1
|
Nowakowska S, Borkowski K, Ruppert C, Hejduk P, Ciritsis A, Landsmann A, Marcon M, Berger N, Boss A, Rossi C. Explainable Precision Medicine in Breast MRI: A Combined Radiomics and Deep Learning Approach for the Classification of Contrast Agent Uptake. Bioengineering (Basel) 2024; 11:556. [PMID: 38927793 PMCID: PMC11200390 DOI: 10.3390/bioengineering11060556] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/15/2024] [Revised: 05/24/2024] [Accepted: 05/29/2024] [Indexed: 06/28/2024] Open
Abstract
In DCE-MRI, the degree of contrast uptake in normal fibroglandular tissue, i.e., background parenchymal enhancement (BPE), is a crucial biomarker linked to breast cancer risk and treatment outcome. In accordance with the Breast Imaging Reporting & Data System (BI-RADS), it should be visually classified into four classes. The susceptibility of such an assessment to inter-reader variability highlights the urgent need for a standardized classification algorithm. In this retrospective study, the first post-contrast subtraction images for 27 healthy female subjects were included. The BPE was classified slice-wise by two expert radiologists. The extraction of radiomic features from segmented BPE was followed by dataset splitting and dimensionality reduction. The latent representations were then utilized as inputs to a deep neural network classifying BPE into BI-RADS classes. The network's predictions were elucidated at the radiomic feature level with Shapley values. The deep neural network achieved a BPE classification accuracy of 84 ± 2% (p-value < 0.00001). Most of the misclassifications involved adjacent classes. Different radiomic features were decisive for the prediction of each BPE class underlying the complexity of the decision boundaries. A highly precise and explainable pipeline for BPE classification was achieved without user- or algorithm-dependent radiomic feature selection.
Collapse
Affiliation(s)
- Sylwia Nowakowska
- Diagnostic and Interventional Radiology, University Hospital Zürich, University Zürich, Rämistrasse 100, 8091 Zürich, Switzerland (C.R.)
| | | | - Carlotta Ruppert
- Diagnostic and Interventional Radiology, University Hospital Zürich, University Zürich, Rämistrasse 100, 8091 Zürich, Switzerland (C.R.)
- b-rayZ AG, Wagistrasse 21, 8952 Schlieren, Switzerland
| | - Patryk Hejduk
- Diagnostic and Interventional Radiology, University Hospital Zürich, University Zürich, Rämistrasse 100, 8091 Zürich, Switzerland (C.R.)
| | - Alexander Ciritsis
- Diagnostic and Interventional Radiology, University Hospital Zürich, University Zürich, Rämistrasse 100, 8091 Zürich, Switzerland (C.R.)
- b-rayZ AG, Wagistrasse 21, 8952 Schlieren, Switzerland
| | - Anna Landsmann
- Diagnostic and Interventional Radiology, University Hospital Zürich, University Zürich, Rämistrasse 100, 8091 Zürich, Switzerland (C.R.)
| | - Magda Marcon
- Diagnostic and Interventional Radiology, University Hospital Zürich, University Zürich, Rämistrasse 100, 8091 Zürich, Switzerland (C.R.)
| | - Nicole Berger
- Diagnostic and Interventional Radiology, University Hospital Zürich, University Zürich, Rämistrasse 100, 8091 Zürich, Switzerland (C.R.)
| | - Andreas Boss
- Diagnostic and Interventional Radiology, University Hospital Zürich, University Zürich, Rämistrasse 100, 8091 Zürich, Switzerland (C.R.)
| | - Cristina Rossi
- Diagnostic and Interventional Radiology, University Hospital Zürich, University Zürich, Rämistrasse 100, 8091 Zürich, Switzerland (C.R.)
