Contrast-enhanced Spectral Mammography Imaging Research Articles

DIAGNOSTIC PERFORMANCE OF CONTRAST-ENHANCED SPECTRAL MAMMOGRAPHY: COMPARISON WITH CONTRAST-ENHANCED DYNAMIC MR IMAGING IN PATIENTS WITH SUSPECTED BREAST CANCER

OBJECTIVE: To compare the diagnostic performance of contrast-enhanced spectral mammography (CESM) with dynamic contrast-enhanced magnetic resonance (MR) imaging in terms of the detection of BI-RADS 4 and 5 lesions suspected of breast cancer. MATERIAL AND METHODS: 92 patients with ACR BI-RADS 4 and 5 lesions underwent CESM, MR Imaging, and consequent core biopsy. Two readers assessed the index lesions which were classified as mass lesions, non-mass lesions, and microcalcifications, and scored using a 7-point scoring system. RESULTS: A total of 98 index lesions were detected, including bilateral lesions in six patients. In histopathological analysis, 56 of the lesions were benign (56/98, 57%,) and 42 of the lesions were malignant (42/98, 43%). 55 of the lesions were classified as mass lesions, 18 as non-mass lesions, and 25 as microcalcifications. CESM scored 28 of the lesions (28,6%) as benign, and 70 (71,4%) of the lesions were malignant whereas these results were 30 (30,6%) and 68 (69,4%) for MR Imaging examinations, respectively. The sensitivity of both CESM and MR imaging for depicting the index cancer was 95 % for both modalities. In ROC (Receiver Operating Characteristic) analysis, AUC (Area Under the Curve) was 0.93 (%95 CI:0.870-0.977) for CESM and 0.94 (%95 CI:0.882-0.982) for MR Imaging. There was no statistically significant difference in AUC values between CESM and MR Imaging (p=0.332; p>0.05). CONCLUSIONS: The diagnostic performance of CESM is similar when compared to MR imaging in the detection of index cancers in patients with BI-RADS 4 and 5 lesions. CESM may be used as a confidential diagnostic tool in this regard.

YOLO-based CAD framework with ViT transformer for breast mass detection and classification in CESM and FFDM images

Breast cancer detection is considered a challenging task for the average experienced radiologist due to the variation of the lesions’ size and shape, especially with the existence of high fibro-glandular tissues. The revolution of deep learning and computer vision contributes recently in introducing systems that can provide an automated diagnosis for breast cancer that can act as a second opinion for doctors/radiologists. The most of previously proposed deep learning-based Computer-Aided Diagnosis (CAD) systems mainly utilized Convolutional Neural Networks (CNN) that focuses on local features. Recently, vision transformers (ViT) have shown great potential in image classification tasks due to its ability in learning the local and global spatial features. This paper proposes a fully automated CAD framework based on YOLOv4 network and ViT transformers for mass detection and classification of Contrast Enhanced Spectral Mammography (CESM) images. CESM is an evolution type of Full Field Digital Mammography (FFDM) images that provides enhanced visualization for breast tissues. Different experiments were conducted to evaluate the proposed framework on two different datasets that are INbreast and CDD-CESM that provides both FFDM and CESM images. The model achieved at mass detection a mean Average Precision (mAP) score of 98.69%, 81.52%, and 71.65% and mass classification accuracy of 95.65%, 97.61%, and 80% for INbreast, CE-CESM, and DM-CESM, respectively. The proposed framework showed competitive results regarding the state-of-the-art models in INbreast. It outperformed the previous work in the literature in terms of the F1-score by almost 5% for mass detection in CESM. Moreover, the experiments showed that the CESM could provide more morphological features that can be more informative, especially with the highly dense breast tissues.

Qualitative assessments of density and background parenchymal enhancement on contrast-enhanced spectral mammography associated with breast cancer risk in high-risk women.

