardo Rundo, Calogero Zarcaro, Mariangela Dimarco, Alessia Angela Maria Orlando, Domenica Matranga, Salvatore Vitabile

Published in: La radiologia medica | Issue 7/2024

Login to get access

Abstract

Purpose

To investigate the feasibility of an artificial intelligence (AI)-based semi-automated segmentation for the extraction of ultrasound (US)-derived radiomics features in the characterization of focal breast lesions (FBLs).

Material and methods

Two expert radiologists classified according to US BI-RADS criteria 352 FBLs detected in 352 patients (237 at Center A and 115 at Center B). An AI-based semi-automated segmentation was used to build a machine learning (ML) model on the basis of B-mode US of 237 images (center A) and then validated on an external cohort of B-mode US images of 115 patients (Center B).

Results

A total of 202 of 352 (57.4%) FBLs were benign, and 150 of 352 (42.6%) were malignant. The AI-based semi-automated segmentation achieved a success rate of 95.7% for one reviewer and 96% for the other, without significant difference (p = 0.839). A total of 15 (4.3%) and 14 (4%) of 352 semi-automated segmentations were not accepted due to posterior acoustic shadowing at B-Mode US and 13 and 10 of them corresponded to malignant lesions, respectively. In the validation cohort, the characterization made by the expert radiologist yielded values of sensitivity, specificity, PPV and NPV of 0.933, 0.9, 0.857, 0.955, respectively. The ML model obtained values of sensitivity, specificity, PPV and NPV of 0.544, 0.6, 0.416, 0.628, respectively. The combined assessment of radiologists and ML model yielded values of sensitivity, specificity, PPV and NPV of 0.756, 0.928, 0.872, 0.855, respectively.

