Microcalcification Research Articles

Mammograms enhancement based on multifractal measures for microcalcifications detection

Breast cancer is the most common cancer in women and represents as a leading cause of death in the world (https://gco.iarc.fr/today/home). Microcalcifications (MCs) are the essential signs of precancerous cells. Their small size makes them difficult to detect and locate, hence the need of developing computer-aided detection systems for early detection of breast cancer. In this paper, an approach of MCs detection is proposed. Our system includes three phases. In the first, we start by a pre-processing step to remove various noises, followed by a step of intensity enhancement based on the haze removal algorithm. The third step is based on multifractal measures to construct the α-image which enhance MCs contrast. The proposed method was tested on three databases with a set of 371 images and evaluated in terms of PSNR and sensitivity. The obtained results are very significant and better compared to other approaches proposed in the literature.

Detection of microcalcifications in photon-counting dedicated breast-CT using a deep convolutional neural network: Proof of principle

ObjectiveIn this study, we investigate the feasibility of a deep Convolutional Neural Network (dCNN), trained with mammographic images, to detect and classify microcalcifications (MC) in breast-CT (BCT) images. MethodsThis retrospective single-center study was approved by the local ethics committee. 3518 icons generated from 319 mammograms were classified into three classes: “no MC” (1121), “probably benign MC” (1332), and “suspicious MC” (1065). A dCNN was trained (70% of data), validated (20%), and tested on a “real-world” dataset (10%). The diagnostic performance of the dCNN was tested on a subset of 60 icons, generated from 30 mammograms and 30 breast-CT images, and compared to human reading. ROC analysis was used to calculate diagnostic performance. Moreover, colored probability maps for representative BCT images were calculated using a sliding-window approach. ResultsThe dCNN reached an accuracy of 98.8% on the “real-world” dataset. The accuracy on the subset of 60 icons was 100% for mammographic images, 60% for “no MC”, 80% for “probably benign MC” and 100% for “suspicious MC”. Intra-class correlation between the dCNN and the readers was almost perfect (0.85). Kappa values between the two readers (0.93) and the dCNN were almost perfect (reader 1: 0.85 and reader 2: 0.82). The sliding-window approach successfully detected suspicious MC with high image quality. The diagnostic performance of the dCNN to classify benign and suspicious MC was excellent with an AUC of 93.8% (95% CI 87, 4%–100%). ConclusionDeep convolutional networks can be used to detect and classify benign and suspicious MC in breast-CT images.

Detection and classification of microcalcifications in mammograms images using difference filter and Yolov4 deep learning model

In this study, we focus on increasing the visibility of microcalcifications (MCs) in mammogram images by means of the difference filter and classifying the visibility-increased MCs by using Yolov4 deep learning model. The same classification experiments are reperformed for also the widely used Faster R-CNN deep learning model to compare with the proposed approach. For this aim, the difference filter is applied to the sections taken from normal and abnormal labeled mammogram images, and the filtered images are used as inputs to Yolov4 and Faster R-CNN models in order to classify as normal and abnormal. In order to show the contribution of the difference filter to the classification success, the experiments are reimplemented without using the difference filter. The difference filter based on the neighborhood relations of the image pixels significantly improves the classification success ratios of the classifier models used in the study since it increases especially the visibility of the rounded edges and makes microcalcifications in the image more prominent. As a result, the experiments show that the use of deep learning models together with the difference filter contributes significantly to the classification success. Finally, this study gives rise to the idea that it can greatly contribute to studies reading of the mammograms with MCs (abnormal) highlighted by the use of difference filter.

Multi-scale cascaded networks for synthesis of mammogram to decrease intensity distortion and increase model-based perceptual similarity.