To investigate the correlation between the risk of breast cancer for high-risk females and the density and background parenchymal enhancement (BPE) on contrast-enhanced spectral mammography (CESM). Females at high-risk, without breast cancer history and received CESM from July 2016 to December 2017 were retrospectively enrolled. The longest follow-up time was 4.5 years, and patients who developed breast cancer with maximized follow-up time were classified as cancer cohort, while females who did not develop breast cancer were categorized as control cohort. These two cohorts were one-to-one matched in age, family and/or genetic history of breast cancer, menopausal status and BRCA status. The density and BPE at CESM imaging were assessed. Conditional logistic regression was applied to evaluate the relationship between imaging features and breast cancer risk. During the follow-up interval, 90 women at high-risk without history of breast cancer were newly diagnosed. Compared with minimal BPE, increasing BPE levels were associated with the risk of breast cancer among high-risk females in a time interval of 4.5 years (mild: odds ratio [OR]=3.2, p = 0.001; moderate: OR = 4.0, p = 0.002; marked: OR = 11.2, p < 0.001). In addition, females with mild, moderate or marked BPE were four times more likely to be diagnosed with breast cancer than females with minimal BPE in a time interval of 4.5 years (OR = 4.0, p < 0.001). Qualitative CESM BPE assessment may be useful in the prediction of breast cancer risk among high-risk females. • Qualitative CESM BPE assessment may be useful in the prediction of breast cancer risk among high-risk women during the follow-up period of 4.5 years. • The significance of breast density as an independent risk factor is not fully established for high-risk women during the follow-up period of 4.5 years.

An efficient hybrid computer-aided breast cancer diagnosis system with wavelet packet transform and synthetically-generated contrast-enhanced spectral mammography images

Contrast-enhanced spectral mammography (CESM) is an emerging modality for breast cancer diagnosis. This work investigates the feasibility of a computationally-efficient computer-aided diagnosis (CAD) system for breast lesion classification. Moreover, to avoid the need for intravenous contrast agents, we investigated the synthesis of contrast-enhanced (SynCESM) images. A total of 504 pairs of low-energy (LE) and CESM images were collected from 160 female subjects. A semi-automatic active-contour method was used for lesion segmentation. Then, 20 morphological and textural features were extracted. To improve the computational efficiency of the proposed system, the wavelet packet transform (WPT) was applied. Then, the same features were extracted from the WPT-approximated segmented lesions. Using LE images, a sigmoid-kernel SVM classifier exhibited a 90.20% accuracy, an 88.39% sensitivity, an 88.26% specificity, and a 0.92 AUC. The per-frame classification time was significantly reduced from 1.1046 s to 0.0734 s with WPT. Using CESM images, we achieved a 93.26% accuracy, a 95.94% sensitivity, a 93.37% specificity, a 0.94 AUC, and a 0.0657 s classification time. Interestingly, with SynCESM images, reasonable performance was still obtained with a 92.14% accuracy, a 93.87% sensitivity, a 91.58% specificity, a 0.93 AUC and a 0.0657 s classification time. Finally, with hybrid pairs of LE and CESM images, we got the best performance with a 96.87% accuracy, a 97.23% sensitivity, a 95.47% specificity, a 0.98 AUC, and a 0.0696 s classification time. The results demonstrate several advantages of the proposed system including its clinical feasibility, lower complexity, and reduced need for contrast agents via synthetic data generation.

Background enhancement in contrast-enhanced spectral mammography (CESM): are there qualitative and quantitative differences between imaging systems?

ObjectiveTo evaluate the impact of the digital mammography imaging system on overall background enhancement on recombined contrast-enhanced spectral mammography (CESM) images, the overall background enhancement of two different mammography systems was compared.MethodsIn a retrospective single-center study, CESM images of n = 129 female patients who underwent CESM between 2016 and 2019 were analyzed independently by two radiologists. Two mammography machines of different manufacturers were compared qualitatively using a Likert-scale from 1 (minimal) to 4 (marked overall background enhancement) and quantitatively by placing a region of interest and measuring the intensity enhancement. Lesion conspicuity was analyzed using a Likert-scale from 1 (lesion not reliably distinguishable) to 5 (excellent lesion conspicuity). A multivariate regression was performed to test for potential biases on the quantitative results.ResultsSignificant differences in qualitative background enhancement measurements between machines A and B were observed for both readers (p = 0.003 and p < 0.001). The quantitative evaluation showed significant differences in background enhancement with an average difference of 75.69 (99%-CI [74.37, 77.02]; p < 0.001). Lesion conspicuity was better for machine A for the first and second reader respectively (p = 0.009 and p < 0.001). The factor machine was the only influencing factor (p < 0.001). The factors contrast agent, breast density, age, and menstrual cycle could be excluded as potential biases.ConclusionMammography machines seem to significantly influence overall background enhancement qualitatively and quantitatively; thus, an impact on diagnostic accuracy appears possible.Key Points• Overall background enhancement on CESM differs between different vendors qualitatively and quantitatively.• Our retrospective single-center study showed consistent results of the qualitative and quantitative data analysis of overall background enhancement.• Lesion conspicuity is higher in cases of lower background enhancement on CESM.