Conclusion

AI-based semi-automated segmentation is feasible, allowing an instantaneous and reproducible extraction of US-derived radiomics features of FBLs. The combination of radiomics and US BI-RADS classification led to a potential decrease of unnecessary biopsy but at the expense of a not negligible increase of potentially missed cancers.
Literature
1.
go back to reference International agency for research on cancer (2020) The global cancer observatory. World Health Organization International agency for research on cancer (2020) The global cancer observatory. World Health Organization
2.
go back to reference Mokhtari-Hessari P, Montazeri A (2020) Health-related quality of life in breast cancer patients: review of reviews from 2008 to 2018. Health Qual Life Outcomes 18:338PubMedPubMedCentralCrossRef Mokhtari-Hessari P, Montazeri A (2020) Health-related quality of life in breast cancer patients: review of reviews from 2008 to 2018. Health Qual Life Outcomes 18:338PubMedPubMedCentralCrossRef
3.
go back to reference Brown C, Nazeer R, Gibbs A et al (2023) Breaking bias: the role of artificial intelligence in improving clinical decision-making. Cureus 15:e36415PubMedPubMedCentral Brown C, Nazeer R, Gibbs A et al (2023) Breaking bias: the role of artificial intelligence in improving clinical decision-making. Cureus 15:e36415PubMedPubMedCentral
4.
go back to reference Smith H, Fotheringham K (2020) Artificial intelligence in clinical decision-making: rethinking liability. Med Law Int 20:096853322094576CrossRef Smith H, Fotheringham K (2020) Artificial intelligence in clinical decision-making: rethinking liability. Med Law Int 20:096853322094576CrossRef
5.
go back to reference Gillies RJ, Kinahan PE, Hricak H (2016) Radiomics: images are more than pictures, they are data. Radiology 278:563–577PubMedCrossRef Gillies RJ, Kinahan PE, Hricak H (2016) Radiomics: images are more than pictures, they are data. Radiology 278:563–577PubMedCrossRef
6.
go back to reference Lambin P, Leijenaar RTH, Deist TM et al (2017) Radiomics: the bridge between medical imaging and personalized medicine. Nat Rev Clin Oncol 14:749–762PubMedCrossRef Lambin P, Leijenaar RTH, Deist TM et al (2017) Radiomics: the bridge between medical imaging and personalized medicine. Nat Rev Clin Oncol 14:749–762PubMedCrossRef
8.
go back to reference Kabiraj S, Raihan M, Alvi N, et al (2020) Breast cancer risk prediction using XGBoost and random forest algorithm. In: 2020 11th International conference on computing, communication and networking technologies (ICCCNT). IEEE Kabiraj S, Raihan M, Alvi N, et al (2020) Breast cancer risk prediction using XGBoost and random forest algorithm. In: 2020 11th International conference on computing, communication and networking technologies (ICCCNT). IEEE
9.
go back to reference Ghiasi MM, Zendehboudi S (2021) Application of decision tree-based ensemble learning in the classification of breast cancer. Comput Biol Med 128:104089PubMedCrossRef Ghiasi MM, Zendehboudi S (2021) Application of decision tree-based ensemble learning in the classification of breast cancer. Comput Biol Med 128:104089PubMedCrossRef
10.
go back to reference Lin A, Kolossváry M, Yuvaraj J et al (2020) myocardial infarction associates with a distinct pericoronary adipose tissue radiomic phenotype: a prospective case-control study. JACC Cardiovasc Imaging 13:2371–2383PubMedPubMedCentralCrossRef Lin A, Kolossváry M, Yuvaraj J et al (2020) myocardial infarction associates with a distinct pericoronary adipose tissue radiomic phenotype: a prospective case-control study. JACC Cardiovasc Imaging 13:2371–2383PubMedPubMedCentralCrossRef
11.
12.
go back to reference Tagliafico AS, Piana M, Schenone D et al (2020) Overview of radiomics in breast cancer diagnosis and prognostication. Breast 49:74–80PubMedCrossRef Tagliafico AS, Piana M, Schenone D et al (2020) Overview of radiomics in breast cancer diagnosis and prognostication. Breast 49:74–80PubMedCrossRef
13.
