Breast mass is one of the most common breast diseases globally. A number of studies have shown that more than 25% of women are affected by breast disease in their lifetimes (1). However, breast masses have a variety of causes, ranging from physiological adenosis to highly invasive malignancy. Breast cancer is the most frequently diagnosed cancer over the world. It is the leading cause of cancer death in women and poses a great public health challenge (2). Early diagnosis and treatment of breast cancer have an important impact on the prognosis of patients. Surgical excision and image-guided biopsy of breast masses are the gold standard for pathological evaluation of breast cancer. The commonly used methods include fine-needle aspiration biopsy, core-needle biopsy, and vacuum-assisted biopsy (3). But the invasiveness of biopsy and possible complications (such as bleeding, infection, etc.) make women uncomfortable.

It is known that angiogenesis appears to be important during cancer progression (4). Previous studies have shown that angiogenesis is a crucial factor affecting local invasion, growth and distant metastasis of breast cancer. The microvascular morphology and functional characteristics of benign and malignant breast masses are inconsistent (5). Related studies had shown that penetrating vessels, branching or disordered vessels may indicate malignant tumors (6). High micro-vessel density (MVD) in masses were more related to the possibility of invasive cancer metastasis and could significantly predict poor survival of breast cancer patients (7). More importantly, malignant breast masses usually have a higher MVD than benign masses (8). At present, the gold standard of MVD detection is obtained by vascular immunohistochemical staining, which is invasive for patients (9). Therefore, it is very important to explore a non-invasive imaging method to detect and evaluate microvasculature in breast masses.

Medical ultrasound imaging is regarded as one of the most common methods for breast masses imaging due to its noninvasiveness, accessibility, affordability (10). Ultrasound is of essential importance in breast cancer detection, diagnosis, and image-guided biopsy for a number of decades. Breast Imaging Reporting and Data System (BI-RADS) is widely used in breast cancer screening in a number of countries. It points out that breast tissue composition, calcification, shape, margin, orientation of masses are important factors to be considered in the classification of masses. However, it is impossible to detect the internal microvascular information of the masses only from the gray scale images. Ultrasound imaging techniques, such as color Doppler flow imaging (CDFI) and contrast-enhanced ultrasound (CEUS), can provide additional blood flow information for the diagnosis of breast masses.

CDFI is the common technique to detect the vascularity and flow information within breast masses noninvasively. It can detect the degree of vascularity within and around breast masses (11). Related studies (12, 13) have shown that most benign breast masses have less blood flow signals, the angiogenesis and the maximum flow velocity in malignant masses are higher than those in benign masses. However, it is challenging for CDFI to detect vessels with relatively slow flows (< 1 cm/s) and relatively small diameters (< 0.1mm) (14, 15). It is unable to display the true micro-vessels and microvascular flow rate (MFR) in the masses. Microvascular flow within the mass does not follow a constant and unidirectional path (16). Therefore, it requires a high-resolution imaging technique to observe the microvascular flow to monitor these functional changes within the masses.

CEUS has been widely used in clinical studies to visualize different vascular structures and tissues. The results showed that CEUS could provide more blood flow information than CDFI (17, 18). These parameters are helpful to differentiate benign and malignant masses and follow-up after local treatment. Some studies have shown that malignant masses tend to be rapid wash-in with hyper-enhancement, enlarged size, present penetrating vessels or crab claw-like pattern, while benign masses tend to be synchronous or slow wash-in with hypo-enhancement, with equal size after enhancement (19). However benign and malignant masses still have an overlap in contrast enhancement patterns. At present, there is no clear classification or guidance of CEUS to avoid the biopsy of indeterminate breast masses (20). Due to the acoustic diffraction limit of the operating ultrasound frequency and insufficient sensitivity, this technique is not capable to visualize the microvasculature at a micron scale. There is currently a lack of a non-invasive diagnostic imaging technique which can visualize microvascular features in clinical practice.