Radiomic Features Applied to Contrast Enhancement Spectral Mammography: Possibility to Predict Breast Cancer Molecular Subtypes in a Non-Invasive Manner

We aimed to investigate the association between the radiomic features of contrast-enhanced spectral mammography (CESM) images and a specific receptor pattern of breast neoplasms. In this single-center retrospective study, we selected patients with neoplastic breast lesions who underwent CESM before a biopsy and surgical assessment between January 2013 and February 2022. Radiomic analysis was performed on regions of interest selected from recombined CESM images. The association between the features and each evaluated endpoint (ER, PR, Ki-67, HER2+, triple negative, G2-G3 expressions) was investigated through univariate logistic regression. Among the significant and highly correlated radiomic features, we selected only the one most associated with the endpoint. From a group of 321 patients, we enrolled 205 malignant breast lesions. The median age at the exam was 50 years (interquartile range (IQR) 45-58). NGLDM_Contrast was the only feature that was positively associated with both ER and PR expression (p-values = 0.01). NGLDM_Coarseness was negatively associated with Ki-67 expression (p-value = 0.02). Five features SHAPE Volume(mL), SHAPE_Volume(vx), GLRLM_RLNU, NGLDM_Busyness and GLZLM_GLNU were all positively and significantly associated with HER2+; however, all of them were highly correlated. Radiomic features of CESM images could be helpful to predict particular molecular subtypes before a biopsy.

Role of contrast-enhanced spectral mammography in the assessment of residual disease following neoadjuvant chemotherapy in patients with breast cancer

BackgroundNeoadjuvant chemotherapy (NAC) is currently one of the most commonly utilized treatment options for patients with breast cancer. Evaluating residual disease following NAC is vital in post-chemotherapy treatment planning. Underestimation or overestimation of the residual disease may result in incomplete surgical resection or unnecessary aggressive surgical treatment which can compromise patient survival or result in unwanted morbidity. Contrast-enhanced spectral mammography (CESM) is a new imaging modality that can provide both morphological and anatomical characterization of neoplastic breast lesions when MRI is unavailable or contraindicated. The aim of this prospective observational study was to evaluate the value of CESM in assessing the extent of residual disease following NAC and predicting complete response (CR) with histopathology as the gold standard.ResultsA total of 142 patients were included in our study with invasive duct carcinoma the most prevalent histopathological type seen in 81.7% of cases. The average tumor reduction in size was 40.8% by CESM and 35.8% by histopathology. Comparison between maximum tumor diameter after NAC by CESM and histopathology revealed a high degree of correlation (R = 0.918, p < 0.01). In 45.1% of cases, CESM underestimated the residual tumor size by an average of 1.1 cm. Comparison between CESM and histopathology in predicting CR yielded CESM sensitivity, specificity, positive predictive value (PPV), and negative predictive values (NPV) of 96.7%, 84.6%, 97.2%, and 100%, respectively.ConclusionsCESM is a new effective imaging modality in assessing residual disease in breast cancer patients receiving NAC. Post-NAC CESM correlated well with residual tumor size by histopathology with a tendency for underestimation. Careful assessment of CESM images should be made to accurately measure the exact tumor size to avoid overestimation or underestimation potentially increasing the extent of the operation. Proper assessment of the true extent of residual disease can help the surgeon to select the optimal surgical technique ensuring radical treatment of the disease improving patient survival and reducing morbidity.

