Optoacoustic Imaging With Decision Support for Differentiation of Benign and Malignant Breast Masses: A 15-Reader Retrospective Study
Abstract
Please see the Editorial Comment by Babita Panigrahi discussing this article.
Chinese (
To listen to the podcast associated with this article, please select one of the following: iTunes, Google Play, or direct download.
BACKGROUND. Overlap in ultrasound features of benign and malignant breast masses yields high rates of false-positive interpretations and benign biopsy results. Optoacoustic imaging is an ultrasound-based functional imaging technique that can increase specificity.
OBJECTIVE. The purpose of this study was to compare specificity at fixed sensitivity of ultrasound images alone and of fused ultrasound and optoacoustic images evaluated with machine learning–based decision support tool (DST) assistance.
METHODS. This retrospective Reader-02 study included 480 patients (mean age, 49.9 years) with 480 breast masses (180 malignant, 300 benign) that had been classified as BI-RADS category 3–5 on the basis of conventional gray-scale ultrasound findings. The patients were selected by stratified random sampling from the earlier prospective 16-site Pioneer-01 study. For that study, masses were further evaluated by ultrasound alone followed by fused ultrasound and optoacoustic imaging between December 2012 and September 2015. For the current study, 15 readers independently reviewed the previously acquired images after training in optoacoustic imaging interpretation. Readers first assigned probability of malignancy (POM) on the basis of clinical history, mammographic findings, and conventional ultrasound findings. Readers then evaluated fused ultrasound and optoacoustic images, assigned scores for ultrasound and optoacoustic imaging features, and viewed a POM prediction score derived by a machine learning–based DST before issuing final POM. Individual and mean specificities at fixed sensitivity of 98% and partial AUC (pAUC) (95–100% sensitivity) were calculated.
RESULTS. Averaged across all readers, specificity at fixed sensitivity of 98% was significantly higher for fused ultrasound and optoacoustic imaging with DST assistance than for ultrasound alone (47.2% vs 38.2%; p = .03). Across all readers, pAUC was higher (p < .001) for fused ultrasound and optoacoustic imaging with DST assistance (0.024 [95% CI, 0.023–0.026]) than for ultrasound alone (0.021 [95% CI, 0.019–0.022]). Better performance using fused ultrasound and optoacoustic imaging with DST assistance than using ultrasound alone was observed for 14 of 15 readers for specificity at fixed sensitivity and for 15 of 15 readers for pAUC.
CONCLUSION. Fused ultrasound and optoacoustic imaging with DST assistance had significantly higher specificity at fixed sensitivity than did conventional ultrasound alone.
CLINICAL IMPACT. Optoacoustic imaging, integrated with reader training and DST assistance, may help reduce the frequency of biopsy of benign breast masses.
HIGHLIGHTS
Key Finding
▪ Fused ultrasound-optoacoustic imaging with DST assistance had higher specificity at 98% sensitivity than ultrasound alone (47.2% vs 38.2%; p = .03). Of 2851 BI-RADS category ≥ 4A assessments of benign masses by ultrasound alone, 825 (28.9%) were correctly downgraded to BI-RADS 2 or 3 with optoacoustic imaging and DST assistance.
Importance
▪ Optoacoustic imaging with decision support may help reduce biopsies of benign breast masses compared with gray-scale ultrasound interpretation alone while maintaining 98% sensitivity for cancer.
Breast cancer is a heterogeneous disease with variable gray-scale ultrasound appearances that overlap with those of benign lesions, resulting in performance of hundreds of thousands of ultrasound-guided biopsies of benign masses each year [1–5]. The potential harms of false-positive results of breast imaging examinations have garnered much attention, resulting in a drive to improve the specificity of imaging [6, 7]. This goal has led to increased use of functional imaging tools that provide supplementary biologic information to help differentiate benign masses from cancer [8].
Optoacoustic images can provide functional data in real time, spatially fused and temporally interleaved with standard ultrasound images without the need for contrast material injection or ionizing radiation [9]. Optoacoustic imaging entails use of a pulsed ultrasound laser at long and short optical wavelengths to generate and detect acoustic signals from oxygenated, deoxygenated, and total hemoglobin [10–15]. Rapid cell proliferation in breast cancers causes hypoxia that triggers neovascularity. Optoacoustic images coregistered with conventional ultrasound images leverage this intrinsic tissue contrast to improve diagnostic precision [16–22]. The combination of ultrasound and optoacoustic imaging was previously found to improve the accuracy of classification of benign breast lesions that had suspicious ultrasound features, thus potentially reducing the number of benign biopsies significantly [23–25].
The Pioneer-01 pivotal study was designed to evaluate the safety and effectiveness of a first-generation system performing combined ultrasound and optoacoustic imaging [24]. Optoacoustic imaging, compared with ultrasound, had a 14.9% increase in specificity, from 28.1% to 43.0%, but a 2.6% decrease in sensitivity, from 98.6% to 96.0%. Although feature scores on optoacoustic images were described as objective observations, conversion of the score on optoacoustic images to a probability of malignancy (POM) and BI-RADS category was at each reader's judgment. In addition, features on optoacoustic images did not contribute equally to POM assessments. Sensitivity, specificity, and POM were therefore likely substantially influenced by readers' training and familiarity with feature categorization on optoacoustic images. The 2.6% decrease in sensitivity may have been due in part to radiologists' difficulty integrating features on ultrasound images with those on optoacoustic images. To facilitate radiologists' evaluations, a machine learning–based decision support tool (DST) was developed to help synthesize findings on ultrasound and optoacoustic images and thereby improve radiologists' BI-RADS assessments. The primary goal of implementing a DST was to help better integrate the multitude of features on ultrasound and optoacoustic images and improve user experience in interpreting features on ultrasound and optoacoustic images in real time. We hypothesized that modifications of reader training and implementation of the DST would collectively help address the 2.6% decrease in sensitivity of optoacoustic imaging that was observed in the initial Pioneer-01 clinical study.
According to BI-RADS, the POM cutoff for recommending biopsy of a finding at a breast imaging examination is 2%, establishing diagnostic performance benchmarks of 98% sensitivity and a 2% false-negative rate (FNR) [26]. The aim of the current study was to compare specificity at fixed sensitivity of ultrasound alone with that of fused ultrasound and optoacoustic imaging evaluated with machine learning–based DST assistance.
Methods
Patient Selection
The Reader-02 study (National Clinical Trials identifier NCT04030104) was a single-arm retrospective multireader study conducted with images previously obtained as part of the Pioneer-01 pivotal study (National Clinical Trials identifier NCT01943916). The Reader-02 study was conducted to evaluate the effect on diagnostic performance of two modifications with respect to the Pioneer-01 study: more extensive reader training and implementation of the DST.