Over the last few years, inspired by optical super-resolution imaging (SRI), ultrasound SRI utilized the localizations of ultrasound contrast agents within blood vessels for the noninvasive display of microvasculature. The introduction of ultrasound SRI technique demonstrated the capability of breaking the ultrasound diffraction limit. After the injection of microbubble contrast agents, the individual microbubbles can be localized and tracked within a subwavelength resolution. The MVD and flow velocity within the masses can then be generated at a spatial scale of micrometers. Previous studies have demonstrated the application of this imaging technique in mouse ear, rat brain and various cancer model (21–23). Opacic et al. and Harput et al. had applied it on human for the first time (24, 25). Several studies had shown that ultrasound SRI technique could be successfully applied on human brain and kidney to further help clinicians for medical diagnosis (26–28). At the same time, ultrasound SRI had been proved to be able to visualize the complex microvasculature in human breast masses and provide rich information. Clinical feasibility of the method was demonstrated for chemotherapy monitoring of breast cancer, the ultrasound SRI can be used to access the early changes of breast cancer after treatment with a vascular-disrupting agent (29). Related researches obtained the MVD and super-resolved velocity map (SRVM) of the human breast masses through ultrasound SRI technique, which further verified the feasibility of using SRI to reveal the microvasculature in breast masses (15, 29). Therefore, by quantifying the extremely low flow rate of single vessels and displaying vascular characteristics at super-resolution, SRI is expected to significantly improve the differential diagnosis of breast masses and the monitoring of breast cancer response to treatment.

However, the feasibility of utilizing ultrasound SRI to differentiate human breast masses remains unknown. Therefore, this study aimed to explore the feasibility and the clinical value of ultrasound SRI compared with CDFI and CEUS images in the differential diagnosis of breast masses.

Ultrasound SRI is a new emerging technique inspired from optical SRI (30, 31). A number of previous studies had demonstrated the capability of applying ultrasound SRI technique in various animals and in human (21–23, 25, 32–34). The value of SRI in differential diagnosis of breast masses remains to be studied. This study applied ultrasound SRI technique on humans and explored its clinical feasibility and diagnostic value in the differential diagnosis of breast masses compared with CEUS and CDFI images. B-mode, CDFI, CEUS and SRI examinations were performed on 46 breast masses respectively. Quantitative parameters were extracted from different datasets and compared with pathological results. The results demonstrated the feasibility of using SRI to visualize microvasculature in human breast masses. Additionally, SRI could display micro-vessels of breast masses at a significantly higher resolution than the conventional CDFI and CEUS images. Furthermore, the MVD and MFR, as quantitative parameters for evaluating the microvasculature, were successfully quantified from high quality SRI and SRVM images respectively. Compared with quantitative parameters (PI, TTP, AUC) obtained from CEUS, MVD has higher diagnostic value in differentiating benign and malignant breast masses.

The feasibility of SRI to display the microvasculature in breast masses at higher resolution relative to CEUS and CDFI was further validated. The result is consistent with the previous studies (15). Relevant studies showed that CDFI could be used to monitor blood flow and perfusion in tissues, but it only provided macro-blood flow information. The actual low-speed blood flow and movement artifacts cannot be distinguished. Therefore, the ability of CDFI to evaluate micro-vessels and microvascular flow rate was very limited (35, 36). CEUS visualizes microcirculation that is usually invisible in CDFI by detecting the enhanced backscatter echo of microbubbles (37). However, due to insufficient sensitivity and the acoustic diffraction limit of the operating US frequencies, neither technique provides a high enough spatial resolution to assess micro-vessels. Ultrasound SRI could achieve a high spatial resolution of vasculature beyond the acoustic diffraction limit (38). Dencks et al. (29) studied the feasibility of SRI for microvascular detection of clinical breast masses in 2019. The results demonstrated that microbubbles tracking-based ultrasound SRI provided visualizations and information beyond the CEUS technique. Ghost et al. (39) and Opacic et al. (24) also proved that ultrasound SRI technology can visualize and characterize the changes of microvascular network in the treatment response of breast cancer mice and patients respectively. These studies fully demonstrated the ability of SRI to show the microvascular of breast masses.

Ultrasound SRI can not only display the micro-vessels of breast masses at the capillary level, but also obtain more information through the quantitative analysis, such as MVD and MFR. In this study, MVD and MFR in breast masses were calculated based on the generation of ultrasound SRI and SRVM images. Chen et al. (38) compared the SRI with histology in MVD estimation of sham kidneys, contralateral kidneys, and injured kidneys at 21-days post injury and injured kidneys at 42-days post injury respectively. The results showed that there was a significant correlation between MVD measured in histology and SRI in the same area, which supported the accuracy of SRI in evaluating micro-vessel density. Song and his colleagues also demonstrated the microvascular perfusion image and blood flow velocity image in vivo rabbit kidney model through SRI technology to calculate MVD and MFR (40). Huang et al. used SRI technology to display the microvascular velocity map in the chorioallantoic membrane of chicken embryos and obtained the MFR (41). Previous studies have proved the accuracy of calculating MVD and MFR using ultrasound SRI.