Mammographic and contrast-enhanced spectral mammography imaging findings of HER2-positive cancers according to hormone receptor status

BackgroundBreast cancer is the leading cause of cancer-related mortality in women. Human epidermal growth factor receptor 2 (HER2) overexpression is seen in 20 out of 100 invasive breast cancers. Among HER2+ patients, two distinct hormone receptor (HR) subtypes can be defined: HR-positive (HR+) and HR-negative (HR−) each of which with unique therapeutic response and survival pattern. Contrast-enhanced spectral mammography (CESM) is an emerging novel imaging modality that offers diagnostic performance comparable to contrast-enhanced MRI. The purpose of this retrospective study was to describe the CESM features of HER2+ breast cancers according to hormone receptor status and to assess whether specific mammographic and CESM imaging features can differentiate between HER2+/HR+ and HER2+/HR− breast cancers potentially aiding treatment planning in HER2+ breast cancer patients.ResultsA total of 61 patients were included. Twenty-nine cases (47.5%) were HER2+/HR+ and 32 cases (52.5%) were HER2+/HR−. No statistically significant difference was found between mammographic imaging presentations and hormonal status. HR- were more likely to be multifocal (P 0.018), rounded or oval (P 0.008), circumscribed (P 0.004), and with associated non-mass enhancement (NME) (P < 0.001). HR+ cancers showed a tendency for irregular shape (P 0.008), spiculated outline (P 0.004), and heterogeneous (P 0.021) or ring (P 0.046) enhancement.ConclusionsHER2+ tumors have different demographic, pathologic and imaging features according to the hormone receptor status. Because the two subtypes of HER2 breast cancer have different clinical outcomes, CESM imaging features can potentially enhance patient outcome by accelerating the diagnosis and treatment.

How Dual-Energy Contrast-Enhanced Spectral Mammography Can Provide Useful Clinical Information About Prognostic Factors in Breast Cancer Patients: A Systematic Review of Literature.

IntroductionIn the past decade, a new technique derived from full-field digital mammography has been developed, named contrast-enhanced spectral mammography (CESM). The aim of this study was to define the association between CESM findings and usual prognostic factors, such as estrogen receptors, progesterone receptors, HER2, and Ki67, in order to offer an updated overview of the state of the art for the early differential diagnosis of breast cancer and following personalized treatments.Materials and MethodsAccording to the PRISMA guidelines, two electronic databases (PubMed and Scopus) were investigated, using the following keywords: breast cancer AND (CESM OR contrast enhanced spectral mammography OR contrast enhanced dual energy mammography) AND (receptors OR prognostic factors OR HER2 OR progesterone OR estrogen OR Ki67). The search was concluded in August 2021. No restriction was applied to publication dates.ResultsWe obtained 28 articles from the research in PubMed and 114 articles from Scopus. After the removal of six replicas that were counted only once, out of 136 articles, 37 articles were reviews. Eight articles alone have tackled the relation between CESM imaging and ER, PR, HER2, and Ki67. When comparing radiological characterization of the lesions obtained by either CESM or contrast-enhanced MRI, they have a similar association with the proliferation of tumoral cells, as expressed by Ki-67. In CESM-enhanced lesions, the expression was found to be 100% for ER and 77.4% for PR, while moderate or high HER2 positivity was found in lesions with non-mass enhancement and with mass closely associated with a non-mass enhancement component. Conversely, the non-enhancing breast cancer lesions were not associated with any prognostic factor, such as ER, PR, HER2, and Ki67, which may be associated with the probability of showing enhancement. Radiomics on CESM images has the potential for non-invasive characterization of potentially heterogeneous tumors with different hormone receptor status.ConclusionsCESM enhancement is associated with the proliferation of tumoral cells, as well as to the expression of estrogen and progesterone receptors. As CESM is a relatively young imaging technique, a few related works were found; this may be due to the “off-label” modality. In the next few years, the role of CESM in breast cancer diagnostics will be more thoroughly investigated.

Multi-feature deep information bottleneck network for breast cancer classification in contrast enhanced spectral mammography

There is considerable variation in the size, shape and location of tumours, which makes it challenging for radiologists to diagnose breast cancer. Automated diagnosis of breast cancer from Contrast Enhanced Spectral Mammography (CESM) can support clinical decision making. However, existing methods fail to obtain an effective representation of the CESM and ignore the relationships between images. In this paper, we investigated for the first time a novel and flexible multimodal representation learning method, multi-feature deep information bottleneck (MDIB), for breast cancer classification in CESM. Specifically, the method incorporated an information bottleneck (IB)-based module to learn the prominent representation that provide concise input while informative for the classification. In addition, we creatively extended IB theory to multi-feature IB, which facilitates the learning of relevant features for classification between CESM images. To validate our method, experiments were conducted on our private and public datasets. The classification results of our method were also compared with those of state-of-the-art methods. The experiment results proved the effectiveness and the efficiency of the proposed method. We release our code at https://github.com/sjq5263/MDIB-for-CESM-classification.

