1
|
Marasinou C, Li B, Paige J, Omigbodun A, Nakhaei N, Hoyt A, Hsu W. Improving the Quantitative Analysis of Breast Microcalcifications: A Multiscale Approach. J Digit Imaging 2023; 36:1016-1028. [PMID: 36820930 PMCID: PMC10287598 DOI: 10.1007/s10278-022-00751-3] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/23/2021] [Revised: 12/04/2022] [Accepted: 12/06/2022] [Indexed: 02/24/2023] Open
Abstract
Accurate characterization of microcalcifications (MCs) in 2D digital mammography is a necessary step toward reducing the diagnostic uncertainty associated with the callback of indeterminate MCs. Quantitative analysis of MCs can better identify MCs with a higher likelihood of ductal carcinoma in situ or invasive cancer. However, automated identification and segmentation of MCs remain challenging with high false positive rates. We present a two-stage multiscale approach to MC segmentation in 2D full-field digital mammograms (FFDMs) and diagnostic magnification views. Candidate objects are first delineated using blob detection and Hessian analysis. A regression convolutional network, trained to output a function with a higher response near MCs, chooses the objects which constitute actual MCs. The method was trained and validated on 435 screening and diagnostic FFDMs from two separate datasets. We then used our approach to segment MCs on magnification views of 248 cases with amorphous MCs. We modeled the extracted features using gradient tree boosting to classify each case as benign or malignant. Compared to state-of-the-art comparison methods, our approach achieved superior mean intersection over the union (0.670 ± 0.121 per image versus 0.524 ± 0.034 per image), intersection over the union per MC object (0.607 ± 0.250 versus 0.363 ± 0.278) and true positive rate of 0.744 versus 0.581 at 0.4 false positive detections per square centimeter. Features generated using our approach outperformed the comparison method (0.763 versus 0.710 AUC) in distinguishing amorphous calcifications as benign or malignant.
Collapse
Affiliation(s)
- Chrysostomos Marasinou
- Medical & Imaging Informatics, Department of Radiological Sciences, David Geffen School of Medicine at UCLA, 924 Westwood Blvd, Ste 420, Los Angeles, 90024, USA
| | - Bo Li
- Department of Radiological Sciences, David Geffen School of Medicine at UCLA, Los Angeles, 90095, CA, USA
| | - Jeremy Paige
- Department of Radiological Sciences, David Geffen School of Medicine at UCLA, Los Angeles, 90095, CA, USA
| | - Akinyinka Omigbodun
- Medical & Imaging Informatics, Department of Radiological Sciences, David Geffen School of Medicine at UCLA, 924 Westwood Blvd, Ste 420, Los Angeles, 90024, USA
| | - Noor Nakhaei
- Department of Computer Science, UCLA Samueli School of Engineering, Los Angeles, 90095, CA, USA
| | - Anne Hoyt
- Department of Radiological Sciences, David Geffen School of Medicine at UCLA, Los Angeles, 90095, CA, USA
| | - William Hsu
- Medical & Imaging Informatics, Department of Radiological Sciences, David Geffen School of Medicine at UCLA, 924 Westwood Blvd, Ste 420, Los Angeles, 90024, USA.