Synthetic digital mammogram (SDM) is a 2D image generated from digital breast tomosynthesis (DBT) and used as a substitute for a full-field digital mammogram (FFDM) to reduce the radiation dose for breast cancer screening. The previous deep learning-based method used FFDM images as the ground truth, and trained a single neural network to directly generate SDM images with similar appearances (e.g., intensity distribution, textures) to the FFDM images. However, the FFDM image has a different texture pattern from DBT. The difference in texture pattern might make the training of the neural network unstable and result in high-intensity distortion, which makes it hard to decrease intensity distortion and increase perceptual similarity (e.g., generate similar textures) at the same time. Clinically, radiologists want to have a 2D synthesized image that feels like an FFDM image in vision and preserves local structures such as both mass and microcalcifications (MCs) in DBT because radiologists have been trained on reading FFDM images for a long time, while local structures are important for diagnosis. In this study, we proposed to use a deep convolutional neural network to learn the transformation to generate SDM from DBT. To decrease intensity distortion and increase perceptual similarity, a multi-scale cascaded network (MSCN) is proposed to generate low-frequency structures (e.g., intensity distribution) and high-frequency structures (e.g., textures) separately. The MSCN consist of two cascaded sub-networks: the first sub-network is used to predict the low-frequency part of the FFDM image; the second sub-network is used to generate a full SDM image with textures similar to the FFDM image based on the prediction of the first sub-network. The mean-squared error (MSE) objective function is used to train the first sub-network, termed low-frequency network, to generate a low-frequency SDM image. The gradient-guided generative adversarial network's objective function is to train the second sub-network, termed high-frequency network, to generate a full SDM image with textures similar to the FFDM image. 1646 cases with FFDM and DBT were retrospectively collected from the Hologic Selenia system for training and validation dataset, and 145 cases with masses or MC clusters were independently collected from the Hologic Selenia system for testing dataset. For comparison, the baseline network has the same architecture as the high-frequency network and directly generates a full SDM image. Compared to the baseline method, the proposed MSCN improves the peak-to-noise ratio from 25.3 to 27.9 dB and improves the structural similarity from 0.703 to 0.724, and significantly increases the perceptual similarity. The proposed method can stabilize the training and generate SDM images with lower intensity distortion and higher perceptual similarity.

Classification of Mammographic ROI for Microcalcification Detection Using Multifractal Approach

Microcalcifications (MCs) are the main signs of precancerous cells. The development of aided-system for their detection has become a challenge for researchers in this field. In this paper, we propose a system for MCs detection based on the multifractal approach that classifies mammographic ROIs into normal (healthy) or abnormal ROIs containing MCs. The proposed method is divided into four main steps: a mammogram pre-processing step based on breast selection, breast density reduction using haze removal algorithm and contrast enhancement using multifractal measures. The second step consists of extracting the normal and abnormal ROIs and calculating the multifractal spectrum of each ROI. The next step represents the extraction of the multifractal features from the multifractal spectrum and the GLCM characteristics of each ROI. The last step is the classification of ROIs where three classifiers are tested (KNN, DT, and SVM). The system is evaluated on images from the INbreast database (308 images) with a total of 2688 extracted ROIs (1344 normal, 1344 with MC) from different BI-RADS classes. In this study, the SVM classifier gave the best classification results with a sensitivity, specificity, and precision of 98.66%, 97.77%, and 98.20% respectively. These results are very satisfactory and remarkable compared to the literature.

Real-Time Ultrasound Detection of Breast Microcalcifications Using Multifocus Twinkling Artifact Imaging.

Detecting microcalcifications (MCs) in real time is important in the guidance of many breast biopsies. Due to its capability in visualizing biopsy needles without radiation hazards, ultrasound imaging is preferred over X-ray mammography, but it suffers from low sensitivity in detecting MCs. Here, we present a new nonionizing method based on real-time multifocus twinkling artifact (MF-TA) imaging for reliably detecting MCs. Our approach exploits time-varying TAs arising from acoustic random scattering on MCs with rough or irregular surfaces. To obtain the increased intensity of the TAs from MCs, in MF-TA, acoustic transmit parameters, such as the transmit frequency, the number of focuses and f-number, were optimized by investigating acoustical characteristics of MCs. A real-time MF-TA imaging sequence was developed and implemented on a programmable ultrasound research system, and it was controlled with a graphical user interface during real-time scanning. From an in-house 3D phantom and ex vivo breast specimen studies, the MF-TA method showed outstanding visibility and high-sensitivity detection for MCs regardless of their distribution or the background tissue. These results demonstrated that this nonionizing, noninvasive imaging technique has the potential to be one of effective image-guidance methods for breast biopsy procedures.