Contrast-Enhanced Spectral Mammography-Based Prediction of Non-Sentinel Lymph Node Metastasis and Axillary Tumor Burden in Patients With Breast Cancer.

PurposeTo establish and evaluate non-invasive models for estimating the risk of non-sentinel lymph node (NSLN) metastasis and axillary tumor burden among breast cancer patients with 1–2 positive sentinel lymph nodes (SLNs).Materials and MethodsBreast cancer patients with 1–2 positive SLNs who underwent axillary lymph node dissection (ALND) and contrast-enhanced spectral mammography (CESM) examination were enrolled between 2018 and 2021. CESM-based radiomics and deep learning features of tumors were extracted. The correlation analysis, least absolute shrinkage and selection operator (LASSO), and analysis of variance (ANOVA) were used for further feature selection. Models based on the selected features and clinical risk factors were constructed with multivariate logistic regression. Finally, two radiomics nomograms were proposed for predicting NSLN metastasis and the probability of high axillary tumor burden.ResultsA total of 182 patients [53.13 years ± 10.03 (standard deviation)] were included. For predicting the NSLN metastasis status, the radiomics nomogram built by 5 selected radiomics features and 3 clinical risk factors including the number of positive SLNs, ratio of positive SLNs, and lymphovascular invasion (LVI), achieved the area under the receiver operating characteristic curve (AUC) of 0.85 [95% confidence interval (CI): 0.71–0.99] in the testing set and 0.82 (95% CI: 0.67–0.97) in the temporal validation cohort. For predicting the high axillary tumor burden, the AUC values of the developed radiomics nomogram are 0.82 (95% CI: 0.66–0.97) in the testing set and 0.77 (95% CI: 0.62–0.93) in the temporal validation cohort.DiscussionCESM images contain useful information for predicting NSLN metastasis and axillary tumor burden of breast cancer patients. Radiomics can inspire the potential of CESM images to identify lymph node metastasis and improve predictive performance.

Comparison between second-look ultrasound and second-look digital breast tomosynthesis in the detection of additional lesions with presurgical CESM.

To compare second-look ultrasound (SL-ultrasound) with second-look digital breast tomosynthesis (SL-DBT) in the detection of additional lesions (ALs) with presurgical contrast-enhanced spectral mammography (CESM). We retrospectively included 121 women with 128 ALs from patients who underwent CESM for presurgical staging at our centre from September 2016 to December 2018. These ALs underwent SL-ultrasound and a retrospective review of DBT (SL-DBT) performed 1-3 weeks prior to CESM to evaluate the performance of each technique individually and in combination. ALs in CESM images were evaluated according to enhancement type (focus, mass, or non-mass), size (<10 mm or >10 mm) and level of suspicion (BI-RADS 2, 3, 4 or 5). Our gold-standard was post-biopsy histology, post-surgical specimen or >24 month negative follow-up. McNemar's test was used for the statistical analysis. Out of the 128 ALs, an imaging correlate was found for 71 (55.5 %) with ultrasound, 79 (61.7%) with DBT, 53 (41.4 %) with DBT and ultrasound, and 97 (75.8%) with ultrasound and/or DBT. SL-DBT demonstrated a higher detection rate vs SL-ultrasound in non-mass enhancement (NME) pattern (p: 0.0325) and ductal carcinoma in situ histological type (p: 0.0081). Adding SL-DBT improved the performance vs SL-ultrasound alone in the overall sample (p: <0.0001) and in every subcategory identified; adding SL-ultrasound to SL-DBT improved the detectability of ALs in the overall sample and in every category except for NME (p: 0.0833), foci (p: 0.0833) and B3 lesions (p: 0.3173). Combined second-look imaging (SL-DBT+ SL-ultrasound) for CESM ALs is superior to SL-DBT alone and SL-ultrasound alone. In B3 lesions, NME, and foci, the analysis of a larger sample could determine whether adding SL-ultrasound to SL-DBT is necessary or not. Thanks to its high sensitivity, CESM is a useful tool in presurgical staging to detect the extent of the disease burden and identify ALs not detected with conventional imaging. Since CESM-guided biopsy systems are still scarcely available in clinical practice, it is necessary to look for other approaches to histologically characterize ALs detected with CESM. In our study, combined second-look imaging (SL-DBT + SL-ultrasound) showed better performance in terms of detectability of ALs, than either SL-DBT or SL-ultrasound alone, and allowed us to identify 91.2% of ALs that turned out to be malignant at final histology; for the remaining 8.8% it was still necessary to perform MRI or MRI-guided biopsy. However, this issue could be solved once CESM-guided biopsies spread in clinical practice. SL-DBT demonstrated a higher detection rate than SL-ultrasound in NME and ductal carcinoma in situ histology.