The Pioneer-01 study was a HIPAA-compliant prospective clinical trial performed at 16 sites that enrolled women age 18 and older presenting with a solid or complex cystic and solid breast mass assessed with conventional ultrasound as BI-RADS category 3–5 between December 2012 and September 2015. The study sample and trial design have been previously reported [24]. The institutional review boards of all participating institutions approved the study. Participants underwent both gray-scale ultrasound and optoacoustic imaging evaluation of the mass in a single session. A handheld duplex probe was used as both a standalone gray-scale ultrasound transducer and a first-generation optoacoustic imaging device (Imagio, Seno Medical Instruments). Biopsy and/or imaging follow-up of the mass was performed according to the study protocol.
In the Pioneer-01 study, 2105 patients with 2191 masses provided written informed consent. After exclusions, the intent-to-diagnose sample included 1739 patients with 1808 masses. Additional masses were excluded from potential inclusion in the Reader-02 study for the following reasons: additional mass in patients already included in study sample (n = 69); mass used for reader training and proficiency testing (n = 116); mass used for DST training; or mass not biopsied, BI-RADS category 3 assigned enrollment, and 12-month follow-up gray-scale ultrasound showed an increase in size or BI-RADS category (based on a retrospective review performed by a panel of radiologists as part of the Pioneer-01 study, as previously described [24]) (n = 8). These exclusions resulted in 1615 patients with 1615 masses who were eligible for potential selection for inclusion in the Reader-02 study. All included masses either were classified as benign (including high risk) or malignant on the basis of histologic evaluation or, in the absence of biopsy of a BI-RADS category 3 mass, were classified as benign on the basis of stable size and BI-RADS category at 12-month follow-up.
For the final study sample, eligible patients were selected by means of stratified random sampling to maintain the same distribution of BI-RADS categories and diagnoses as in the Pioneer-01 study. This stratified random sampling was used to construct blocks of 120 masses, each containing 20, 75, and 25 masses with BI-RADS categories of 3, 4, and 5, respectively, or of 45 malignant masses and 75 masses classified as benign (72 on the basis of benign histology or 12-month follow-up and three on the basis of high-risk histology). Four such blocks were constructed to provide a total of 480 masses, reflecting the prior determined sample size. The final study sample comprised these 480 patients (mean age, 49.9 years) with 480 masses (Fig. 1).
Design of Reader Study
Ultrasound examinations were independently interpreted by 15 breast radiologists (seven academic, eight nonacademic) with 4–38 years of posttraining experience. The interpretations were conducted between July 30, 2019, and November 3, 2019. The readers were blinded to the reference standard outcome of each mass. Readers reviewed static images and video sweeps of ultrasound images alone and of fused ultrasound and optoacoustic imaging in orthogonal planes. The order of the masses was individually randomized within each block. All 15 readers interpreted the masses in the same order. Before starting the study interpretations, the readers underwent training in scoring and interpretation of features for both ultrasound and optoacoustic imaging, as described in the Supplemental Methods (available in the online supplement).
The readers assessed five features on conventional ultrasound images (peripheral zone, boundary zone vessel, shape, internal texture, and sound transmission) and five features on optoacoustic images (external peripheral radiating vessels, boundary zone vessels, internal vessels, internal hemoglobin, and internal blush). Readers were provided with the following clinical variables: age, breast cancer history, indication for ultrasound, and mass location. Readers were also provided with preultrasound mammograms, if available.
Readers first assigned the mass a POM and BI-RADS category on the basis of review of ultrasound images alone. They were instructed to consider clinical information and any available mammograms when formulating these assessments. During this assessment, readers also measured the maximum diameter and the depth from skin to the posterior margin of each mass. No other ultrasound features were evaluated at this stage. Next, using an electronic form, readers evaluated fused ultrasound and optoacoustic images and recorded scores for the five ultrasound features and five optoacoustic imaging features, assigning scores on integer scales from 0 to 5 or 0 to 6. The DST then displayed a predicted POM, computed with the scores that the reader had just entered. The reader then assigned a final POM and BI-RADS category, considering the DST results. Readers were allowed to assign a final POM different from the DST-predicted POM. Both the DST-predicted POM and reader-assigned final POM, assisted by DST, were recorded.
The reader study design is further described in the Supplemental Methods.
Decision Support Tool
The machine learning–based DST was produced with the eX-treme Gradient Boosting (XGBoost) algorithm and trained with feature score data from seven independent readers in the Pioneer-01 study using distinct cases from those included in the Reader-02 study [27]. The DST used reader-assigned scores for ultrasound and optoacoustic features and patient age, mass size (as measured by the reader), depth to posterior mass wall (as measured by the reader), and, if available, mammographic BI-RADS category (based on the clinical report, as entered by the reader in the DST interface). On the basis of this information, the interface graphically displayed a predicted POM (and its 95% CI), which ranged from 0% to 100%, and mapped this POM to a predicted FNR (and its 95% CI). Figure S1 (available in the online supplement) shows screen shots of the DST interface. Development of the DST is further described in the Supplemental Methods.
Statistical Analysis
To reflect the counterbalance between increase in specificity and loss in sensitivity, the primary endpoint of the study was specificity at fixed sensitivity of 98%, and the secondary endpoint was the partial AUC (pAUC) within a range of interest of 95–100% sensitivity. Readers' observed specificity and sensitivity for ultrasound alone and for fused ultrasound and optoacoustic imaging with DST assistance were reported first. Then, model-adjusted specificity for both image sets was reported at 98% sensitivity (i.e., 2% FNR). Generalized estimating equations were used to determine the model-based results. Sample size calculations were based on the primary endpoint of evaluating the differences in specificity at fixed 98% sensitivity. To detect a 10% absolute increase in specificity at fixed sensitivity with 80% power—a hypothesis test with a two-sided 5% alpha value among 15 readers by use of previously observed intrareader and interreader variance estimates—a sample size of 480 masses was estimated. Statistical analysis was performed with SAS software (version 9.4, SAS Institute) for the generalized estimating equations and OR-DBM MRMC 2.51 software (Medical Image Perception Laboratory) [28] for sample size determination and for calculation of specificity at fixed sensitivity of 98%. Additional aspects of the statistical analysis are described in the Supplemental Methods.