Breast cancer segmentation using K-means clustering and optimized region-growing technique

Breast cancer is one of the major causes of death among women, and early detection may decrease the aggressiveness of the disease. The goal of this paper is to create an automated system that can classify digital mammogram images into benign and malignant. This paper presents a new detection technique of micro-calcifications in mammogram images. An automated technique for identifying breast microcalcifications (MCs) proposed utilizing two-level segmentation processes, first crop the breast area from the image using k-means clustering, then, an optimized region growing (ORG) approach has been used, where multi-seed points and thresholds are generated optimally depending on the color values of the image pixels. Then the texture features are extracted based on Haralick definitions of texture analysis. In addition, three features (cross-correlation coefficient, pearson correlation, and average area of segmented spots) are obtained from the segmented image. Support vector machine (SVM) classifier evaluate the efficiency of the system utilizing the images from the digital database for screening mammography (DDSM) dataset. The results were obtained by utilizing 355 images for training and 85 images for testing. The proposed system's sensitivity reached up to 97.05%, the specificity obtained is 98.52%, and accuracy is 98.2%.

A Mammography-Based Nomogram for Prediction of Malignancy in Breast Suspicious Calcification

To establish a predictive nomogram for malignancy risk stratification of micro-calcifications (MCCs) detected on mammography. Consecutive mammograms from January 2017 to March 2021 were retrospectively reviewed. Traditional clinical features were recorded and mammographic features were estimated according to the 5th BI-RADS. A nomogram was developed to graphically predict the malignancy risk based on multivariate logistic regression analysis. The discrimination and calibration performance of the prediction model was assessed. There were 123 cases of suspicious MCCs with final pathological results identified with a malignancy rate of 55.2%. The malignancy rates of subgroups divided according to the morphology and distribution of MCCs, age, menopausal status and the maximum diameter of MCCs were significantly different. Multivariate logistic analysis showed that a menopause status of postmenopausal, maximum diameters of MCCs ≥2 cm, the morphology of MCCs as fine pleomorphic or fine linear or branching, and the distribution of MCCs as linear or segmental were predictive of a higher probability of malignancy. A prediction nomogram was developed based on four risk factors, including menopausal status as well as the maximum diameters, distribution and morphology of the MCCs. The AUC of that nomogram was 0.839 (95%CI:0.771-0.903). In mammography, the morphology, distribution and maximum diameter of MCCs, and the menopausal status are independent predictors of malignant suspicious MCCs and are readily available in the clinical setting. The nomogram developed in this study for individualized malignancy risk stratification of suspicious MCCs shows a reliable discrimination performance.

Automatic Pectoral Muscle Removal and Microcalcification Localization in Digital Mammograms.

ObjectivesBreast cancer is the most common cancer diagnosed in women, and microcalcification (MCC) clusters act as an early indicator. Thus, the detection of MCCs plays an important role in diagnosing breast cancer.MethodsThis paper presents a methodology for mammogram preprocessing and MCC detection. The preprocessing method employs automatic artefact deletion and pectoral muscle removal based on region-growing segmentation and polynomial contour fitting. The MCC detection method uses a convolutional neural network for region-of-interest (ROI) classification, along with morphological operations and wavelet reconstruction to reduce false positives (FPs).ResultsThe methodology was evaluated using the mini-MIAS and UTP datasets in terms of segmentation accuracy in the preprocessing phase, as well as sensitivity and the mean FP rate per image in the MCC detection phase. With the mini-MIAS dataset, the proposed methods achieved accuracy scores of 99% for breast segmentation and 95% for pectoral segmentation, a sensitivity score of 82% for MCC detection, and an FP rate per image of 3.27. With the UTP dataset, the methods achieved accuracy scores of 97% for breast segmentation and 91% for pectoral segmentation, a sensitivity score of 78% for MCC detection, and an FP rate per image of 0.74.ConclusionsThe proposed preprocessing method outperformed the state-of-the-art methods for breast segmentation and achieved relatively good results for pectoral muscle removal. Furthermore, the MCC detection module achieved the highest test accuracy in identifying potential ROIs with MCCs compared to other methods.