- b-rayZ AG, Wagistrasse 21, 8952 Schlieren, Switzerland
| |
Collapse
|
2
|
Heinze S, Rudnicki WK, Paluchowska J, Szpor J, Łuczyńska E. Enhancing diagnostic precision: comparative analysis of MR-guided breast biopsies performed in two centres. Pol J Radiol 2024; 89:e235-e239. [PMID: 38938661 PMCID: PMC11210378 DOI: 10.5114/pjr/186862] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/11/2024] [Accepted: 04/05/2024] [Indexed: 06/29/2024] Open
Abstract
Purpose Breast lesions that remain elusive in traditional imaging techniques such as ultrasound and mammography pose a diagnostic challenge. In such cases, magnetic resonance (MR)-guided breast biopsy emerges as a crucial tool for accurate histopathological verification. This article presents a comparative study conducted at 2 centres, exploring the results of MR-guided breast biopsies performed by experienced radiologists, based on inside and external referrals. Material and methods The study involved 228 patients, 120 of whom underwent biopsies at Centre 1, where the same radiologist performed both the qualification and biopsy. The remaining 108 patients were biopsied at Centre 2, based on referrals from different institutions. Uniform examination protocols were adopted at both centres, and all biopsies underwent histopathological verification. Results The distribution of lesion types was found to be independent of the apparatus used for biopsies (p = 0.759). Interestingly, Centre 1 exhibited a higher prevalence of infiltrating carcinomas compared to Centre 2 (p = 0.12). Furthermore, the analysis demonstrated a significant variance in the nature of the lesions in relation to breast structure and biopsy centre (p < 0.001). Conclusions MR-guided breast biopsy serves as a remarkable tool for verifying lesions that evade detection through conventional imaging methods and physical examinations. The study findings underscore the crucial role of radiologist experience in determining the efficacy of MR-guided breast biopsies.
Collapse
Affiliation(s)
- Sylwia Heinze
- Department of Radiology, Maria Sklodowska-Curie Memorial Cancer Centre and Institute of Oncology, Cracow Branch, Cracow, Poland
| | | | | | - Joanna Szpor
- Department of Pathomorphology, Jagiellonian University Medical College, Cracow, Poland
| | - Elżbieta Łuczyńska
- Department of Electroradiology, Jagiellonian University Medical College, Cracow, Poland
| |
Collapse
|
3
|
Breast MRI: Clinical Indications, Recommendations, and Future Applications in Breast Cancer Diagnosis. Curr Oncol Rep 2023; 25:257-267. [PMID: 36749493 DOI: 10.1007/s11912-023-01372-x] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Accepted: 11/28/2022] [Indexed: 02/08/2023]
Abstract
PURPOSE OF REVIEW This article aims to provide an updated overview of the indications for diagnostic breast magnetic resonance imaging (MRI), discusses the available and novel imaging exams proposed for breast cancer detection, and discusses considerations when performing breast MRI in the clinical setting. RECENT FINDINGS Breast MRI is superior in identifying lesions in women with a very high risk of breast cancer or average risk with dense breasts. Moreover, the application of breast MRI has benefits in numerous other clinical cases as well; e.g., the assessment of the extent of disease, evaluation of response to neoadjuvant therapy (NAT), evaluation of lymph nodes and primary occult tumor, evaluation of lesions suspicious of Paget's disease, and suspicious discharge and breast implants. Breast cancer is the most frequently detected tumor among women around the globe and is often diagnosed as a result of abnormal findings on mammography. Although effective multimodal therapies significantly decline mortality rates, breast cancer remains one of the leading causes of cancer death. A proactive approach to identifying suspicious breast lesions at early stages can enhance the efficacy of anti-cancer treatments, improve patient recovery, and significantly improve long-term survival. However, the currently applied mammography to detect breast cancer has its limitations. High false-positive and false-negative rates are observed in women with dense breasts. Since approximately half of the screening population comprises women with dense breasts, mammography is often incorrectly used. The application of breast MRI should significantly impact the correct cases of breast abnormality detection in women. Radiomics provides valuable data obtained from breast MRI, further improving breast cancer diagnosis. Introducing these constantly evolving algorithms in clinical practice will lead to the right breast detection tool, optimized surveillance program, and individualized breast cancer treatment.
Collapse
|
4
|
Iacob R, Manolescu DL, Stoicescu ER, Fabian A, Malita D, Oancea C. Breast Cancer—How Can Imaging Help? Healthcare (Basel) 2022; 10:healthcare10071159. [PMID: 35885686 PMCID: PMC9323053 DOI: 10.3390/healthcare10071159] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/25/2022] [Revised: 06/20/2022] [Accepted: 06/21/2022] [Indexed: 11/16/2022] Open
Abstract
Breast cancer is the most common malignant disease among women, causing death and suffering worldwide. It is known that, for the improvement of the survival rate and the psychological impact it has on patients, early detection is crucial. For this to happen, the imaging techniques should be used at their full potential. We selected and examined 44 articles that had as subject the use of a specific imaging method in breast cancer management (mammography, ultrasound, MRI, ultrasound-guided biopsy, PET-CT). After analyzing their data, we summarized and concluded which are the best ways to use each one of the mentioned techniques for a good outcome. We created a simplified algorithm with easy steps that can be followed by radiologists when facing this type of neoplasia.
