5
|
Uddin KMS, Zhang M, Anastasio M, Zhu Q. Optimal breast cancer diagnostic strategy using combined ultrasound and diffuse optical tomography. BIOMEDICAL OPTICS EXPRESS 2020; 11:2722-2737. [PMID: 32499955 PMCID: PMC7249842 DOI: 10.1364/boe.389275] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 01/27/2020] [Revised: 03/19/2020] [Accepted: 03/31/2020] [Indexed: 05/02/2023]
Abstract
Ultrasound (US)-guided near-infrared diffuse optical tomography (DOT) has demonstrated great potential as an adjunct breast cancer diagnosis tool to US imaging alone, especially in reducing unnecessary benign biopsies. However, DOT data processing and image reconstruction speeds remain slow compared to the real-time speed of US. Real-time or near real-time diagnosis with DOT is an important step toward the clinical translation of US-guided DOT. Here, to address this important need, we present a two-stage diagnostic strategy that is both computationally efficient and accurate. In the first stage, benign lesions are identified in near real-time by use of a random forest classifier acting on the DOT measurements and the radiologists' US diagnostic scores. Any lesions that cannot be reliably classified by the random forest classifier will be passed on to the second stage which begins with image reconstruction. Functional information from the reconstructed hemoglobin concentrations is employed by a Support Vector Machine (SVM) classifier for diagnosis at the end of the second stage. This two-step classification approach which combines both perturbation data and functional features, results in improved classification, as denoted by the receiver operating characteristic (ROC) curve. Using this two-step approach, the area under the ROC curve (AUC) is 0.937 ± 0.009, with a sensitivity of 91.4% and specificity of 85.7%. In comparison, using functional features and US score yields an AUC of 0.892 ± 0.027, with a sensitivity of 90.2% and specificity of 74.5%. Most notably, the specificity is increased by more than 10% due to the implementation of the random forest classifier.
Collapse
Affiliation(s)
- K. M. Shihab Uddin
- Biomedical Engineering Department, Washington University in St. Louis, 1 Brooking Dr, St. Louis, MO 63130, USA
| | - Menghao Zhang
- Electrical and System Engineering Department, Washington University in St. Louis, 1 Brooking Dr, St. Louis, MO 63130, USA
| | - Mark Anastasio
- Biomedical Engineering Department, Washington University in St. Louis, 1 Brooking Dr, St. Louis, MO 63130, USA
- Department of Bioengineering, University of Illinois at Urbana-Champaign, 1406 W Green St, Urbana, IL 61801, USA
| | - Quing Zhu
- Biomedical Engineering Department, Washington University in St. Louis, 1 Brooking Dr, St. Louis, MO 63130, USA
| |
Collapse
|
9
|
Zhu Q, Ricci A, Hegde P, Kane M, Cronin E, Merkulov A, Xu Y, Tavakoli B, Tannenbaum S. Assessment of Functional Differences in Malignant and Benign Breast Lesions and Improvement of Diagnostic Accuracy by Using US-guided Diffuse Optical Tomography in Conjunction with Conventional US. Radiology 2016; 280:387-97. [PMID: 26937708 PMCID: PMC4976463 DOI: 10.1148/radiol.2016151097] [Citation(s) in RCA: 38] [Impact Index Per Article: 4.8] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/17/2022]
Abstract
Purpose To investigate ultrasonography (US)-guided diffuse optical tomography to distinguish the functional differences of hemoglobin concentrations in a wide range of malignant and benign breast lesions and to improve breast cancer diagnosis in conjunction with conventional US. Materials and Methods The study protocol was approved by the institutional review boards and was HIPAA compliant. Written informed consent was obtained from all patients. Patients (288 women; mean age, 50 years; range, 17-94 years) who underwent US-guided biopsy were imaged with a handheld US and optical probe. The US-imaged lesion was used to guide reconstruction of light absorption maps at four wavelengths, and total hemoglobin (tHb), oxygenated hemoglobin (oxyHb), and deoxygenated hemoglobin (deoxyHb) were computed from the absorption maps. A threshold (80 μmol/L) was chosen on the basis of this study population. Two radiologists retrospectively evaluated US images on the basis of the US Breast Imaging Reporting and Data System lexicon, and a lesion was considered malignant when a score of 4C or 5 was given or a lesion had tHb greater than 80 μmol/L. A two-sample t test was used to calculate significance between groups, and Spearman ρ was computed between hemoglobin parameters and tumor pathologic grades. Results Three tumors were Tis, 37 were T1, 19 were T2-T4 carcinomas, and 233 were benign lesions. The mean maximum tHb, oxyHb, and deoxyHb of Tis-T1 and T2-T4 groups were 89.3 μmol/L ± 20.2 (standard deviation), 65.0 μmol/L ± 20.8, and 33.5 μmol/L ± 11.3, respectively, and 84.7 μmol/L ± 32.8, 57.1 μmol/L ± 19.8, and 34.7 μmol/L ± 18.9, respectively. The corresponding values of benign lesions were 54.1 μmol/L ± 23.5, 38.0 μmol/L ± 17.4, and 25.2 μmol/L ± 13.8, respectively. The mean maximum tHb, oxyHb, and deoxyHb were significantly higher in the malignant groups than the benign group (P <.001, <.001, and .041, respectively). For malignant lesions, the mean maximum tHb moderately correlated with tumor histologic grade and nuclear grade (ρ = 0.283 and 0.315, respectively). The mean maximum oxyHb moderately correlated with tumor nuclear grade (ρ = 0.267). When radiologists' US diagnosis and the tHb were used together, the sensitivity, specificity, positive predictive value, and negative predictive value were 96.6%-100%, 77.3%-83.3%, 52.7%-59.4%, and 99.0%-100%, respectively, for the combined malignant group. Conclusion The tHb and oxyHb correlate with breast cancer pathologic grade and can be used as an adjunct to US to improve sensitivity and negative predictive value in breast cancer diagnosis. (©) RSNA, 2016 Online supplemental material is available for this article.