Categorized contrast enhanced mammography dataset for diagnostic and artificial intelligence research

Contrast-enhanced spectral mammography (CESM) is a relatively recent imaging modality with increased diagnostic accuracy compared to digital mammography (DM). New deep learning (DL) models were developed that have accuracies equal to that of an average radiologist. However, most studies trained the DL models on DM images as no datasets exist for CESM images. We aim to resolve this limitation by releasing a Categorized Digital Database for Low energy and Subtracted Contrast Enhanced Spectral Mammography images (CDD-CESM) to evaluate decision support systems. The dataset includes 2006 images, with an average resolution of 2355 × 1315, consisting of 310 mass images, 48 architectural distortion images, 222 asymmetry images, 238 calcifications images, 334 mass enhancement images, 184 non-mass enhancement images, 159 postoperative images, 8 post neoadjuvant chemotherapy images, and 751 normal images, with 248 images having more than one finding. This is the first dataset to incorporate data selection, segmentation annotation, medical reports, and pathological diagnosis for all cases. Moreover, we propose and evaluate a DL-based technique to automatically segment abnormal findings in images.

Deep learning analysis of contrast-enhanced spectral mammography to determine histoprognostic factors of malignant breast tumours.

ObjectiveTo evaluate if a deep learning model can be used to characterise breast cancers on contrast-enhanced spectral mammography (CESM).MethodsThis retrospective mono-centric study included biopsy-proven invasive cancers with an enhancement on CESM. CESM images include low-energy images (LE) comparable to digital mammography and dual-energy subtracted images (DES) showing tumour angiogenesis. For each lesion, histologic type, tumour grade, estrogen receptor (ER) status, progesterone receptor (PR) status, HER-2 status, Ki-67 proliferation index, and the size of the invasive tumour were retrieved. The deep learning model used was a CheXNet-based model fine-tuned on CESM dataset. The area under the curve (AUC) of the receiver operating characteristic (ROC) curve was calculated for the different models: images by images and then by majority voting combining all the incidences for one tumour.ResultsIn total, 447 invasive breast cancers detected on CESM with pathological evidence, in 389 patients, which represented 2460 images analysed, were included. Concerning the ER, the deep learning model on the DES images had an AUC of 0.83 with the image-by-image analysis and of 0.85 for the majority voting. For the triple-negative analysis, a high AUC was observable for all models, in particularity for the model on LE images with an AUC of 0.90 for the image-by-image analysis and 0.91 for the majority voting. The AUC for the other histoprognostic factors was lower.ConclusionDeep learning analysis on CESM has the potential to determine histoprognostic tumours makers, notably estrogen receptor status, and triple-negative receptor status.Key Points• A deep learning model developed for chest radiography was adapted by fine-tuning to be used on contrast-enhanced spectral mammography.• The adapted models allowed to determine for invasive breast cancers the status of estrogen receptors and triple-negative receptors.• Such models applied to contrast-enhanced spectral mammography could provide rapid prognostic and predictive information.Supplementary InformationThe online version contains supplementary material available at 10.1007/s00330-022-08538-4.