A new conditional region growing approach for microcalcification delineation in mammograms.

Microcalcifications (MCs) are considered as the first indicator of breast cancer development. Their morphology, in terms of shape and size, is considered as the most important criterion that determines their malignity degrees. Therefore, the accurate delineation of MC is a cornerstone step in their automatic diagnosis process. In this paper, we propose a new conditional region growing (CRG) approach with the ability of finding the accurate MC boundaries starting from selected seed points. The starting seed points are determined based on regional maxima detection and superpixel analysis. The region growing step is controlled by a set of criteria that are adapted to MC detection in terms of contrast and shape variation. These criteria are derived from prior knowledge to characterize MCs and can be divided into two categories. The first one concerns the neighbourhood searching size. The second one deals with the analysis of gradient information and shape evolution within the growing process. In order to prove the effectiveness and the reliability in terms of MC detection and delineation, several experiments have been carried out on MCs of various types, with both qualitative and quantitative analysis. The comparison of the proposed approach with state-of-the art proves the importance of the used criteria in the context of MC delineation, towards a better management of breast cancer. Graphical Abstract Flowchart of the proposed approach.

Raman analysis of microcalcifications in male breast cancer

Microcalcifications (MCs) are important disease markers for breast cancer. Many studies were conducted on their characterization in female breast cancer (FBC), but no information is available on their composition in male breast cancer (MBC). Raman spectroscopy (RS) is a molecular spectroscopy that can rapidly explore the biochemical composition of MCs without requiring any staining protocol.In this study, we optimized an algorithm to identify the mineral components present in MCs from Raman images. The algorithm was then used to study and compare MCs identified on breast cancer pieces from male and female patients. In total, we analyzed 41 MCs from 5 invasive MBC patients and 149 MCs from 13 invasive FBC patients.Results show that hydroxyapatite is the most abundant type of calcium both in MBC and FBC. However, some differences in the amount and distribution of calcium minerals are present between the two groups. Besides, we observed that MCs in MBC have a higher amount of organic material (collagen) than FBC.To the best of our knowledge, this study provides the first overview of the composition of MCs present in MBC patients; and suggests that these patients have specific features that differentiate them from the previously studied FBC. Our result support thus the need for studies designed explicitly to the understanding of MBC.

Characterization of Suspicious Microcalcifications on Mammography Using 2D Shear-Wave Elastography

Our aim was to investigate the correlations between the findings of two-dimensional shear-wave elastography (2D-SWE) and the histopathologic results of microcalcifications (MCs) visualized using ultrasonography (USG). Fifty people with suspicious MCs without accompanying mass were evaluated. They underwent USG and 2D-SWE before USG-guided tru-cut biopsy. SWE values and histopathologic features were compared statistically. The variables between groups were analyzed using the Mann–Whitney U test. Receiver operating characteristic analysis was performed and cut-off values determined to discriminate malignancy, invasiveness and high grade. Pathology confirmed 27 malignant lesions (18 invasive ductal carcinomas, one invasive lobular and eight ductal carcinomas in situ) and 23 benign ones. There was a statistically significant difference between the SWE values of malignant and benign MCs (p < 0.001). The diagnostic performance of SWE for malignancy, invasiveness and high grade were as follows, repectively: sensitivity (93%, 83%, 88%), specificity (91%, 88%, 53%), positive predictive value (93%, 94%, 44%), negative predictive value (91%, 70%, 90%) and area under the curve (0.952, 0.885, 0.776). Cut-off values were determined as 57 kPa for malignancy, 124 kPa for invasiveness and 124.5 kPa for high grade. In conclusion, SWE is a useful method in clinical practice for characterizing MCs that can be visualized with USG.