Collapse
Affiliation(s)
- Quing Zhu
- From the Department of Electrical and Biomedical Engineering (Q.Z.) and Department of Electrical and Computer Engineering (Y.X., B.T.), University of Connecticut, 371 Fairfield Rd, U4157, Storrs, CT 06269; Departments of Pathology (A.R.) and Radiology (E.C.), Hartford Hospital, Hartford, Conn; and Department of Pathology (P.H.), Department of Radiology (M.K., A.M.), and Carole & Ray Neag Comprehensive Cancer Center (S.T.), University of Connecticut Health Center, Farmington, Conn
| | - Andrew Ricci
- From the Department of Electrical and Biomedical Engineering (Q.Z.) and Department of Electrical and Computer Engineering (Y.X., B.T.), University of Connecticut, 371 Fairfield Rd, U4157, Storrs, CT 06269; Departments of Pathology (A.R.) and Radiology (E.C.), Hartford Hospital, Hartford, Conn; and Department of Pathology (P.H.), Department of Radiology (M.K., A.M.), and Carole & Ray Neag Comprehensive Cancer Center (S.T.), University of Connecticut Health Center, Farmington, Conn
| | - Poornima Hegde
- From the Department of Electrical and Biomedical Engineering (Q.Z.) and Department of Electrical and Computer Engineering (Y.X., B.T.), University of Connecticut, 371 Fairfield Rd, U4157, Storrs, CT 06269; Departments of Pathology (A.R.) and Radiology (E.C.), Hartford Hospital, Hartford, Conn; and Department of Pathology (P.H.), Department of Radiology (M.K., A.M.), and Carole & Ray Neag Comprehensive Cancer Center (S.T.), University of Connecticut Health Center, Farmington, Conn
| | - Mark Kane
- From the Department of Electrical and Biomedical Engineering (Q.Z.) and Department of Electrical and Computer Engineering (Y.X., B.T.), University of Connecticut, 371 Fairfield Rd, U4157, Storrs, CT 06269; Departments of Pathology (A.R.) and Radiology (E.C.), Hartford Hospital, Hartford, Conn; and Department of Pathology (P.H.), Department of Radiology (M.K., A.M.), and Carole & Ray Neag Comprehensive Cancer Center (S.T.), University of Connecticut Health Center, Farmington, Conn
| | - Edward Cronin
- From the Department of Electrical and Biomedical Engineering (Q.Z.) and Department of Electrical and Computer Engineering (Y.X., B.T.), University of Connecticut, 371 Fairfield Rd, U4157, Storrs, CT 06269; Departments of Pathology (A.R.) and Radiology (E.C.), Hartford Hospital, Hartford, Conn; and Department of Pathology (P.H.), Department of Radiology (M.K., A.M.), and Carole & Ray Neag Comprehensive Cancer Center (S.T.), University of Connecticut Health Center, Farmington, Conn
| | - Alex Merkulov
- From the Department of Electrical and Biomedical Engineering (Q.Z.) and Department of Electrical and Computer Engineering (Y.X., B.T.), University of Connecticut, 371 Fairfield Rd, U4157, Storrs, CT 06269; Departments of Pathology (A.R.) and Radiology (E.C.), Hartford Hospital, Hartford, Conn; and Department of Pathology (P.H.), Department of Radiology (M.K., A.M.), and Carole & Ray Neag Comprehensive Cancer Center (S.T.), University of Connecticut Health Center, Farmington, Conn
| | - Yan Xu
- From the Department of Electrical and Biomedical Engineering (Q.Z.) and Department of Electrical and Computer Engineering (Y.X., B.T.), University of Connecticut, 371 Fairfield Rd, U4157, Storrs, CT 06269; Departments of Pathology (A.R.) and Radiology (E.C.), Hartford Hospital, Hartford, Conn; and Department of Pathology (P.H.), Department of Radiology (M.K., A.M.), and Carole & Ray Neag Comprehensive Cancer Center (S.T.), University of Connecticut Health Center, Farmington, Conn
| | - Behnoosh Tavakoli
- From the Department of Electrical and Biomedical Engineering (Q.Z.) and Department of Electrical and Computer Engineering (Y.X., B.T.), University of Connecticut, 371 Fairfield Rd, U4157, Storrs, CT 06269; Departments of Pathology (A.R.) and Radiology (E.C.), Hartford Hospital, Hartford, Conn; and Department of Pathology (P.H.), Department of Radiology (M.K., A.M.), and Carole & Ray Neag Comprehensive Cancer Center (S.T.), University of Connecticut Health Center, Farmington, Conn
| | - Susan Tannenbaum
- From the Department of Electrical and Biomedical Engineering (Q.Z.) and Department of Electrical and Computer Engineering (Y.X., B.T.), University of Connecticut, 371 Fairfield Rd, U4157, Storrs, CT 06269; Departments of Pathology (A.R.) and Radiology (E.C.), Hartford Hospital, Hartford, Conn; and Department of Pathology (P.H.), Department of Radiology (M.K., A.M.), and Carole & Ray Neag Comprehensive Cancer Center (S.T.), University of Connecticut Health Center, Farmington, Conn
| |
Collapse
|