Quantifying lesion enhancement on contrast-enhanced mammography: a review of published data

To review and discuss the current published data on FUNCTIONAL DATA DERIVED FROM contrast-enhanced spectral mammography (CESM) for investigation of breast lesions. Literature searches were conducted in MEDLINE and PUBMED. Due to the novel nature of CESM and sparsity of published literature pertaining to associated functional data, the Medical Subject Headings (MeSH) used were intentionally broad. After inclusion and exclusion criteria, 23 papers were included; 13 pertained to assessment of intensity or pattern of lesion enhancement, and 10 considered textural analysis for lesion assessment, including those using computer-aided detection (CAD) software. Meta-analysis of data was not possible due to heterogeneity of methodology. There is consistent evidence that benign lesions tend to demonstrate different enhancement characteristics to cancers, with benign lesions tending to demonstrate weaker, homogeneous contrast medium uptake. Limited evidence suggests malignant lesions exhibit "wash-out" or decreasing pattern of enhancement, and benign lesions a progressively enhancing one. The application of textural analysis and radiomics to CESM images shows promising results for differentiating benign and malignant lesions, with potential to predict immunohistological features. A large-scale multicentre study, ideally using multivendor CESM equipment, will be needed to confirm this.

Intratumoral and peritumoral radiomics for preoperative prediction of neoadjuvant chemotherapy effect in breast cancer based on contrast-enhanced spectral mammography.

To investigative the performance of intratumoral and peritumoral radiomics based on contrast-enhanced spectral mammography (CESM) to preoperatively predict the effect of the neoadjuvant chemotherapy (NAC) of breast cancers. A total of 118 patients with breast cancer who underwent preoperative CESM and NAC from July 2017 to June 2020 were retrospectively analyzed, and the patients were grouped into training (n= 81) and test sets (n= 37) according to the CESM examination time. NAC effect for each patient was assessed by pathology. Intratumoral and peritumoral radiomics features were extracted from CESM images, and feature selection was performed through the Mann-Whitney U test and least absolute shrinkage and selection operator regression (LASSO). Five radiomics signatures based on intratumoral regions, 5-mm peritumoral regions, 10-mm peritumoral regions, intratumoral regions + 5-mm peritumoral regions, and intratumoral regions + 10-mm peritumoral regions were calculated through a linear combination of selected features weighted by their respective coefficients. The prediction performance of radiomics signatures was assessed by the area under the receiver operator characteristic (ROC) curve, the precision-recall (P-R) curve, the calibration curve, and decision curve analysis (DCA). Ten radiomics features were selected to establish the radiomics signature of intratumoral regions + 5-mm peritumoral regions, which yielded a maximum AUC of 0.85 (95% CI, 0.72-0.98) in the test set. The calibration curves, P-R curves, and DCA showed favorable predictive performance of the five radiomics signatures. The intratumoral and peritumoral radiomics based on CESM exhibited potential for predicting the NAC effect in breast cancer, which could guide treatment decisions. • The intratumoral and peritumoral CESM-based radiomics signatures show good performance in predicting the NAC effect in breast cancer.

Diagnostic Value of Radiomics Analysis in Contrast-Enhanced Spectral Mammography for Identifying Triple-Negative Breast Cancer.

PurposeTo evaluate the value of radiomics analysis in contrast-enhanced spectral mammography (CESM) for the identification of triple-negative breast cancer (TNBC).MethodCESM images of 367 pathologically confirmed breast cancer patients (training set: 218, testing set: 149) were retrospectively analyzed. Cranial caudal (CC), mediolateral oblique (MLO), and combined models were built on the basis of the features extracted from subtracted images on CC, MLO, and the combination of CC and MLO, respectively, in the tumour region. The performance of the models was evaluated through receiver operating characteristic (ROC) curve analysis, the Hosmer-Lemeshow test, and decision curve analysis (DCA). The areas under ROC curves (AUCs) were compared through the DeLong test.ResultsThe combined CC and MLO model had the best AUC and sensitivity of 0.90 (95% confidence interval: 0.85–0.96) and 0.97, respectively. The Hosmer–Lemeshow test yielded a non-significant statistic with p-value of 0.59. The clinical usefulness of the combined CC and MLO model was confirmed if the threshold was between 0.02 and 0.81 in the DCA.ConclusionsMachine learning models based on subtracted images in CESM images were valuable for distinguishing TNBC and NTNBC. The model with the combined CC and MLO features had the best performance compared with models that used CC or MLO features alone.