Predicting the pathological status of mammographic microcalcifications through a radiomics approach

Objective The study aimed to develop a machine learning (ML)-coupled interpretable radiomics signature to predict the pathological status of non-palpable suspicious breast microcalcifications (MCs).Methods We enrolled 463 digital mammographical view images from 260 consecutive patients detected with non-palpable MCs and BI-RADS scored at 4 (training cohort, n = 428; independent testing cohort, n = 35) in the First Affiliated Hospital of Nanjing Medical University between September 2010 and January 2019. Subsequently, 837 textures and 9 shape features were subsequently extracted from each view and finally selected by an XGBoost-embedded recursive feature elimination technique (RFE), followed by four machine learning-based classifiers to build the radiomics signature.Results Ten radiomic features constituted a malignancy-related signature for breast MCs as logistic regression (LR) and support vector machine (SVM) yielded better positive predictive value (PPV)/sensitivity (SE), 0.904 (95% CI, 0.865–0.949)/0.946 (95% CI, 0.929–0.977) and 0.891 (95% CI, 0.822–0.939)/0.939 (95% CI, 0.907–0.973) respectively, outperforming their negative predictive value (NPV)/specificity (SP) from 10-fold cross-validation (10FCV) of the training cohort. The optimal prognostic model was obtained by SVM with an area under the curve (AUC) of 0.906 (95% CI, 0.834–0.969) and accuracy (ACC) 0.787 (95% CI, 0.680–0.855) from 10FCV against AUC 0.810 (95% CI, 0.760–0.960) and ACC 0.800 from the testing cohort.Conclusion The proposed radiomics signature dependens on a set of ML-based advanced computational algorithms and is expected to identify pathologically cancerous cases from mammographically undecipherable MCs and thus offer prospective clinical diagnostic guidance.

Detection of Microcalcifications in Spiral Breast Computed Tomography with Photon-Counting Detector Is Feasible: A Specimen Study.

The primary objective of the study was to compare a spiral breast computed tomography system (SBCT) to digital breast tomosynthesis (DBT) for the detection of microcalcifications (MCs) in breast specimens. The secondary objective was to compare various reconstruction modes in SBCT. In total, 54 breast biopsy specimens were examined with mammography as a standard reference, with DBT, and with a dedicated SBCT containing a photon-counting detector. Three different reconstruction modes were applied for SBCT datasets (Recon1 = voxel size (0.15 mm)3, smooth kernel; Recon2 = voxel size (0.05 mm)3, smooth kernel; Recon3 = voxel size (0.05 mm)3, sharp kernel). Sensitivity and specificity of DBT and SBCT for the detection of suspicious MCs were analyzed, and the McNemar test was used for comparisons. Diagnostic confidence of the two readers (Likert Scale 1 = not confident; 5 = completely confident) was analyzed with ANOVA. Regarding detection of MCs, reader 1 had a higher sensitivity for DBT (94.3%) and Recon2 (94.9%) compared to Recon1 (88.5%; p < 0.05), while sensitivity for Recon3 was 92.4%. Respectively, reader 2 had a higher sensitivity for DBT (93.0%), Recon2 (92.4%), and Recon3 (93.0%) compared to Recon1 (86.0%; p < 0.05). Specificities ranged from 84.7–94.9% for both readers (p > 0.05). The diagnostic confidence of reader 1 was better with SBCT than with DBT (DBT 4.48 ± 0.88, Recon1 4.77 ± 0.66, Recon2 4.89 ± 0.44, and Recon3 4.75 ± 0.72; DBT vs. Recon1/2/3: p < 0.05), while reader 2 found no differences. Sensitivity and specificity for the detection of MCs in breast specimens is equal for DBT and SBCT when a small voxel size of (0.05 mm)3 is used with an equal or better diagnostic confidence for SBCT compared to DBT.

Deep Convolutional Neural Network With Adversarial Training for Denoising Digital Breast Tomosynthesis Images.