Collapse
Affiliation(s)
- Roxana Iacob
- Department of Radiology and Medical Imaging, ‘Victor Babes’ University of Medicine and Pharmacy Timisoara, 300041 Timișoara, Romania; (R.I.); (E.R.S.); (A.F.); (D.M.)
| | - Diana Luminita Manolescu
- Department of Radiology and Medical Imaging, ‘Victor Babes’ University of Medicine and Pharmacy Timisoara, 300041 Timișoara, Romania; (R.I.); (E.R.S.); (A.F.); (D.M.)
- Center for Research and Innovation in Precision Medicine of Respiratory Diseases (CRIPMRD), ‘Victor Babeș’ University of Medicine and Pharmacy, 300041 Timișoara, Romania;
- Correspondence:
| | - Emil Robert Stoicescu
- Department of Radiology and Medical Imaging, ‘Victor Babes’ University of Medicine and Pharmacy Timisoara, 300041 Timișoara, Romania; (R.I.); (E.R.S.); (A.F.); (D.M.)
- Research Center for Pharmaco-Toxicological Evaluations, ‘Victor Babes’ University of Medicine and Pharmacy Timisoara, 300041 Timișoara, Romania
| | - Antonio Fabian
- Department of Radiology and Medical Imaging, ‘Victor Babes’ University of Medicine and Pharmacy Timisoara, 300041 Timișoara, Romania; (R.I.); (E.R.S.); (A.F.); (D.M.)
| | - Daniel Malita
- Department of Radiology and Medical Imaging, ‘Victor Babes’ University of Medicine and Pharmacy Timisoara, 300041 Timișoara, Romania; (R.I.); (E.R.S.); (A.F.); (D.M.)
| | - Cristian Oancea
- Center for Research and Innovation in Precision Medicine of Respiratory Diseases (CRIPMRD), ‘Victor Babeș’ University of Medicine and Pharmacy, 300041 Timișoara, Romania;
- Department of Pulmonology, ‘Victor Babes’ University of Medicine and Pharmacy, 300041 Timișoara, Romania
| |
Collapse
|
5
|
de Faria Castro Fleury E, Castro C, do Amaral MSC, Roveda Junior D. Management of Non-Mass Enhancement at Breast Magnetic Resonance in Screening Settings Referred for Magnetic Resonance-Guided Biopsy. BREAST CANCER: BASIC AND CLINICAL RESEARCH 2022; 16:11782234221095897. [PMID: 35602239 PMCID: PMC9118420 DOI: 10.1177/11782234221095897] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/06/2021] [Accepted: 03/30/2022] [Indexed: 11/17/2022] Open
Abstract
Rationale and Objectives According to the Breast Imaging and Reporting Data System (BI-RADS), one of the main limitations of MRI is diagnosing the non-mass enhancement (NME). The NME lesion is challenging since it is unique to the MRI lexicon. This study aims to report our experience with NME lesions diagnosed by MRI referred for MRI-guided biopsies and discuss the management and follow-up of these lesions. Materials and Methods We retrospectively evaluated all MRI-guide breast biopsies. We included all patients referred for NME breast MRI-guided biopsy in screening settings. All patients had a negative second-look mammography or ultrasonography. We correlated the distribution and internal enhancement pattern (IEP) of the NME lesions with histology. Invasive ductal carcinomas (IDC) of no special type and ductal carcinoma in situ (DCIS) were considered malignant lesions. Results From January-2018 to July-2021, we included 96 women with a total of 96 lesions in the study. There were 90 benign and 6 malignant lesions with DCIS prevalence (5/6 cancers). The most frequent benign lesion type was fibrocystic changes. There were no NME lesions with diffuse or multiple area distribution features referred to MRI-guided biopsy. The positive-predictive values (PPV) were respectively 0.0%, 2.5%, 9.0%, and 11.0% for linear, focal, regional, and segmental distribution describers, and 0.0, 3.0%, 7.9%, and 50% for homogenous, heterogeneous, clumped, and clustered-ring enhancement patterns. Conclusion We observe the high potential risk for malignancy in the clustered-ring enhancement followed by the clumped pattern. Segmental distribution presented the highest predictive-positive values.