Results
A total of 480 masses in 480 women—180 (37.5%) malignant on the basis of histopathology (175 invasive, five ductal carcinoma in situ); 300 (62.5%) classified as benign (237 on the basis of benign histopathology, 12 on the basis of high-risk histopathology, 51 on the basis of 12-month follow-up)—from the original Pioneer-01 study intent-to-diagnose sample were randomly selected with stratification based on disease prevalence. The 15 readers provided a total of 7200 reads (2700 for malignant masses, 4500 for masses classified as benign) for ultrasound alone and for fused ultrasound and optoacoustic imaging with DST assistance. Mammography was available for 178 of 180 (98.9%) malignant masses and 252 of 300 (84.0%) benign masses. Patient and mass characteristics are summarized in Table 1. The relative importance of scores for the ultrasound and optoacoustic imaging features used with the DST are shown in Table S1 (available in the online supplement).
| Characteristic | All Masses (n = 480) | Benign Masses (n = 300)a | Malignant Masses (n = 180) |
|---|---|---|---|
| Mammography present | |||
| Yes | 430 (89.6) | 252 (84.0) | 178 (98.9) |
| No | 50 (10.4) | 48 (16.0) | 2 (1.1) |
| BI-RADS category | |||
| 2 | 2 (0.4) | 2 (0.7) | 0 (0.0) |
| 3 | 74 (15.4) | 73 (24.3) | 1 (0.6) |
| 4A | 129 (26.9) | 125 (41.7) | 4 (2.2) |
| 4B | 84 (17.5) | 70 (23.3) | 14 (7.8) |
| 4C | 87 (18.1) | 22 (7.3) | 65 (36.1) |
| 5 | 102 (21.3) | 6 (2.0) | 96 (53.3) |
| Missing | 2 (0.4) | 2 (0.7) | 0 (0.0) |
| Breast density 1 | 24 (5.0) | 10 (3.3) | 14 (7.8) |
| 2 | 168 (35.0) | 93 (31.0) | 75 (41.7) |
| 3 | 187 (39.0) | 116 (38.7) | 71 (39.4) |
| 4 | 40 (8.3) | 26 (8.7) | 14 (7.8) |
| Missing | 61 (12.7) | 55 (18.3) | 6 (3.3) |
| Palpable | |||
| Yes | 210 (43.8) | 121 (40.3) | 89 (49.4) |
| No | 199 (41.5) | 130 (43.3) | 69 (38.3) |
| Unknown | 71 (14.8) | 49 (16.3) | 22 (12.2) |
| Breast implants | |||
| Yes | 14 (2.9) | 8 (2.7) | 6 (3.3) |
| No | 466 (97.1) | 292 (97.3) | 174 (96.7) |
| Postmenopausal | |||
| Yes | 219 (45.6) | 90 (30.0) | 129 (71.7) |
| No | 258 (53.8) | 208 (69.3) | 50 (27.8) |
| Unknown | 3 (0.6) | 2 (0.7) | 1 (0.6) |
| Relevant medical historyb | |||
| Yes | 245 (51.0) | 136 (45.3) | 109 (60.6) |
| No | 235 (49.0) | 164 (54.7) | 71 (39.4) |
| Age | |||
| Mean ± SD | 49.9 ± 14.4 | 44.2 ± 12.5 | 59.4 ± 12.2 |
| Median | 49.0 | 45.0 | 59.5 |
| Minimum, maximum | 18, 88 | 18, 77 | 28, 88 |
| Mass maximal diameter (cm) | |||
| Mean ± SD | 1.5 ± 0.8 | 1.4 ± 0.8 | 1.5 ± 0.8 |
| Median | 1.3 | 1.2 | 1.3 |
| Minimum, maximum | 0.3, 3.9 | 0.3, 3.9 | 0.4, 3.9 |
| Distance from nipple (cm) | |||
| Mean ± SD | 5.3 ± 3.2 | 4.8 ± 3.1 | 6.0 ± 3.3 |
| Median | 5.0 | 4.0 | 6.0 |
| Minimum, maximum | 0.0, 17.0 | 0.0, 15.0 | 0.0, 17.0 |
| Depth from skin to posterior margin of mass (cm) | |||
| Mean ± SD | 1.7 ± 0.7 | 1.6 ± 0.6 | 2.0 ± 0.7 |
| Median | 1.7 | 1.6 | 1.9 |
| Minimum, maximum | 0.5, 3.8 | 0.5, 3.6 | 0.6, 3.8 |
Note—Except where otherwise indicated, data are number of patients with percentage in parentheses. Percentages do not total 100 owing to rounding.
a
Includes 237 masses that were benign by histopathologic evaluation, 51 masses with stability in size and BI-RADS category 3 assessment at 12-month follow-up, and 12 lesions with high-risk results based on histopathologic evaluation.
b
At least one condition recorded under medical history (e.g., cervical cancer or partial hysterectomy) by Pioneer-01 investigators.
Specificity and Sensitivity
The observed specificity was 36.6% (1649/4500) for ultrasound alone versus 50.1% (2256/4500) for fused ultrasound and optoacoustic imaging with DST assistance (p < .001). Table 2 shows individual and aggregate reader-observed specificity for benign masses. The observed sensitivity was 98.1% (2648/2700) for ultrasound alone and 97.7% (2637/2700) for fused ultrasound and optoacoustic imaging with DST assistance (p = .35).
| Reader | Observed Specificity | Observed Sensitivity | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Ultrasound Alone | Optoacoustic Imaging With DST | Difference | Ultrasound Alone | Optoacoustic Imaging With DST | Difference | |||||||
| Mean | 95% CI | Mean | 95% CI | Mean | 95% CI | Mean | 95% CI | Mean | 95% CI | Mean | 95% CI | |
| 1 | 32.3 (97/300) | 27.0–37.6 | 50.7 (152/300) | 45.0–56.3 | 18.3 | 12.9–23.8 | 98.3 (177/180) | 95.2–99.7 | 97.2 (175/180) | 93.6–99.1 | −1.1 | −3.3 to 1.1 |
| 2 | 53.7 (161/300) | 48.0–59.3 | 57.7 (173/300) | 52.1–63.3 | 4.0 | 0.3–7.7 | 93.3 (168/180) | 88.6–96.5 | 96.1 (173/180) | 92.2–98.4 | 2.8 | −0.5 to 6.0 |
| 3 | 24.7 (74/300) | 19.8–29.5 | 38.7 (116/300) | 33.2–44.2 | 14.0 | 8.8–19.2 | 98.3 (177/180) | 95.2–99.7 | 98.3 (177/180) | 95.2–99.7 | 0.0 | −2.2 to 2.2 |
| 4 | 39.7 (119/300) | 34.1–45.2 | 60.3 (181/300) | 54.8–65.9 | 20.7 | 15.2–26.2 | 97.8 (176/180) | 94.4–99.4 | 96.7 (174/180) | 92.9–98.8 | −1.1 | −2.6 to 0.4 |
| 5 | 29.3 (88/300) | 24.2–34.5 | 32.0 (96/300) | 26.7–37.3 | 2.7 | −2.0 to 7.4 | 97.8 (176/180) | 94.4–99.4 | 98.9 (178/180) | 96.0–99.9 | 1.1 | −0.4 to 2.6 |