| |
Collapse
|
2
|
Konz N, Buda M, Gu H, Saha A, Yang J, Chłędowski J, Park J, Witowski J, Geras KJ, Shoshan Y, Gilboa-Solomon F, Khapun D, Ratner V, Barkan E, Ozery-Flato M, Martí R, Omigbodun A, Marasinou C, Nakhaei N, Hsu W, Sahu P, Hossain MB, Lee J, Santos C, Przelaskowski A, Kalpathy-Cramer J, Bearce B, Cha K, Farahani K, Petrick N, Hadjiiski L, Drukker K, Armato SG, Mazurowski MA. A Competition, Benchmark, Code, and Data for Using Artificial Intelligence to Detect Lesions in Digital Breast Tomosynthesis. JAMA Netw Open 2023; 6:e230524. [PMID: 36821110 PMCID: PMC9951043 DOI: 10.1001/jamanetworkopen.2023.0524] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 02/24/2023] Open
Abstract
IMPORTANCE An accurate and robust artificial intelligence (AI) algorithm for detecting cancer in digital breast tomosynthesis (DBT) could significantly improve detection accuracy and reduce health care costs worldwide. OBJECTIVES To make training and evaluation data for the development of AI algorithms for DBT analysis available, to develop well-defined benchmarks, and to create publicly available code for existing methods. DESIGN, SETTING, AND PARTICIPANTS This diagnostic study is based on a multi-institutional international grand challenge in which research teams developed algorithms to detect lesions in DBT. A data set of 22 032 reconstructed DBT volumes was made available to research teams. Phase 1, in which teams were provided 700 scans from the training set, 120 from the validation set, and 180 from the test set, took place from December 2020 to January 2021, and phase 2, in which teams were given the full data set, took place from May to July 2021. MAIN OUTCOMES AND MEASURES The overall performance was evaluated by mean sensitivity for biopsied lesions using only DBT volumes with biopsied lesions; ties were broken by including all DBT volumes. RESULTS A total of 8 teams participated in the challenge. The team with the highest mean sensitivity for biopsied lesions was the NYU B-Team, with 0.957 (95% CI, 0.924-0.984), and the second-place team, ZeDuS, had a mean sensitivity of 0.926 (95% CI, 0.881-0.964). When the results were aggregated, the mean sensitivity for all submitted algorithms was 0.879; for only those who participated in phase 2, it was 0.926. CONCLUSIONS AND RELEVANCE In this diagnostic study, an international competition produced algorithms with high sensitivity for using AI to detect lesions on DBT images. A standardized performance benchmark for the detection task using publicly available clinical imaging data was released, with detailed descriptions and analyses of submitted algorithms accompanied by a public release of their predictions and code for selected methods. These resources will serve as a foundation for future research on computer-assisted diagnosis methods for DBT, significantly lowering the barrier of entry for new researchers.
Collapse
Affiliation(s)
- Nicholas Konz
- Department of Electrical and Computer Engineering, Duke University, Durham, North Carolina
| | - Mateusz Buda
- Department of Radiology, Duke University Medical Center, Durham, North Carolina
- Faculty of Mathematics and Information Science, Warsaw University of Technology, Warsaw, Poland
| | - Hanxue Gu
- Department of Electrical and Computer Engineering, Duke University, Durham, North Carolina
| | - Ashirbani Saha
- Department of Radiology, Duke University Medical Center, Durham, North Carolina
- Department of Oncology, McMaster University, Hamilton, Ontario, Canada
| | | | - Jakub Chłędowski
- Jagiellonian University, Kraków, Poland
- Department of Radiology, NYU Grossman School of Medicine, New York, New York
| | - Jungkyu Park
- Department of Radiology, NYU Grossman School of Medicine, New York, New York
| | - Jan Witowski
- Department of Radiology, NYU Grossman School of Medicine, New York, New York
| | - Krzysztof J. Geras
- Department of Radiology, NYU Grossman School of Medicine, New York, New York
| | - Yoel Shoshan
- Medical Image Analytics, IBM Research, Haifa, Israel
| | | | - Daniel Khapun
- Medical Image Analytics, IBM Research, Haifa, Israel
| | - Vadim Ratner
- Medical Image Analytics, IBM Research, Haifa, Israel
| | - Ella Barkan
- Medical Image Analytics, IBM Research, Haifa, Israel
| | | | - Robert Martí
- Institute of Computer Vision and Robotics, University of Girona, Girona, Spain
| | - Akinyinka Omigbodun
- Medical and Imaging Informatics Group, Department of Radiological Sciences, David Geffen School of Medicine, University of California Los Angeles
| | - Chrysostomos Marasinou
- Medical and Imaging Informatics Group, Department of Radiological Sciences, David Geffen School of Medicine, University of California Los Angeles
| | - Noor Nakhaei
- Medical and Imaging Informatics Group, Department of Radiological Sciences, David Geffen School of Medicine, University of California Los Angeles
| | - William Hsu
- Medical and Imaging Informatics Group, Department of Radiological Sciences, David Geffen School of Medicine, University of California Los Angeles
