Search

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.

BACKGROUND: The present study aimed to investigate whether deep bone suppression imaging (BSI) could increase the diagnostic performance for solitary pulmonary nodule detection compared with digital tomosynthesis (DTS), dual-energy subtraction (DES) radiography, and conventional chest radiography (CCR). METHODS: A total of 256 patients (123 with a solitary pulmonary nodule, 133 with normal findings) were included in the study. The confidence score of 6 observers determined the presence or absence of pulmonary nodules in each patient. These were first analyzed using a CCR image, then with CCR plus deep BSI, then with CCR plus DES radiography, and finally with DTS images. Receiver-operating characteristic curves were used to evaluate the performance of the 6 observers in the detection of pulmonary nodules. RESULTS: For the 6 observers, the average area under the curve improved significantly from 0.717 with CCR to 0.848 with CCR plus deep BSI (P

ELF5 is known to regulate the specification and differentiation of epithelial cells in the embryonic lung. However, the pathological function of ELF5 in lung cancer remains elusive. In the present study, ELF5 expression was found to be significantly higher in lung adenocarcinoma than that in corresponding adjacent normal tissues. We performed cell and animal experiments to investigate the role of ELF5 in lung adenocarcinoma cells. The results indicated that the overexpression of ELF5 increased the proliferation of lung adenocarcinoma cells, on the contrary, the reduction of ELF5 decreased the proliferation of lung adenocarcinoma cells. Mechanistically, we advance ELF5 promoted lung cancer cell proliferation through inhibiting APC2 and increased Cyclin D1 expression which is a critical downstream target of Wnt pathway. Taken together, these findings supported that ELF5 plays an essential role in the proliferation of lung adenocarcinoma cells and may act as a candidate of therapeutic target to treat lung adenocarcinoma.

Synthetic digital mammography (SDM), a 2D image generated from digital breast tomosynthesis (DBT), is used as a potential substitute for full-field digital mammography (FFDM) in clinic to reduce the radiation dose for breast cancer screening. Previous studies exploited projection geometry and fused projection data and DBT volume, with different post-processing techniques applied on re-projection data which may generate different image appearance compared to FFDM. To alleviate this issue, one possible solution to generate an SDM image is using a learning-based method to model the transformation from the DBT volume to the FFDM image using current DBT/FFDM combo images. In this study, we proposed to use a deep convolutional neural network (DCNN) to learn the transformation to generate SDM using current DBT/FFDM combo images. Gradient guided conditional generative adversarial networks (GGGAN) objective function was designed to preserve subtle MCs and the perceptual loss was exploited to improve the performance of the proposed DCNN on perceptual quality. We used various image quality criteria for evaluation, including preserving masses and MCs which are important in mammogram. Experiment results demonstrated progressive performance improvement of network using different objective functions in terms of those image quality criteria. The methodology we exploited in the SDM generation task to analyze and progressively improve image quality by designing objective functions may be helpful to other image generation tasks.

Objective: accurately classifying the malignancy of lesions detected in a screening scan is critical for reducing false positives. Radiomics holds great potential to differentiate malignant from benign tumors by extracting and analyzing a large number of quantitative image features. Since not all radiomic features contribute to an effective classifying model, selecting an optimal feature subset is critical. Methods: this work proposes a new multi-objective based feature selection (MO-FS) algorithm that considers sensitivity and specificity simultaneously as the objective functions during feature selection. For MO-FS, we developed a modified entropy-based termination criterion that stops the algorithm automatically rather than relying on a preset number of generations. We also designed a solution selection methodology for multi-objective learning that uses the evidential reasoning approach (SMOLER) to automatically select the optimal solution from the Pareto-optimal set. Furthermore, we developed an adaptive mutation operation to generate the mutation probability in MO-FS automatically. Results: we evaluated the MO-FS for classifying lung nodule malignancy in low-dose CT and breast lesion malignancy in digital breast tomosynthesis. Conclusion: the experimental results demonstrated that the feature set selected by MO-FS achieved better classification performance than features selected by other commonly used methods. Significance: the proposed method is general and more effective radiomic feature selection strategy.