Digital breast tomosynthesis (DBT) is a quasi-three-dimensional imaging modality that can reduce false negatives and false positives in mass lesion detection caused by overlapping breast tissue in conventional two-dimensional (2D) mammography. The patient dose of a DBT scan is similar to that of a single 2D mammogram, while acquisition of each projection view adds detector readout noise. The noise is propagated to the reconstructed DBT volume, possibly obscuring subtle signs of breast cancer such as microcalcifications (MCs). This study developed a deep convolutional neural network (DCNN) framework for denoising DBT images with a focus on improving the conspicuity of MCs as well as preserving the ill-defined margins of spiculated masses and normal tissue textures. We trained the DCNN using a weighted combination of mean squared error (MSE) loss and adversarial loss. We configured a dedicated x-ray imaging simulator in combination with digital breast phantoms to generate realistic in silico DBT data for training. We compared the DCNN training between using digital phantoms and using real physical phantoms. The proposed denoising method improved the contrast-to-noise ratio (CNR) and detectability index (d') of the simulated MCs in the validation phantom DBTs. These performance measures improved with increasing training target dose and training sample size. Promising denoising results were observed on the transferability of the digital-phantom-trained denoiser to DBT reconstructed with different techniques and on a small independent test set of human subject DBT images.

Assessment of task-based performance from five clinical DBT systems using an anthropomorphic breast phantom.

Digital breast tomosynthesis (DBT) is a limited-angle tomographic breast imaging modality that can be used for breast cancer screening in conjunction with full-field digital mammography (FFDM) or synthetic mammography (SM). Currently, there are five commercial DBT systems that have been approved by the U.S. FDA for breast cancer screening, all varying greatly in design and imaging protocol. Because the systems are different in technical specifications, there is a need for a quantitative approach for assessing them. In this study, the DBT systems are assessed using a novel methodology with an inkjet-printed anthropomorphic phantom and four alternative forced choice (4AFC) study scheme. A breast phantom was fabricated using inkjet printing and parchment paper. The phantom contained 5-mm spiculated masses fabricated with potassium iodide (KI)-doped ink and microcalcifications (MCs) made with calcium hydroxyapatite. Images of the phantom were acquired on all five systems with DBT, FFDM, and SM modalities where available using beam settings under automatic exposure control. A 4AFC study was conducted to assess reader performance with each signal under each modality. Statistical analysis was performed on the data to determine proportion correct (PC), standard deviations, and levels of significance. For masses, overall detection was highest with DBT. The difference in PC was statistically significant between DBT and SM for most systems. A relationship was observed between increasing PC and greater gantry span. For MCs, performance was highest with DBT and FFDM compared to SM. The difference between PC of DBT and PC of SM was statistically significant for all manufacturers. This methodology represents a novel approach for evaluating systems. This study is the first of its kind to use an inkjet-printed anthropomorphic phantom with realistic signals to assess performance of clinical DBT imaging systems.

Deep Learning Features for Discriminating Between Benign and Malignant Microcalcification Lesions

Accurate diagnosis of microcalcification (MC) lesions in mammograms as benign or malignant is a challenging clinical task. In this study we investigate the potential discriminative power of deep learning features in MC lesion diagnosis. We consider two types of deep learning networks, of which one is a convolutional neural network developed for MC detection and the other is a denoising autoencoder network. In the experiments, we evaluated both the separability between malignant and benign lesions and the classification performance of image features from these two networks using Fisher's linear discriminant analysis on a set of mammographic images. The results demonstrate that the deep learning features from the MC detection network are most discriminative for classification of MC lesions when compared to both features from the autoencoder network and traditional handcrafted texture features.

3D microcalcification detection using a color Doppler twinkling artifact with optimized transmit conditions: Preliminary results.

Mammography is the only method that has been proven to detect breast microcalcifications (MCs), but the sensitivity of mammography varies according to breast density. This paper proposes an ultrasound (US) color Doppler twinkling artifact (CDTA) method with optimized transmit conditions to identify breast MCs without ionizing radiation. The transmit conditions for US color Doppler imaging (CDI) were optimized to enhance the sensitivity of the twinkling artifact (TA) that arises from random scattering on rough surfaces of breast MCs. To validate the proposed breast MC detection method, a chicken breast phantom with MC particles (groups of particles <400 and <240 ) was fabricated and scanned by a digital mammography system and an US research platform by an L11-5v linear array probe with a three-dimensional (3D) motion tracking system. From the phantom experiment, the proposed 3D CDTA imaging method with optimized transmit conditions (i.e., a center frequency of 5.0MHz, an f-number of 1.3, and a peak negative pressure of 1.83MPa) successfully detected all 16 MC particles, comparable to detection with mammography. For a human breast surgical specimen in the ex vivo study, all 10 MC clusters, marked by a radiologist on the mammogram, were identified with the proposed 3D CDTA imaging method. In the phantom and ex vivo breast specimen studies, the proposed 3D CDTA imaging method successfully detected MCs, and the spatial localization was highly correlated with the mammogram results. These results indicate that the proposed 3D CDTA imaging method has great potential for the detection of MCs without ionizing radiation.