In addition, quantitative analysis on CEUS and ultrasound SRI were performed to evaluate the ability of each parameter to differentiate between malignant and benign breast masses. It was found that all five parameters (PI, TTP, AUC, MVD and MFR) had statistical differences, among which MVD had the highest correlation. For CEUS, the results showed that the PI and AUC of malignant breast masses were significantly higher than those of benign breast masses, while TTP of malignant breast masses was significantly lower than that of benign breast masses. These findings can be explained by the earlier, faster and higher enhancement of malignant masses. This result showed the consistency with a number of previous studies (3). The research of Jung et al. indicated the CEUS perfusion parameters PI and AUC provided more information for assessing the risk of malignant breast masses (42). Janu et al. also proved that malignant masses showed statistically significantly lower TTP parameters than benign masses (18). For SRI, the results suggested that the MVD and MFR within the malignant breast masses were significantly higher than those within the benign breast masses. This may be related to the high density of blood vessels and the high incidence of arteriovenous shunts in breast malignant masses. It is well-known that angiogenesis is a hallmark of cancer and promotes tumor progression and metastasis (43). Relevant studies have shown that MVD is often regarded as a surrogate marker of angiogenesis in tumors (44) and is considered as a risk factor of metastasis and predicts poor prognosis of breast cancer patients (45). The Felix et al. determined MVD by counting the CD31-positive vessels in sections of breast biopsies and found that women with in situ or invasive breast cancer were more likely to have higher tissue MVD than women with benign masses. MVD, as a risk factor of breast cancer, is positively associated with breast cancer incidence (46). Studies of Krüger et al. (43) and Tolaney et al. (47) implicated that MVD could significantly predicted response to neoadjuvant in breast cancer. In addition, SRI of the tumor vasculature had also been clinically verified for chemotherapy monitoring of breast cancer (24, 29). However, there is no study to distinguish malignant masses from benign masses by showing the microvascular networks in breast masses using SRI. This study showed that SRI had a potential to become a new imaging method for differential diagnosis of breast masses by revealing microvascular and calculating quantitative indicators (such as MVD and MFR).

There are some limitations existing in this study. First, a larger sample size may be required to obtain a more solid conclusion to differentiate the benign and malignant breast masses using ultrasound SRI. However, the data acquisition of SRI dataset is challenging as this applies to all SRI studies on humans (15, 29). There were only less than 10 SRI cases shown in the previous studies. Second, the SRI generation is time-consuming and requires off-line processing. The graphics processing unit (GPU) parallel processing may be required to be integrated into the SRI processing in the future to further accelerate the SRI processing. Third, the ultrasound SRI and SRVM obtained in this study were two-dimensional imaging, therefore, the out of plane microvasculature could not been revealed. A 3D SRI technique equipped with a 2D array probe is expected to overcome this problem in the future. Finally, pathological underestimation may happen using ultrasound-guided breast biopsy. This is because either ultrasound-guided breast biopsy or surgical excision was used in this study to obtain the histopathological results.

In conclusion, it is feasible for SRI to show the details of micro-vessels in breast masses at a submicron resolution level compared with CDFI and CEUS. Meanwhile, more micro-vessel information can be obtained by calculating MVD and MFR. The MFR and MVD within the malignant breast masses were significantly higher than those within the benign breast masses respectively. Compared with the quantitative parameters of CEUS, MVD had a stronger significance in differentiating benign and malignant breast masses. This work provides an improved approach for the differential diagnosis of breast masses besides conventional B-mode, CDFI and CEUS images. With the study of breast masses with larger sample size, and the extraction of more microvascular parameters, ultrasound SRI technology is expected to further contribute to the classification of different pathological types of breast masses in clinics.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by The Institutional Review Board (IRB) of China Resources & Wisco General Hospital. The patients/participants provided their written informed consent to participate in this study. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

GZ, Y-ML, NL, JY, X-YJ, M-HY, and H-MH contributed to the collection of relevant literature. Y-ML and NL contributed to the literature analysis and manuscript preparation. GZ and Y-ML sorted out the literature and wrote the manuscript. S-EZ, X-WC, and H-RY were responsible for design of this article and provided data acquisition, analysis, and interpretation. All authors contributed to the article and approved the submitted version.

This work was supported by the Key Research and Development Project of Hubei Province (No.2020BCB022), the joint fund project of the Hubei Provincial Health and Family Planning Commission (No. WJ2019H197) and the research project of Wuhan Municipal Health Commission (No.WX21C07).

The authors would like to thank the helps from Prof. Mengxing Tang at Imperial College London in the completion of this article and the reviewers for reviewing this article.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.