go back to reference Militello C, Rundo L, Dimarco M et al (2022) 3D DCE-MRI radiomic analysis for malignant lesion prediction in breast cancer patients. Acad Radiol 29:830–840PubMedCrossRef Militello C, Rundo L, Dimarco M et al (2022) 3D DCE-MRI radiomic analysis for malignant lesion prediction in breast cancer patients. Acad Radiol 29:830–840PubMedCrossRef
14.
go back to reference Gu J, Jiang TA (2022) Ultrasound radiomics in personalized breast management: current status and future prospects. Front Oncol 17(12):963612CrossRef Gu J, Jiang TA (2022) Ultrasound radiomics in personalized breast management: current status and future prospects. Front Oncol 17(12):963612CrossRef
15.
go back to reference Guo Y, Hu Y, Qiao M et al (2018) Radiomics analysis on ultrasound for prediction of biologic behavior in breast invasive ductal carcinoma. Clin Breast Cancer 18:e335–e344PubMedCrossRef Guo Y, Hu Y, Qiao M et al (2018) Radiomics analysis on ultrasound for prediction of biologic behavior in breast invasive ductal carcinoma. Clin Breast Cancer 18:e335–e344PubMedCrossRef
16.
go back to reference Bove S, Comes MC, Lorusso V et al (2022) A ultrasound-based radiomic approach to predict the nodal status in clinically negative breast cancer patients. Sci Rep 12:7914PubMedPubMedCentralCrossRef Bove S, Comes MC, Lorusso V et al (2022) A ultrasound-based radiomic approach to predict the nodal status in clinically negative breast cancer patients. Sci Rep 12:7914PubMedPubMedCentralCrossRef
17.
go back to reference Jiang M, Zhang D, Tang S-C et al (2021) Deep learning with convolutional neural network in the assessment of breast cancer molecular subtypes based on US images: a multicenter retrospective study. Eur Radiol 31:3673–3682PubMedCrossRef Jiang M, Zhang D, Tang S-C et al (2021) Deep learning with convolutional neural network in the assessment of breast cancer molecular subtypes based on US images: a multicenter retrospective study. Eur Radiol 31:3673–3682PubMedCrossRef
18.
go back to reference Ciritsis A, Rossi C, Eberhard M et al (2019) Automatic classification of ultrasound breast lesions using a deep convolutional neural network mimicking human decision-making. Eur Radiol 29:5458–5468PubMedCrossRef Ciritsis A, Rossi C, Eberhard M et al (2019) Automatic classification of ultrasound breast lesions using a deep convolutional neural network mimicking human decision-making. Eur Radiol 29:5458–5468PubMedCrossRef
20.
go back to reference O’Connell AM, Bartolotta TV, Orlando A et al (2022) Diagnostic performance of an artificial intelligence system in breast ultrasound. J Ultrasound Med 41:97–105PubMedCrossRef O’Connell AM, Bartolotta TV, Orlando A et al (2022) Diagnostic performance of an artificial intelligence system in breast ultrasound. J Ultrasound Med 41:97–105PubMedCrossRef
21.
go back to reference Bartolotta TV, Orlando A, Cantisani V et al (2018) Focal breast lesion characterization according to the BI-RADS US lexicon: role of a computer-aided decision-making support. Radiol Med 123:498–506PubMedCrossRef Bartolotta TV, Orlando A, Cantisani V et al (2018) Focal breast lesion characterization according to the BI-RADS US lexicon: role of a computer-aided decision-making support. Radiol Med 123:498–506PubMedCrossRef
22.
go back to reference World Medical Association (2013) World medical association declaration of Helsinki: ethical principles for medical research involving human subjects. JAMA 310:2191–2194CrossRef World Medical Association (2013) World medical association declaration of Helsinki: ethical principles for medical research involving human subjects. JAMA 310:2191–2194CrossRef
23.
go back to reference D’Orsi C, Bassett L, Feig S, Others (2018) Breast imaging reporting and data system (BI-RADS). Breast imaging atlas, 4th edn American College of Radiology, Reston D’Orsi C, Bassett L, Feig S, Others (2018) Breast imaging reporting and data system (BI-RADS). Breast imaging atlas, 4th edn American College of Radiology, Reston
24.