| 6 | 46.0 (138/300) | 40.4–51.6 | 42.7 (128/300) | 37.1–48.3 | −3.3 | −7.7 to 1.1 | 97.8 (176/180) | 94.4–99.4 | 98.3 (177/180) | 95.2–99.7 | 0.6 | −0.5 to 1.6 |
| 7 | 41.3 (124/300) | 35.8–46.9 | 55.7 (167/300) | 50.0–61.3 | 14.3 | 9.8–18.9 | 99.4 (179/180) | 96.9–100.0 | 98.3 (177/180) | 95.2–99.7 | −1.1 | −2.6 to 0.4 |
| 8 | 25.0 (75/300) | 20.1–29.9 | 51.3 (154/300) | 45.7–57.0 | 26.3 | 20.9–31.7 | 100.0 (180/180) | 98.0–100.0 | 98.3 (177/180) | 95.2–99.7 | −1.7 | −3.5 to 0.2 |
| 9 | 40.7 (122/300) | 35.1–46.2 | 59.7 (179/300) | 54.1–65.2 | 19.0 | 13.4–24.6 | 98.3 (177/180) | 95.2–99.7 | 96.1 (173/180) | 92.2–98.4 | −2.2 | −5.3 to 0.8 |
| 10 | 50.3 (151/300) | 44.7–56.0 | 44.3 (133/300) | 38.7–50.0 | −6.0 | −10.9 to 1.1 | 96.7 (174/180) | 92.9–98.8 | 98.3 (177/180) | 95.2–99.7 | 1.7 | −0.8 to 4.1 |
| 11 | 38.0 (114/300) | 32.5–43.5 | 44.7 (134/300) | 39.0–50.3 | 6.7 | 2.2–11.1 | 98.9 (178/180) | 96.0–99.9 | 98.9 (178/180) | 96.0–99.9 | 0.0 | −2.2 to 2.2 |
| 12 | 60.0 (180/300) | 54.5–65.5 | 64.0 (192/300) | 58.6–69.4 | 4.0 | −0.5 to 8.5 | 96.7 (174/180) | 92.9–98.8 | 95.0 (171/180) | 90.7–97.7 | −1.7 | −4.9 to 1.6 |
| 13 | 20.3 (61/300) | 15.8–24.9 | 44.0 (132/300) | 38.4–49.6 | 23.7 | 18.3–29.1 | 100.0 (180/180) | 98.0–100.0 | 99.4 (179/180) | 96.9–100 | −0.6 | −1.6 to 0.5 |
| 14 | 38.3 (115/300) | 32.8–43.8 | 57.3 (172/300) | 51.7–52.9 | 19.0 | 13.0–25.0 | 98.3 (177/180) | 95.2–99.7 | 96.7 (174/180) | 92.9–98.8 | −1.7 | −4.1 to 0.8 |
| 15 | 10.0 (30/300) | 6.6–13.4 | 49.0 (147/300) | 43.3–54.7 | 39.0 | 33.5–44.5 | 99.4 (179/180) | 96.9–100.0 | 98.3 (177/180) | 95.2–99.7 | −1.1 | −2.6 to 0.4 |
| All | 36.6 (1649/4500) | 33.4–40.0 | 50.1 (2256/4500) | 46.1–54.1 | 13.5 | 11.3–15.7 | 98.1 (2648/2700) | 96.6–98.9 | 97.7 (2637/2700) | 95.8–98.7 | −0.4 | −1.5 to 0.4 |
Note—Data are percentage with number of reads in parentheses. Optoacoustic images were fused to ultrasound images. Differences were computed as value for fused ultrasound and optoacoustic imaging with decision support tool (DST) assistance minus value for ultrasound alone.
Fused ultrasound and optoacoustic imaging with DST assistance had similar sensitivity for masses of varying sizes (96.8% [752/777] for lesions smaller than 1 cm, 97.5% [1177/1207] for lesions measuring 1–2 cm, 98.9% [708/716] for lesions larger than 2 cm; p = .28). Fused ultrasound and optoacoustic imaging with DST assistance had lower sensitivity for masses located less than 1 cm from the skin (93.3% [42/45]) than for masses located 1 cm or more from the skin (97.7% [2067/2115]) (p = .02).
Based on mean values across readers, specificity at fixed 98% sensitivity was significantly higher for fused ultrasound and optoacoustic imaging with DST assistance than for ultrasound alone (47.2% vs 38.2%) (difference, 9.0%; 95% CI, 1.0–16.9%; p = .03) (Table 3). Specificity was higher for fused ultrasound and optoacoustic imaging with DST assistance than for ultrasound alone for 14 of the 15 readers (Fig. 2A); the difference in specificity for individual readers ranged from −1.2% to 16.4%. Among masses with discordant POM scores between the two modalities (i.e., one modality positive and the other negative for predicting malignancy), fused ultrasound and optoacoustic imaging with DST assistance had 607 fewer false-positives (13.5% [607/4500] increase in specificity) and 11 more false-negatives (0.4% [11/2700] loss in sensitivity) than ultrasound alone.
| Reader | Specificity at Fixed Sensitivity of 98% | pAUC (Sensitivity, 95–100%) | ||||
|---|---|---|---|---|---|---|
| Ultrasound Alone | Optoacoustic Imaging With DST | Difference | Ultrasound Alone | Optoacoustic Imaging With DST | Difference | |
| 1 | 36.9 | 35.7 | −1.2 | 0.019 | 0.021 | 0.002 |
| 2 | 38.0 | 43.7 | 5.7 | 0.019 | 0.023 | 0.004 |
| 3 | 27.2 | 40.1 | 12.9 | 0.015 | 0.020 | 0.005 |
| 4 | 36.1 | 52.5 | 16.4 | 0.023 | 0.027 | 0.004 |
| 5 | 25.1 | 34.9 | 9.8 | 0.016 | 0.023 | 0.007 |
| 6 | 42.9 | 46.7 | 3.8 | 0.023 | 0.024 | 0.001 |
| 7 | 45.6 | 57.8 | 12.2 | 0.023 | 0.027 | 0.004 |
| 8 | 39.8 | 53.9 | 14.1 | 0.021 | 0.028 | 0.007 |
| 9 | 42.0 | 44.6 | 2.6 | 0.021 | 0.023 | 0.002 |
| 10 | 37.5 | 52.9 | 15.4 | 0.021 | 0.026 | 0.005 |
| 11 | 43.3 | 48.2 | 4.9 | 0.023 | 0.024 | 0.001 |
| 12 | 37.1 | 40.2 | 3.1 | 0.021 | 0.022 | 0.001 |
| 13 | 43.2 | 54.5 | 11.3 | 0.023 | 0.027 | 0.004 |
| 14 | 41.3 | 48.9 | 7.6 | 0.021 | 0.024 | 0.003 |
| 15 | 37.3 | 53.2 | 15.9 | 0.020 | 0.026 | 0.006 |
| All | 38.2 (24.9–51.6) | 47.2 (35.9–58.5) | 9.0 (1.0–16.9) | 0.021 (0.019–0.022) | 0.024 (0.019–0.022) | 0.003 |
Note—Values in parentheses are 95% CIs. For all readers combined, difference was statistically significant for specificity at fixed sensitivity (p = .03) and for partial AUC (pAUC) (p < .001). Optoacoustic images were fused to ultrasound images. Differences were computed as value for fused ultrasound and optoacoustic imaging with decision support tool (DST) assistance minus value for ultrasound alone.
Fig. 2A —Specificity at fixed sensitivity and partial AUC (pAUC).