Improving the classification ability of network utilizing fusion technique in contrast-enhanced spectral mammography.

Contrast-enhanced spectral mammography (CESM) is an effective tool for diagnosing breast cancer with the benefit of its multiple types of images. However, few models simultaneously utilize this feature in deep learning-based breast cancer classification methods. To combine multiple features of CESM and thus aid physicians in making accurate diagnoses, we propose a hybrid approach by taking advantages of both fusion and classificationmodels. We evaluated the proposed method on a CESM dataset obtained from 95 patients between ages ranging from 21 to 74 years, with a total of 760 images. The framework consists of two main parts: a generative adversarial network based image fusion module and a Res2Net-based classification module. The aim of the fusion module is to generate a fused image that combines the characteristics of dual-energy subtracted (DES) and low-energy (LE) images, and the classification module is developed to classify the fused image into benign ormalignant. Based on the experimental results, the fused images contained complementary information of the images of both types (DES and LE), whereas the model for classification achieved accurate classification results. In terms of qualitative indicators, the entropy of the fused images was 2.63, and the classification model achieved an accuracy of 94.784%, precision of 95.016%, recall of 95.912%, specificity of 0.945, F1_score of 0.955, and area under curve of 0.947 on the test dataset,respectively. We conducted extensive comparative experiments and analyses on our in-house dataset, and demonstrated that our method produces promising results in the fusion of CESM images and is more accurate than the state-of-the-art methods in classification of fusedCESM.

Contrast-Enhanced Spectral Mammography Assessment of Patients Treated with Neoadjuvant Chemotherapy for Breast Cancer.

Background: Evaluating the tumor response to neoadjuvant chemotherapy is key to planning further therapy of breast cancer. Our study aimed to evaluate the effectiveness of low-energy and subtraction contrast-enhanced spectral mammography (CESM) images in the detection of complete response (CR) for neoadjuvant chemotherapy (NAC) in breast cancer. Methods: A total of 63 female patients were qualified for our retrospective analysis. Low-energy and subtraction CESM images just before the beginning of NAC and as a follow-up examination 2 weeks before the end of chemotherapy were compared with one another and assessed for compliance with the postoperative histopathological examination (HP). The response to preoperative chemotherapy was evaluated based on the RECIST 1.1 criteria (Response Evaluation Criteria in Solid Tumors). Results: Low-energy images tend to overestimate residual lesions (6.28 mm) and subtraction images tend to underestimate them (2.75 mm). The sensitivity of low-energy images in forecasting CR amounted to 33.33%, while the specificity was 92.86%. In the case of subtraction CESM, the sensitivity amounted to 85.71% and the specificity to 71.42%. Conclusions: CESM is characterized by high sensitivity in the assessment of CR after NAC. The use of only morphological assessment is insufficient. CESM correlates well with the size of residual lesions on histopathological examination but tends to underestimate the dimensions.

Multiview multimodal network for breast cancer diagnosis in contrast-enhanced spectral mammography images.

CESM (contrast-enhanced spectral mammography) is an efficient tool for detecting breast cancer because of its image characteristics. However, among most deep learning-based methods for breast cancer classification, few models can integrate both its multiview and multimodal features. To effectively utilize the image features of CESM and thus help physicians to improve the accuracy of diagnosis, we propose a multiview multimodal network (MVMM-Net). The experiment is carried out to evaluate the in-house CESM images dataset taken from 95 patients aged 21-74 years with 760 images. The framework consists of three main stages: the input of the model, image feature extraction, and image classification. The first stage is to preprocess the CESM to utilize its multiview and multimodal features effectively. In the feature extraction stage, a deep learning-based network is used to extract CESM images features. The last stage is to integrate different features for classification using the MVMM-Net model. According to the experimental results, the proposed method based on the Res2Net50 framework achieves an accuracy of 96.591%, sensitivity of 96.396%, specificity of 96.350%, precision of 96.833%, F1_score of 0.966, and AUC of 0.966 on the test set. Comparative experiments illustrate that the classification performance of the model can be improved by using multiview multimodal features. We proposed a deep learning classification model that combines multiple features of CESM. The results of the experiment indicate that our method is more precise than the state-of-the-art methods and produces accurate results for the classification of CESM images.