- Department of Radiological Sciences, David Geffen School of Medicine, University of California Los Angeles
- Department of Bioengineering, University of California Los Angeles Samueli School of Engineering
| | - Pranjal Sahu
- Department of Computer Science, Stony Brook University, Stony Brook, New York
| | - Md Belayat Hossain
- Department of Radiology, University of Pittsburgh, Pittsburgh, Pennsylvania
| | - Juhun Lee
- Department of Radiology, University of Pittsburgh, Pittsburgh, Pennsylvania
| | - Carlos Santos
- Department of Radiology, Duke University Medical Center, Durham, North Carolina
| | - Artur Przelaskowski
- Faculty of Mathematics and Information Science, Warsaw University of Technology, Warsaw, Poland
| | - Jayashree Kalpathy-Cramer
- Department of Radiology, Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown
| | - Benjamin Bearce
- Department of Radiology, Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown
| | - Kenny Cha
- US Food and Drug Administration, Silver Spring, Maryland
| | - Keyvan Farahani
- Center for Biomedical Informatics and Information Technology, National Cancer Institute, Bethesda, Maryland
| | | | | | - Karen Drukker
- Department of Radiology, University of Chicago, Chicago, Illinois
| | - Samuel G. Armato
- Department of Radiology, University of Chicago, Chicago, Illinois
| | - Maciej A. Mazurowski
- Department of Electrical and Computer Engineering, Duke University, Durham, North Carolina
- Department of Radiology, Duke University Medical Center, Durham, North Carolina
- Department of Computer Science, Duke University, Durham, North Carolina
- Department of Biostatistics and Bioinformatics, Duke University Medical Center, Durham, North Carolina
| |
Collapse
|
3
|
Marathe K, Marasinou C, Li B, Nakhaei N, Li B, Elmore JG, Shapiro L, Hsu W. Automated quantitative assessment of amorphous calcifications: Towards improved malignancy risk stratification. Comput Biol Med 2022; 146:105504. [PMID: 35525068 PMCID: PMC9839357 DOI: 10.1016/j.compbiomed.2022.105504] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/11/2021] [Revised: 03/07/2022] [Accepted: 04/05/2022] [Indexed: 01/17/2023]
Abstract
BACKGROUND Amorphous calcifications noted on mammograms (i.e., small and indistinct calcifications that are difficult to characterize) are associated with high diagnostic uncertainty, often leading to biopsies. Yet, only 20% of biopsied amorphous calcifications are cancer. We present a quantitative approach for distinguishing between benign and actionable (high-risk and malignant) amorphous calcifications using a combination of local textures, global spatial relationships, and interpretable handcrafted expert features. METHOD Our approach was trained and validated on a set of 168 2D full-field digital mammography exams (248 images) from 168 patients. Within these 248 images, we identified 276 image regions with segmented amorphous calcifications and a biopsy-confirmed diagnosis. A set of local (radiomic and region measurements) and global features (distribution and expert-defined) were extracted from each image. Local features were grouped using an unsupervised k-means clustering algorithm. All global features were concatenated with clustered local features and used to train a LightGBM classifier to distinguish benign from actionable cases. RESULTS On the held-out test set of 60 images, our approach achieved a sensitivity of 100%, specificity of 35%, and a positive predictive value of 38% when the decision threshold was set to 0.4. Given that all of the images in our test set resulted in a recommendation of a biopsy, the use of our algorithm would have identified 15 images (25%) that were benign, potentially reducing the number of breast biopsies. CONCLUSIONS Quantitative analysis of full-field digital mammograms can extract subtle shape, texture, and distribution features that may help to distinguish between benign and actionable amorphous calcifications.
Collapse
Affiliation(s)
- Kalyani Marathe
- Department of Electrical and Computer Engineering, University of Washington, Seattle, WA, USA
| | - Chrysostomos Marasinou
- Medical & Imaging Informatics, Department of Radiological Sciences, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA
| | - Beibin Li
- Paul G. Allen School of Computer Science and Engineering, University of Washington, Seattle, WA, USA
| | - Noor Nakhaei
- Medical & Imaging Informatics, Department of Radiological Sciences, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA
| | - Bo Li
- Department of Radiological Sciences, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA
| | - Joann G. Elmore
- Department of Medicine, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA
| | - Linda Shapiro
- Department of Electrical and Computer Engineering, University of Washington, Seattle, WA, USA,Paul G. Allen School of Computer Science and Engineering, University of Washington, Seattle, WA, USA
| | - William Hsu
- Medical & Imaging Informatics, Department of Radiological Sciences, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA,Corresponding author. (W. Hsu)
| |
Collapse
|