A, Graphs show specificity at fixed sensitivity of 98% (A) and pAUC for range of interest of sensitivity of 95–100% (B) for ultrasound alone and for fused ultrasound and optoacoustic imaging with decision support tool (DST) assistance for individual readers (blue) and for all readers combined (red). Specificity at fixed sensitivity was higher for fused ultrasound and optoacoustic imaging with DST assistance than for ultrasound alone for 14 of 15 readers, and pAUC was higher for fused ultrasound and optoacoustic imaging with DST assistance than for ultrasound alone for all 15 readers.
Fig. 2B —Specificity at fixed sensitivity and partial AUC (pAUC).
B, Graphs show specificity at fixed sensitivity of 98% (A) and pAUC for range of interest of sensitivity of 95–100% (B) for ultrasound alone and for fused ultrasound and optoacoustic imaging with decision support tool (DST) assistance for individual readers (blue) and for all readers combined (red). Specificity at fixed sensitivity was higher for fused ultrasound and optoacoustic imaging with DST assistance than for ultrasound alone for 14 of 15 readers, and pAUC was higher for fused ultrasound and optoacoustic imaging with DST assistance than for ultrasound alone for all 15 readers.
Based on the output display of the DST, a POM greater than 5% by DST corresponded to an FNR greater than 2% for classifying masses as malignant; this POM threshold was used for assessing DST performance (Table S2, available in the online supplement). When this threshold was used and metrics were averaged across readers, DST had specificity of 44.1% at fixed sensitivity of 98% and a pAUC of 0.023. Application of the DST decision threshold yielded mean sensitivity of 97.0% (95% CI, 96–97.9%) and mean specificity of 50.9% (95% CI, 46–55.8%).
Likelihood Ratios
Table S3 (available in the online supplement) summarizes data on negative (NLR) and positive (PLR) likelihood ratios. The mean NLR for all readers was 0.05 (95% CI, 0.04–0.07) for ultrasound alone and 0.05 (95% CI, 0.03–0.06) for fused ultrasound and optoacoustic imaging with DST assistance. Thus, fused ultrasound and optoacoustic imaging with DST assistance, compared with ultrasound alone, was associated with a relative decrease in NLR of 0.90 (95% CI, 0.69–1.11; p = .33). Mean PLR for all readers was 1.55 (95% CI, 1.50–1.60) for ultrasound alone and 1.96 (95% CI, 1.87–2.05) for fused ultrasound and optoacoustic imaging with DST assistance. Thus, fused ultrasound and optoacoustic imaging with DST assistance, compared with ultrasound images alone, had a relative increase in PLR of 1.28 (95% CI, 1.23–1.30; p < .001).
Partial AUC
For all 15 readers, pAUC was higher for fused ultrasound and optoacoustic imaging with DST assistance than for ultrasound alone (Table 3 and Fig. 2B). Figure 3 shows readers' mean partial ROC curves. Across all readers, mean pAUC was higher (p < .001) for fused ultrasound and optoacoustic imaging with DST assistance (0.024 [95% CI, 0.023–0.026]) than for ultrasound alone (0.021 [95% CI, 0.019–0.022]) with an absolute difference of 0.004 and relative difference of 19%.
Interreader Variability
Table S4 (available in the online supplement) summarizes data regarding interreader variability. The mean intraclass correlation coefficient for POM scores for ultrasound alone versus for fused ultrasound and optoacoustic imaging with DST assistance was 0.77 versus 0.80 for all masses, 0.56 versus 0.62 for benign masses and 0.54 versus 0.59 for malignant masses.
Agreement Between Reader and Decision Support Tool Results
Readers agreed with the DST results in 98.8% (2667/2700) of reads of malignant masses, 93.3% (4197/4500) of reads of benign masses, and 95.3% (6864/7200) of all reads. Reader and DST results disagreed in 33 of 2700 (1.2%) reads of malignant masses. Among these, the reader was correct in 26 reads and the DST in seven. Reader and DST results disagreed in 303 of 4500 (6.7%) reads of benign masses. Among these, the reader was correct in 135 reads and the DST in 168. Table 4 shows further stratification of concordant and discordant reads based on both negative and positive reads for readers and DST.
| Malignancy Prediction | Total No. of Reads | ||
|---|---|---|---|
| Reader | DST | Malignant Masses (n = 2700) | Benign Masses (n = 4500)a |
| Negative | Negative | 56 (2.1) | 2121 (47.1) |
| Negative | Positive | 7 (0.3) | 135 (3.0) |
| Positive | Negative | 26 (1.0) | 168 (3.7) |
| Positive | Positive | 2611 (96.7) | 2076 (46.1) |
Note—Data are number of reads with percentages in parentheses. Percentages do not total 100 owing to rounding.
a
Includes 237 masses that were benign by histopathologic evaluation, 51 masses with stability in size and BI-RADS category 3 assessment at 12-month follow-up, and 12 lesions with high-risk results by histopathologic evaluation.
Distributions of Probability of Malignancy Scores
Table S5 (available in the online supplement) shows reader-assigned POM scores for various groups of masses. The mean POM scores for ultrasound alone and for fused ultrasound and optoacoustic imaging with DST assistance were 65.2 and 70.1 for malignant masses (p < .001) versus 16.4 and 15.9 for benign masses (p = .21). Thus, the difference between mean POM scores for malignant and benign masses was 48.8 for ultrasound alone versus 54.2 for fused ultrasound and optoacoustic imaging with DST assistance. The subgroup of high-risk lesions had a mean POM score of 36.5 for ultrasound alone and 43.9 for fused ultrasound and optoacoustic imaging with DST assistance. Most histologically confirmed benign masses had similar or lower POM scores for fused ultrasound and optoacoustic imaging with DST assistance than for ultrasound alone. However, masses exhibiting benign inflammation on histopathology had a mean POM score of 34.1 for ultrasound alone and 43.3 for fused ultrasound and optoacoustic imaging with DST assistance.
Distributions of BI-RADS Categories
Table 5 summarizes reclassifications of BIRADS categories from initial assessments on ultrasound alone to final assessments on fused ultrasound and optoacoustic imaging with DST assistance. Of 2851 reads of benign masses that were given a false-positive assessment of BI-RADS category 4A or greater on ultrasound alone, 825 (28.9%) were correctly down-classified to BI-RADS category 2 or 3 on fused ultrasound and optoacoustic imaging with DST assistance. A total of 1649 reads of benign masses were assessed BI-RADS category 2 or 3 on ultrasound alone with incorrect up-classification of 218 (13.2%) reads to BI-RADS category 4A or greater on fused ultrasound and optoacoustic imaging with DST assistance.
| Initial BI-RADS Category | Final BI-RADS Category 2 or 3 | Final BI-RADS Category 4A, 4B, 4C, or 5 |
|---|---|---|
| Benign massesa | ||
| 2, 3 | 86.8 (1431/1649) | 13.2 (218/1649) |
| 4A, 4B, 4C, 5 | 28.9 (825/2851) | 71.1 (2026/2851) |
| All | 50.1 (2256/4500) | 49.9 (2244/4500) |
| Malignant masses | ||
| 2, 3 | 51.9 (27/52) | 48.1 (25/520) |
| 4A, 4B, 4C, 5 | 1.4 (36/2648) | 98.8 (2615/2648) |
| All | 2.3 (63/2700) | 97.7 (2637/2700) |
Note—Data are percentage with number of reads in parentheses. BI-RADS categories 2 and 3 and BI-RADS categories 4 (all three subgroups) and 5 are grouped given the recommendation for biopsy in the latter categories.
a
Includes 237 masses that were benign at histopathologic evaluation, 51 masses with stability in size and BI-RADS category 3 assessment at 12-month stability, and 12 lesions with high-risk results by histopathologic evaluation.