go back to reference Bartolotta TV, Orlando AAM, Di Vittorio ML et al (2021) S-Detect characterization of focal solid breast lesions: a prospective analysis of inter-reader agreement for US BI-RADS descriptors. J Ultrasound 24:143–150PubMedCrossRef Bartolotta TV, Orlando AAM, Di Vittorio ML et al (2021) S-Detect characterization of focal solid breast lesions: a prospective analysis of inter-reader agreement for US BI-RADS descriptors. J Ultrasound 24:143–150PubMedCrossRef
25.
go back to reference Zwanenburg A, Vallières M, Abdalah MA et al (2020) The image biomarker standardization initiative: standardized quantitative radiomics for high-throughput image-based phenotyping. Radiology 295:328–338PubMedCrossRef Zwanenburg A, Vallières M, Abdalah MA et al (2020) The image biomarker standardization initiative: standardized quantitative radiomics for high-throughput image-based phenotyping. Radiology 295:328–338PubMedCrossRef
26.
go back to reference Haralick RM, Shanmugam K, Dinstein IH (1973) Textural features for image classification. IEEE Trans Syst Man Cybernet 6:610–621CrossRef Haralick RM, Shanmugam K, Dinstein IH (1973) Textural features for image classification. IEEE Trans Syst Man Cybernet 6:610–621CrossRef
27.
go back to reference Thibault G, Angulo J, Meyer F (2014) Advanced statistical matrices for texture characterization: application to cell classification. IEEE Trans Biomed Eng 61:630–637PubMedCrossRef Thibault G, Angulo J, Meyer F (2014) Advanced statistical matrices for texture characterization: application to cell classification. IEEE Trans Biomed Eng 61:630–637PubMedCrossRef
28.
go back to reference Galloway MM (1975) Texture analysis using gray level run lengths. Comput Graph Image Process 4:172–179CrossRef Galloway MM (1975) Texture analysis using gray level run lengths. Comput Graph Image Process 4:172–179CrossRef
29.
go back to reference Sun C, Wee WG (1983) Neighboring gray level dependence matrix for texture classification. Comput Vis Graph Image Process 23:341–352CrossRef Sun C, Wee WG (1983) Neighboring gray level dependence matrix for texture classification. Comput Vis Graph Image Process 23:341–352CrossRef
30.
go back to reference Amadasun M, King R (1989) Textural features corresponding to textural properties. IEEE Trans Syst Man Cybernet 19(5):1264–1274CrossRef Amadasun M, King R (1989) Textural features corresponding to textural properties. IEEE Trans Syst Man Cybernet 19(5):1264–1274CrossRef
31.
33.
go back to reference Liang W, Luo S, Zhao G, Wu H (2020) Predicting hard rock pillar stability using GBDT, XGBoost, and LightGBM algorithms. Sci China Ser A Math 8:765 Liang W, Luo S, Zhao G, Wu H (2020) Predicting hard rock pillar stability using GBDT, XGBoost, and LightGBM algorithms. Sci China Ser A Math 8:765
34.
go back to reference Shwartz-Ziv R, Armon A (2022) Tabular data: deep learning is not all you need. Inf Fusion 81:84–90CrossRef Shwartz-Ziv R, Armon A (2022) Tabular data: deep learning is not all you need. Inf Fusion 81:84–90CrossRef
36.
go back to reference Menze BH, Kelm BM, Masuch R et al (2009) A comparison of random forest and its Gini importance with standard chemometric methods for the feature selection and classification of spectral data. BMC Bioinformatics 10:213PubMedPubMedCentralCrossRef Menze BH, Kelm BM, Masuch R et al (2009) A comparison of random forest and its Gini importance with standard chemometric methods for the feature selection and classification of spectral data. BMC Bioinformatics 10:213PubMedPubMedCentralCrossRef
38.
39.
go back to reference Wei Q, Yan Y-J, Wu G-G et al (2022) The diagnostic performance of ultrasound computer-aided diagnosis system for distinguishing breast masses: a prospective multicenter study. Eur Radiol 32:4046–4055PubMedCrossRef Wei Q, Yan Y-J, Wu G-G et al (2022) The diagnostic performance of ultrasound computer-aided diagnosis system for distinguishing breast masses: a prospective multicenter study. Eur Radiol 32:4046–4055PubMedCrossRef
40.