Collapse
Affiliation(s)
| | - Caio Castro
- Department of Radiology, Femme-Laboratório da Mulher, São Paulo, Brazil
| | | | | |
Collapse
|
6
|
Nissan N, Bauer E, Moss Massasa EE, Sklair-Levy M. Breast MRI during pregnancy and lactation: clinical challenges and technical advances. Insights Imaging 2022; 13:71. [PMID: 35397082 PMCID: PMC8994812 DOI: 10.1186/s13244-022-01214-7] [Citation(s) in RCA: 11] [Impact Index Per Article: 5.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/28/2022] [Accepted: 03/21/2022] [Indexed: 12/12/2022] Open
Abstract
The breast experiences substantial changes in morphology and function during pregnancy and lactation which affects its imaging properties and may reduce the visibility of a concurrent pathological process. The high incidence of benign gestational-related entities may further add complexity to the clinical and radiological evaluation of the breast during the period. Consequently, pregnancy-associated breast cancer (PABC) is often a delayed diagnosis and carries a poor prognosis. This state-of-the-art pictorial review illustrates how despite currently being underutilized, technical advances and new clinical evidence support the use of unenhanced breast MRI during pregnancy and both unenhanced and dynamic-contrast enhanced (DCE) during lactation, to serve as effective supplementary modalities in the diagnostic work-up of PABC.
Collapse
Affiliation(s)
- Noam Nissan
- Radiology Department, Sheba Medical Center, 5265601, Tel Hashomer, Israel.
- Sackler Medicine School, Tel Aviv University, Tel Aviv, Israel.
| | - Ethan Bauer
- Sackler Medicine School, New-York Program, Tel Aviv University, Tel Aviv, Israel
| | - Efi Efraim Moss Massasa
- Joint Medicine School Program of Sheba Medical Center, St. George's, University of London and the University of Nicosia, Sheba Medical Center, Tel Hashomer, Israel
| | - Miri Sklair-Levy
- Radiology Department, Sheba Medical Center, 5265601, Tel Hashomer, Israel
- Sackler Medicine School, Tel Aviv University, Tel Aviv, Israel
| |
Collapse
|
7
|
MRI of the Lactating Breast: Computer-Aided Diagnosis False Positive Rates and Background Parenchymal Enhancement Kinetic Features. Acad Radiol 2021; 29:1332-1341. [PMID: 34857455 DOI: 10.1016/j.acra.2021.11.003] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/30/2021] [Revised: 11/01/2021] [Accepted: 11/01/2021] [Indexed: 12/28/2022]
Abstract
RATIONALE AND OBJECTIVES To investigate the application of computer-added diagnosis (CAD) in dynamic contrast-enhanced (DCE) MRI of the healthy lactating breast, focusing on false-positive rates and background parenchymal enhancement (BPE) coloring patterns in comparison with breast cancer features in non-lactating patients. MATERIALS AND METHODS The study population was composed of 58 healthy lactating patients and control groups of 113 healthy premenopausal non-lactating patients and 55 premenopausal non-lactating patients with newly-diagnosed breast cancer. Patients were scanned on 1.5-T MRI using conventional DCE protocol. A retrospective analysis of DCE-derived CAD properties was conducted using a commercial software that is regularly utilized in our routine radiological work-up. Qualitative morphological characterization and automatically-obtained quantitative parametric measurements of the BPE-induced CAD coloring were categorized and subgroups' trends and differences between the lactating and cancer cohorts were statistically assessed. RESULTS CAD false-positive coloring was found in the majority of lactating cases (87%). Lactation BPE coloring was characteristically non-mass enhancement (NME)-like shaped (87%), bilateral (79%) and symmetric (64%), whereas, unilateral coloring was associated with prior irradiation (p <0.0001). Inter-individual variability in CAD appearance of both scoring-grade and kinetic-curve dominance was found among the lactating cohort. When compared with healthy non-lactating controls, CAD false positive probability was significantly increased [Odds ratio 40.2, p <0001], while in comparison with the breast cancer cohort, CAD features were mostly inconclusive, even though increased size parameters were significantly associated with lactation-BPE (p <0.00001). CONCLUSION BPE was identified as a common source for false-positive CAD coloring on breast DCE-MRI among lactating population. Despite several typical characteristics, overlapping features with breast malignancy warrant a careful evaluation and clinical correlation in all cases with suspected lactation induced CAD coloring.
Collapse
|