Among 52 reads of malignant masses that were given a false-negative assessment of BIRADS category 2 or 3 on ultrasound alone, 25 (48.1%) were correctly up-classified to BIRADS category 4 or greater on fused ultrasound and optoacoustic imaging with DST assistance. Table S6 (available in the online supplement) summarizes cancer type, molecular subtype, and histologic grade of these 25 malignant masses that were correctly up-classified on fused images with DST assistance. Of 2648 reads of malignant masses correctly assessed BI-RADS category 4A or greater on ultrasound alone, 36 (1.4%) reads were incorrectly down-classified to BI-RADS category 2 or 3 on fused ultrasound and optoacoustic imaging with DST assistance.
Among 2700 reads of malignant masses, 36 (1.3%) false-negative reads on fused ultrasound and optoacoustic imaging with DST assistance occurred in 16 of 180 (8.9%) unique malignant masses. The masses with false-negative reads, compared with those with true-positive reads, had significantly lower median distance to the nipple (2.0 cm [IQR, 1.0–5.0 cm] vs 6.0 cm [IQR, 4.0–6.0 cm]; p = .03). The masses with false-negative reads, compared with those with true-positive reads, occurred in younger patients, smaller cancers, cancers with higher Ki-67 level, and cancers with a shorter depth from skin, although these differences were not statistically significant (all p > .05) (Table S7, available in the online supplement).
Fig. 5A —63-year-old woman with lump in left breast.
A, Mediolateral oblique (left) and craniocaudal (right) mammograms of left breast show superficial and periareolar focal asymmetry in area of concern (triangular radiopaque marker). Mass was assessed BI-RADS category 4A. Mass was further evaluated with fused ultrasound and optoacoustic imaging with decision support tool (DST) assistance.
Fig. 5B —63-year-old woman with lump in left breast.
B, Screenshot shows single frame captured during video sweep of mass. Ultrasound image (top left) shows superficial hypoechoic ovoid mass abutting dermis with prominent distal acoustic enhancement. Fused optoacoustic combined image (top right) shows little internal zone vascularity or deoxygenation along anterior to central aspect of mass; findings are not suspicious for malignancy. However, other findings represent suspicious optoacoustic imaging features. For example, combined image (top right) as well as fused total hemoglobin image (lower left) show multiple large tortuous oxygenated and deoxygenated boundary-zone vessels and multiple radiating peripheral zone vessels (arrows). Display also includes fused optoacoustic relative image (lower right). On fused ultrasound and optoacoustic imaging with DST assistance, mass was assessed BI-RADS category 4A or greater by 9 of 15 readers and as BI-RADS category 3 by 6 of 15 readers. Ultrasound-guided biopsy revealed grade 1 mucinous carcinoma that was strongly estrogen and progesterone receptor positive. White ROI demarcates internal zone–boundary zone border; blue ROI denotes boundary zone–peripheral zone border. L 10 N 2 ARAD SAX = left breast 10-o'clock position 2 cm from nipple, antiradial scan plane, video sweep parallel to short axis of probe.
Discussion
Optoacoustic images provide complementary functional information to conventional ultrasound images that is intended to improve breast cancer diagnosis and reduce the rate of false-positive biopsies while maintaining sensitivity. In this study we reassessed previously collected images from the Pioneer-01 study after dedicated reader training in optoacoustic imaging and implementation of a DST. The study showed a significant difference for the primary endpoint in that specificity at fixed sensitivity of 98% was significantly higher for fused ultrasound and optoacoustic imaging with DST assistance (47.2%) than for ultrasound alone (38.2%). These results compare favorably with those of the original Pioneer-01 study with respect to performance of ultrasound alone and of fused ultrasound and optoacoustic imaging [24]. After undergoing didactic and interactive case training, 14 of 15 readers achieved higher specificity at fixed sensitivity using fused ultrasound and optoacoustic imaging with DST assistance without significant loss in sensitivity. The pAUC was higher for fused ultrasound and optoacoustic imaging with DST assistance for all 15 readers, also comparing favorably with the Pioneer-01 study results [28–30].
Changes in observed sensitivity and specificity are difficult to assess because decreases in sensitivity offset increases in specificity. This trade-off is especially relevant at sensitivity of 95% or higher, where very small losses in sensitivity can offset a large gain in specificity. The fixed sensitivity of 98% used in this study represents the generally accepted POM threshold to recommend tissue sampling. The primary endpoint of specificity at fixed sensitivity and the secondary endpoint of pAUC account for the trade-off between sensitivity and specificity. In particular, use of fixed sensitivity indicates that fused ultrasound and optoacoustic imaging with DST support can achieve significantly improved specificity without significant loss in sensitivity. That is, the 9.0% gain in specificity at fixed sensitivity of fused ultrasound and optoacoustic imaging with DST assistance accounts for the 0.4% loss in observed sensitivity of the method. Improved specificity without loss of sensitivity can help reduce the frequency of benign biopsies.
The increase in specificity with supplemental optoacoustic ultrasound in the current study was lower than previously observed for elastography [31, 32]. This comparison likely reflects differences in study design and population, given that studies of elastography included larger numbers of BI-RADS category 2 masses and cysts. The current study showed higher sensitivity of the studied method than for the previously found 80–95% range of sensitivities of elastography [33–35]; such sensitivities may be too low for use of elastography as an adjunctive test for down-classifying masses. Accordingly, elastography is likely to be used clinically for up-classification or for targeted applications, such as differentiating a complicated cyst with echogenic fluid from a fibroadenoma or other solid mass. The design of the current study allowed radiologists to be confident that the gain in specificity with fused ultrasound and optoacoustic imaging with DST assistance outweighed any potential loss of sensitivity when the technique was used for adjunctive diagnosis, leading to comfort in use of the method for down-classification. Additionally, elastography has had marked reduction in sensitivity (possibly to < 80%) in masses measuring 1 cm or smaller [33, 34]. In the current study, fused ultrasound and optoacoustic imaging with DST assistance had observed sensitivity of 96.8% for masses smaller than 1 cm, compared with 98.9% for masses larger than 2 cm. This sensitivity for small masses could be particularly important at institutions that have active supplemental MRI or automated breast ultrasound screening programs, in which small masses are frequently detected and subsequent adjunctive diagnosis is needed.