go back to reference Dong F, She R, Cui C et al (2021) One step further into the blackbox: a pilot study of how to build more confidence around an AI-based decision system of breast nodule assessment in 2D ultrasound. Eur Radiol 31:4991–5000PubMedCrossRef Dong F, She R, Cui C et al (2021) One step further into the blackbox: a pilot study of how to build more confidence around an AI-based decision system of breast nodule assessment in 2D ultrasound. Eur Radiol 31:4991–5000PubMedCrossRef
41.
go back to reference Li J-W, Cao Y-C, Zhao Z-J et al (2022) Prediction for pathological and immunohistochemical characteristics of triple-negative invasive breast carcinomas: the performance comparison between quantitative and qualitative sonographic feature analysis. Eur Radiol 32:1590–1600PubMedCrossRef Li J-W, Cao Y-C, Zhao Z-J et al (2022) Prediction for pathological and immunohistochemical characteristics of triple-negative invasive breast carcinomas: the performance comparison between quantitative and qualitative sonographic feature analysis. Eur Radiol 32:1590–1600PubMedCrossRef
42.
go back to reference Jiang M, Li C-L, Chen R-X et al (2021) Management of breast lesions seen on US images: dual-model radiomics including shear-wave elastography may match performance of expert radiologists. Eur J Radiol 141:109781PubMedCrossRef Jiang M, Li C-L, Chen R-X et al (2021) Management of breast lesions seen on US images: dual-model radiomics including shear-wave elastography may match performance of expert radiologists. Eur J Radiol 141:109781PubMedCrossRef
43.
go back to reference Qian X, Pei J, Zheng H et al (2021) Prospective assessment of breast cancer risk from multimodal multiview ultrasound images via clinically applicable deep learning. Nat Biomed Eng 5:522–532PubMedCrossRef Qian X, Pei J, Zheng H et al (2021) Prospective assessment of breast cancer risk from multimodal multiview ultrasound images via clinically applicable deep learning. Nat Biomed Eng 5:522–532PubMedCrossRef
44.
go back to reference Kapetas P, Clauser P, Woitek R et al (2019) Quantitative multiparametric breast ultrasound: application of contrast-enhanced ultrasound and elastography leads to an improved differentiation of benign and malignant lesions. Invest Radiol 54:257–264PubMedPubMedCentralCrossRef Kapetas P, Clauser P, Woitek R et al (2019) Quantitative multiparametric breast ultrasound: application of contrast-enhanced ultrasound and elastography leads to an improved differentiation of benign and malignant lesions. Invest Radiol 54:257–264PubMedPubMedCentralCrossRef
45.
go back to reference Shen Y, Shamout FE, Oliver JR et al (2021) Artificial intelligence system reduces false-positive findings in the interpretation of breast ultrasound exams. Nat Commun 12:5645PubMedPubMedCentralCrossRef Shen Y, Shamout FE, Oliver JR et al (2021) Artificial intelligence system reduces false-positive findings in the interpretation of breast ultrasound exams. Nat Commun 12:5645PubMedPubMedCentralCrossRef
46.
go back to reference Romeo V, Cuocolo R, Apolito R et al (2021) Clinical value of radiomics and machine learning in breast ultrasound: a multicenter study for differential diagnosis of benign and malignant lesions. Eur Radiol 31:9511–9519PubMedPubMedCentralCrossRef Romeo V, Cuocolo R, Apolito R et al (2021) Clinical value of radiomics and machine learning in breast ultrasound: a multicenter study for differential diagnosis of benign and malignant lesions. Eur Radiol 31:9511–9519PubMedPubMedCentralCrossRef
47.
go back to reference Gu Y, Xu W, Liu T et al (2023) Ultrasound-based deep learning in the establishment of a breast lesion risk stratification system: a multicenter study. Eur Radiol 33:2954–2964PubMedCrossRef Gu Y, Xu W, Liu T et al (2023) Ultrasound-based deep learning in the establishment of a breast lesion risk stratification system: a multicenter study. Eur Radiol 33:2954–2964PubMedCrossRef
48.
go back to reference Dietzel M, Clauser P, Kapetas P et al (2021) Images are data: a breast imaging perspective on a contemporary paradigm. Rofo 193:898–908PubMedCrossRef Dietzel M, Clauser P, Kapetas P et al (2021) Images are data: a breast imaging perspective on a contemporary paradigm. Rofo 193:898–908PubMedCrossRef
49.