3D Context-Aware Convolutional Neural Network for False Positive Reduction in Clustered Microcalcifications Detection.

False positives (FPs) reduction is indispensable for clustered microcalcifications (MCs) detection in digital breast tomosynthesis (DBT), since there might be excessive false candidates in the detection stage. Considering that DBT volume has an anisotropic resolution, we proposed a novel 3D context-aware convolutional neural network (CNN) to reduce FPs, which consists of a 2D intra-slices feature extraction branch and a 3D inter-slice features fusion branch. In particular, 3D anisotropic convolutions were designed to learn representations from DBT volumes and inter-slice information fusion isonly performed on the feature map level, which could avoid the influence of anisotropic resolution of DBT volume. The proposed method was evaluated on a large-scale Chinese women population of 877 cases with 1754 DBT volumes and compared with 8 related methods. Experimental results show that the proposed network achieved the best performance with an accuracy of 92.68% for FPs reduction with an AUC of 97.65%, and the FPs are 0.0512 per DBT volume at a sensitivity of 90%. This also proved that making full use of 3D contextual information of DBT volume can improve the performance of the classification algorithm.

Clinical analysis of 42 primary hyperthyroidism patients complicated with thyroid microcarcinoma

Objective To analyze the clinical data of primary hyperthyroidism patients complicated with thyroid microcarcinoma, and to guide clinical diagnosis and treatment. Methods From January 2013 to March 2019, the clinical data of 42 cases of hyperthyroidism with thyroid microcarcinoma(study group) and 410 cases of thyroid microcarcinoma without hyperthyroidism(control group) who underwent surgical treatment in the People's Hospital Affiliated to Hubei University of Medicine were retrospectively analyzed.The epidemiological characteristics, clinical diagnosis and treatment, clinical stage, B-ultrasound characteristics, thyroid function and antibody, pathological data and prognosis of the two groups were compared. Results In study group and control group, the proportion of microcalcification detected by B-ultrasonography was 61.9%(26/42) and 33.4%(137/410), respectively, the difference was statistically significant between the two groups(χ2=13.411, P<0.05). In study group and control group, 47.62%(20/42) and 69.02%(283/410) of multifocal nodules were detected by B-ultrasound, the difference was statistically significant between the two groups(χ2=7.899, P<0.05). The diagnostic rates of intraoperative frozen-section examination of the study group and the control group were 61.9%(26/42) and 66.1%(271/410), respectively, the difference was statistically significant between the two groups(χ2=4.460, P<0.05). The lymph node metastasis rates of the study group and the control group were 9.5%(4/42) and 26.8%(110/410), respectively, the difference was statistically significant between the two groups(χ2=6.049, P<0.05). The middle follow-up period was 37 months(1 month to 74 months). There was no recurrence of thyroid cancer in the study group, 2 cases with recurrence of hyperthyroidism after operation, and 3 cases with recurrence in the control group, with secondary surgical resection.There were no deaths in the two groups during follow-up. Conclusion Thyroid cancer should be paid attention to when hyperthyroidism combined with nodules, but the prognosis of hyperthyroidism with microcarcinoma is better.The operation method tends to be bilateral lobectomy, which can prevent the recurrence of hyperthyroidism.In order to reduce the chance of parathyroid gland and recurrent laryngeal nerve injury, there may be no need of preventive lymph node dissection in area VI. Key words: Hyperthyroidism; Thyroid neoplasms; Epidemiology; Ultrasonography; Pathology, clinical; Calcium compounds; Neoplasm staging; Lymphatic metastasis; Prognosis