In the Pioneer-01 study, use of fused ultrasound and optoacoustic ultrasound images was associated with observed sensitivity loss of 2.6% [24]. In the current study, which incorporated dedicated training and use of a DST, the loss of observed sensitivity was only 0.4%. Moreover, although the improvement in NLR between ultrasound alone and fused ultrasound and optoacoustic imaging assisted by DST was not statistically significant in the current study, the mean NLR (0.047; 15 readers) was higher than the mean NLR for fused ultrasound and optoacoustic imaging in the Pioneer-01 study (0.094; seven readers) [24].
Fused ultrasound and optoacoustic imaging with DST assistance allowed correct down-classification of 28.9% of reads of benign masses that were initially categorized BI-RADS 4A or greater on ultrasound alone to BI-RADS category 2 or 3. Benign histologies that induce physiologic angiogenesis may contribute to false-positive results that persist when optoacoustic imaging is applied. For example, for masses exhibiting inflammation on biopsy, POM scores were higher for fused ultrasound and optoacoustic imaging with DST assistance than for ultrasound alone. The use of a DST cannot overcome the loss of specificity that results from benign physiologic causes associated with new blood vessel development. Hence, such lesions will likely continue to require needle biopsy.
Fused ultrasound and optoacoustic imaging with DST assistance yielded a false-negative read in 1.3% of malignant masses. Cancers with false-negative reads had significantly shorter distances to the nipple; lesions close to the nipple may be affected by nipple or near-field artifacts that prevent optimal detection of optoacoustic signal. Likewise, fused ultrasound and optoacoustic imaging with DST assistance had lower sensitivity for masses located less than 1 cm from the skin than for masses located 1 cm or farther from the skin, possibly reflecting relative undercolorization of superficial lesions due to overcolorization of adjacent skin and subcutaneous tissues rich in vessels [36]. False-negative reads had a nonsignificant association with smaller tumor size; small cancers may undergo minimal angiogenesis and relative deoxygenation, preventing detection with optoacoustic imaging.
The readers' 95.3% overall agreement with DST results indicates a high degree of reader confidence in the DST. This confidence is expected given that the DST relied heavily on reader input. At the decision threshold, DST had mean sensitivity of 97.0% and mean specificity of 50.9%. By comparison, readers had mean sensitivity of 97.7% and mean specificity of 50.1%. The DST had 44.1% specificity at fixed 98% sensitivity and a pAUC of 0.023, closely aligning with the readers' results for combined ultrasound and optoacoustic imaging with DST assistance (47.2% and 0.024).
The process of using fused ultrasound and optoacoustic imaging to estimate POM requires two steps. The first depends on visual recognition and proper scoring of the ultrasound and optoacoustic imaging features by the radiologist. In the second step, the radiologist uses the machine learning–based DST to objectively and precisely estimate POM from the feature scores. Although DST has the potential to improve POM prediction and/or FNR if the scores are well recognized and assigned, it cannot be expected to improve estimation of POM and/or FNR if the ultrasound and/or optoacoustic imaging features are not properly scored. Thus, training in the scoring of the ultrasound and optoacoustic imaging features, as used in this study, is an important step in implementing the new technology.
The improvement in specificity of fused ultrasound and optoacoustic imaging with DST assistance would likely favorably translate to supplemental screening by means of handheld ultrasound. In this study, 37.5% of masses were malignant, whereas the prevalence of malignancy in the general population, and therefore the pretest probability of a positive result of screening ultrasound, is much lower [37]. Among patients with low cancer prevalence who have undergone mammography with negative results, fused ultrasound and optoacoustic imaging with DST assistance would be anticipated to yield even higher specificity than observed in this study.
This study had limitations. The DST was used only to support readers, and in a small fraction of cases, on the basis of their clinical judgment, the readers did not use the DST-predicted POM. In addition, consistent with the reference standard for the Pioneer-01 study, benign masses included masses assessed BI-RADS category 3 that were stable at 12-month follow-up. In comparison, in clinical practice the standard-of-care follow-up for BIRADS category 3 masses is 2 years. Nonetheless, earlier investigators suggested that 1-year follow-up may be sufficient for probably benign masses, even in patients at high risk [38]. Also, this study did not show that it is feasible to prospectively apply optoacoustic imaging to avoid benign breast biopsies. However, the retrospective design represents typical breast imaging work-flow whereby patient demographics, mammographic findings, and ultrasound findings are integrated to determine whether the POM is greater than 2% and thus whether the sonographic lesion warrants tissue sampling. Finally, although the performance of fused ultrasound and optoacoustic imaging with DST assistance was indirectly compared with the performance of fused ultrasound and optoacoustic imaging without DST assistance from the Pioneer-01 study, the performance of fused ultrasound and optoacoustic imaging without DST assistance was not directly assessed in the current study.
Conclusion
The current study showed that fused ultrasound and optoacoustic imaging with DST assistance, compared with ultrasound alone, resulted in a significant improvement in the primary endpoint of specificity at fixed sensitivity of 98%. The findings indicate that after dedicated reader training and implementation of the DST, the technology has a role in differentiating benign and malignant breast masses. This first-generation FDA-approved system thus has the potential to reduce false-positives reads and biopsies of benign masses without loss of sensitivity for cancer.
Acknowledgments
We thank Mary Hayes, Mary Karst, Su-Ju Lee, Madelyn Lefranc, Jessica Leung, Sharp Malak, Rakesh Parbhu, and Nitin Tanna for their contribution to generating gray-scale and optoacoustic reader study data; Roger Aitchison, for statistical support and manuscript review; Shaan Schaeffer, Seno Medical Instruments, who oversaw the Reader-02 clinical trial; and Thomas Stavros, Seno Medical Instruments, who contributed time and expertise in training the readers and aided with data interpretation and analysis and manuscript review.
Footnotes
Provenance and review: Not solicited; externally peer reviewed.
Peer reviewers: All peer reviewers chose not to disclose their identities.
Supplemental Content
File (22_28470_suppl.pdf)
- Download
- 1.83 MB
References
1.
Perou CM, Sørlie T, Eisen MB, et al. Molecular portraits of human breast tumours. Nature 2000; 406:747–752
2.
Polyak K. Heterogeneity in breast cancer. J Clin Invest 2011; 121:3786–3788
3.
Russnes HG, Lingjærde OC, Børresen-Dale AL, Caldas C. Breast cancer molecular stratification: from intrinsic subtypes to integrative clusters. Am J Pathol 2017; 187:2152–2162
4.
Sørlie T, Perou CM, Tibshirani R, et al. Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci USA 2001; 98:10869–10874
5.
Tirada N, Aujero M, Khorjekar G, et al. Breast cancer tissue markers, genomic profiling, and other prognostic factors: a primer for radiologists. Radio-Graphics 2018; 38:1902–1920
6.