go back to reference Rodriguez-Ruiz A, Lång K, Gubern-Merida A et al (2019) Stand-alone artificial intelligence for breast cancer detection in mammography: comparison with 101 radiologists. J Natl Cancer Inst 111:916–922PubMedPubMedCentralCrossRef Rodriguez-Ruiz A, Lång K, Gubern-Merida A et al (2019) Stand-alone artificial intelligence for breast cancer detection in mammography: comparison with 101 radiologists. J Natl Cancer Inst 111:916–922PubMedPubMedCentralCrossRef
50.
go back to reference Stelzer PD, Steding O, Raudner MW et al (2020) Combined texture analysis and machine learning in suspicious calcifications detected by mammography: potential to avoid unnecessary stereotactical biopsies. Eur J Radiol 132:109309PubMedCrossRef Stelzer PD, Steding O, Raudner MW et al (2020) Combined texture analysis and machine learning in suspicious calcifications detected by mammography: potential to avoid unnecessary stereotactical biopsies. Eur J Radiol 132:109309PubMedCrossRef
51.
go back to reference Dietzel M, Schulz-Wendtland R, Ellmann S et al (2020) Automated volumetric radiomic analysis of breast cancer vascularization improves survival prediction in primary breast cancer. Sci Rep 10:3664PubMedPubMedCentralCrossRef Dietzel M, Schulz-Wendtland R, Ellmann S et al (2020) Automated volumetric radiomic analysis of breast cancer vascularization improves survival prediction in primary breast cancer. Sci Rep 10:3664PubMedPubMedCentralCrossRef
52.
go back to reference Pötsch N, Dietzel M, Kapetas P et al (2021) An A.I. Classifier derived from 4D radiomics of dynamic contrast-enhanced breast MRI data: potential to avoid unnecessary breast biopsies. Eur Radiol 31:5866–5876PubMedPubMedCentralCrossRef Pötsch N, Dietzel M, Kapetas P et al (2021) An A.I. Classifier derived from 4D radiomics of dynamic contrast-enhanced breast MRI data: potential to avoid unnecessary breast biopsies. Eur Radiol 31:5866–5876PubMedPubMedCentralCrossRef
53.
go back to reference Tahmassebi A, Wengert GJ, Helbich TH et al (2019) Impact of machine learning with multiparametric magnetic resonance imaging of the breast for early prediction of response to neoadjuvant chemotherapy and survival outcomes in breast cancer patients. Invest Radiol 54:110–117PubMedPubMedCentralCrossRef Tahmassebi A, Wengert GJ, Helbich TH et al (2019) Impact of machine learning with multiparametric magnetic resonance imaging of the breast for early prediction of response to neoadjuvant chemotherapy and survival outcomes in breast cancer patients. Invest Radiol 54:110–117PubMedPubMedCentralCrossRef
54.
go back to reference Mao L, Chen H, Liang M et al (2019) Quantitative radiomic model for predicting malignancy of small solid pulmonary nodules detected by low-dose CT screening. Quant Imaging Med Surg 9:263–272PubMedPubMedCentralCrossRef Mao L, Chen H, Liang M et al (2019) Quantitative radiomic model for predicting malignancy of small solid pulmonary nodules detected by low-dose CT screening. Quant Imaging Med Surg 9:263–272PubMedPubMedCentralCrossRef
55.
go back to reference Kamiya A, Murayama S, Kamiya H et al (2014) Kurtosis and skewness assessments of solid lung nodule density histograms: differentiating malignant from benign nodules on CT. Jpn J Radiol 32:14–21PubMedCrossRef Kamiya A, Murayama S, Kamiya H et al (2014) Kurtosis and skewness assessments of solid lung nodule density histograms: differentiating malignant from benign nodules on CT. Jpn J Radiol 32:14–21PubMedCrossRef
Metadata
Title
Artificial intelligence-based, semi-automated segmentation for the extraction of ultrasound-derived radiomics features in breast cancer: a prospective multicenter study
Authors
Tommaso Vincenzo Bartolotta
Carmelo Militello
Francesco Prinzi
Fabiola Ferraro
Leonardo Rundo
Calogero Zarcaro
Mariangela Dimarco
Alessia Angela Maria Orlando
Domenica Matranga
Salvatore Vitabile
Publication date
09-05-2024
Publisher
Springer Milan
Published in
La radiologia medica / Issue 7/2024
Print ISSN: 0033-8362
Electronic ISSN: 1826-6983
DOI
https://doi.org/10.1007/s11547-024-01826-7