Brennan M, Houssami N. Discussing the benefits and harms of screening mammography. Maturitas 2016; 92:150–153
7.
Nelson HD, O'Meara ES, Kerlikowske K, Balch S, Miglioretti D. Factors associated with rates of false-positive and false-negative results from digital mammography screening: an analysis of registry data. Ann Intern Med 2016; 164:226–235
8.
Butler R, Lavin PT, Tucker FL, et al. Optoacoustic breast imaging: imaging-pathology correlation of optoacoustic features in benign and malignant breast masses. AJR 2018; 211:1155–1170
9.
Schellenberg MW, Hunt HK. Hand-held optoacoustic imaging: a review. Photoacoustics 2018; 11:14–27
10.
Ermilov SA, Khamapirad T, Conjusteau A, et al. Laser optoacoustic imaging system for detection of breast cancer. J Biomed Opt 2009; 14:024007
11.
Kruger RA, Lam RB, Reinecke DR, Del Rio SP, Doyle RP. Photoacoustic angiography of the breast. Med Phys 2010; 37:6096–6100
12.
Menke J. Photoacoustic breast tomography prototypes with reported human applications. Eur Radiol 2015; 25:2205–2213
13.
Oraevsky A. Optoacoustic tomography of the breast. In: Wang L, ed. Photo-acoustic imaging and spectroscopy. Taylor & Francis, 2009:411–430
14.
Su JL, Wang B, Wilson KE, et al. Advances in clinical and biomedical applications of photoacoustic imaging. Expert Opin Med Diagn 2010; 4:497–510
15.
Valluru KS, Wilson KE, Willmann JK. Photoacoustic imaging in oncology: translational preclinical and early clinical experience. Radiology 2016; 280:332–349
16.
Choi JS, Kim MJ, Youk JH, Moon HJ, Suh HJ, Kim EK. US-guided optical tomography: correlation with clinicopathologic variables in breast cancer. Ultrasound Med Biol 2013; 39:233–240
17.
Folkman J. Tumor angiogenesis: therapeutic implications. N Engl J Med 1971; 285:1182–1186
18.
Kitai T, Torii M, Sugie T, et al. Photoacoustic mammography: initial clinical results. Breast Cancer 2014; 21:146–153
19.
Manohar S, Vaartjes SE, van Hespen JC, et al. Initial results of in vivo non-invasive cancer imaging in the human breast using near-infrared photo-acoustics. Opt Express 2007; 15:12277–12285
20.
Zhu Q, Cronin EB, Currier AA, et al. Benign versus malignant breast masses: optical differentiation with US-guided optical imaging reconstruction. Radiology 2005; 237:57–66
21.
Zhu Q, Ricci A Jr, Hegde P, et al. 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; 280:387–397
22.
Zhu Q, You S, Jiang Y, et al. Detecting angiogenesis in breast tumors: comparison of color Doppler flow imaging with ultrasound-guided diffuse optical tomography. Ultrasound Med Biol 2011; 37:862–869
23.
Menezes GLG, Pijnappel RM, Meeuwis C, et al. Downgrading of breast masses suspicious for cancer by using optoacoustic breast imaging. Radiology 2018; 288:355–365
24.
Neuschler EI, Butler R, Young CA, et al. A pivotal study of optoacoustic imaging to diagnose benign and malignant breast masses: a new evaluation tool for radiologists. Radiology 2018; 287:398–412
25.
Neuschler EI, Lavin PT, Tucker FL, et al. Downgrading and upgrading gray-scale ultrasound BI-RADS categories of benign and malignant masses with optoacoustics: a pilot study. AJR 2018; 211:689–700
26.
D'Orsi CJ, Sickles EA, Mendelson EB, et al. ACR BI-RADS Atlas, Breast Imaging Reporting and Data System. American College of Radiology, 2013
27.
Chen T, Guestrin C. XGBoost: a scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Association for Computing Machinery, 2016:785–794
28.
Berg WA, Zhang Z, Cormack JB, Mendelson EB. Multiple bilateral circumscribed masses at screening breast US: consider annual follow-up. Radiology 2013; 268:673–683
29.
Chan CH, Coopey SB, Freer PE, Hughes KS. False-negative rate of combined mammography and ultrasound for women with palpable breast masses. Breast Cancer Res Treat 2015; 153:699–702
30.
Sickles EA. Management of probably benign breast lesions. Radiol Clin North Am 1995; 33:1123–1130
31.
Sadigh G, Carlos RC, Neal CH, Dwamena BA. Ultrasonographic differentiation of malignant from benign breast lesions: a meta-analytic comparison of elasticity and BIRADS scoring. Breast Cancer Res Treat 2012; 133:23–35
32.
Sinha D, Kundaragi NG, Sharma S, Kale SK. Can strain elastography be used in reclassification of indeterminate breast lesions in BIRADS lexicon? A prospective study. Indian J Radiol Imaging 2020; 30:493–499
33.
Pu H, Zhang XL, Xiang LH, et al. The efficacy of added shear wave elastography (SWE) in breast screening for women with inconsistent mammography and conventional ultrasounds (US). Clin Hemorheol Microcirc 2019; 71:83–94
34.
Sadigh G, Carlos RC, Neal CH, Wojcinski S, Dwamena BA. Impact of breast mass size on accuracy of ultrasound elastography vs. conventional B-mode ultrasound: a meta-analysis of individual participants. Eur Radiol 2013; 23:1006–1014
35.
Suvannarerg V, Chitchumnong P, Apiwat W, et al. Diagnostic performance of qualitative and quantitative shear wave elastography in differentiating malignant from benign breast masses, and association with the histological prognostic factors. Quant Imaging Med Surg 2019; 9:386–398
36.
Oraevsky AA, Clingman B, Zalev J, Stavros AT, Yang WT, Parikh JR. Clinical optoacoustic imaging combined with ultrasound for coregistered functional and anatomical mapping of breast tumors. Photoacoustics 2018; 12:30–45
37.
Berg WA, Blume JD, Cormack JB, et al.; ACRIN 6666 Investigators. Combined screening with ultrasound and mammography vs mammography alone in women at elevated risk of breast cancer. JAMA 2008; 299:2151–2163
38.
Barr RG, Zhang Z, Cormack JB, Mendelson EB, Berg WA. Probably benign lesions at screening breast US in a population with elevated risk: prevalence and rate of malignancy in the ACRIN 6666 trial. Radiology 2013; 269:701–712
Information & Authors
Information
Published In
Presented at
Based on a presentation at the Radiological Society of North America 2021 annual meeting, Chicago, IL.
Copyright
© American Roentgen Ray Society.
History
Submitted: August 28, 2022
Revision requested: September 16, 2022
Revision received: October 17, 2022
Accepted: November 20, 2022
Version of record online: December 7, 2022
Keywords
Authors
Funding Information
Supported by Seno Medical Instruments.
Metrics & Citations
Metrics
Citations
Export Citations
To download the citation to this article, select your reference manager software.