As the coronavirus disease 2019 (COVID-19) epidemic spreads around the world, the demand for imaging examinations increases accordingly. The value of conventional chest radiography (CCR) remains unclear. In this study, we aimed to investigate the diagnostic value of CCR in the detection of COVID-19 through a comparative analysis of CCR and CT. This study included 49 patients with 52 CT images and chest radiographs of pathogen-confirmed COVID-19 cases and COVID-19-suspected cases that were found to be negative (non-COVID-19). The performance of CCR in detecting COVID-19 was compared to CT imaging. The major signatures that allowed for differentiation between COVID-19 and non-COVID-19 cases were also evaluated. Approximately 75% (39/52) of images had positive findings on the chest x-ray examinations, while 80.7% (42/52) had positive chest CT scans. The COVID-19 group accounted for 88.4% (23/26) of positive chest X-ray examinations and 96.1% (25/26) of positive chest CT scans. The sensitivity, specificity, and accuracy of CCR for abnormal shadows were 88%, 80%, and 87%, respectively, for all patients. For the COVID-19 group, the accuracy of CCR was 92%. The primary signature on CCR was flocculent shadows in both groups. The shadows were primarily in the bi-pulmonary, which was significantly different from non-COVID-19 patients ( p = 0.008). The major CT finding of COVID-19 patients was ground-glass opacities in both lungs, while in non-COVID-19 patients, consolidations combined with ground-glass opacities were more common in one lung than both lungs ( p = 0.0001). CCR showed excellent performance in detecting abnormal shadows in patients with confirmed COVID-19. However, it has limited value in differentiating COVID-19 patients from non-COVID-19 patients. Through the typical epidemiological history, laboratory examinations, and clinical symptoms, combined with the distributive characteristics of shadows, CCR may be useful to identify patients with possible COVID-19. This will allow for the rapid identification and quarantine of patients.

To build and validate deep learning and machine learning fusion models to classify benign, malignant, and intermediate bone tumors based on patient clinical characteristics and conventional radiographs of the lesion. In this retrospective study, data were collected with pathologically confirmed diagnoses of bone tumors between 2012 and 2019. Deep learning and machine learning fusion models were built to classify tumors as benign, malignant, or intermediate using conventional radiographs of the lesion and potentially relevant clinical data. Five radiologists compared diagnostic performance with and without the model. Diagnostic performance was evaluated using the area under the curve (AUC). A total of 643 patients’ (median age, 21 years; interquartile range, 12–38 years; 244 women) 982 radiographs were included. In the test set, the binary category classification task, the radiological model of classification for benign/not benign, malignant/nonmalignant, and intermediate/not intermediate had AUCs of 0.846, 0.827, and 0.820, respectively; the fusion models had an AUC of 0.898, 0.894, and 0.865, respectively. In the three-category classification task, the radiological model achieved a macro average AUC of 0.813, and the fusion model had a macro average AUC of 0.872. In the observation test, the mean macro average AUC of all radiologists was 0.819. With the three-category classification fusion model support, the macro AUC improved by 0.026. We built, validated, and tested deep learning and machine learning models that classified bone tumors at a level comparable with that of senior radiologists. Model assistance may somewhat help radiologists’ differential diagnoses of bone tumors. • The deep learning model can be used to classify benign, malignant, and intermediate bone tumors. • The machine learning model fusing information from radiographs and clinical characteristics can improve the classification capacity for bone tumors. • The diagnostic performance of the fusion model is comparable with that of senior radiologists and is potentially useful as a complement to radiologists in a bone tumor differential diagnosis.

Mammography is the first option for screening breast cancer. However, some lesions may be missed due to superimposition of breast parenchymal patterns/tissues. Current enhancement methods can highlight some specific tissues, but the edges are weakened or extra noise is added. For improving the solution, this paper proposed a multiscale contrast enhancement for mammogram using dynamic unsharp masking (UM) in Laplacian pyramid. Laplacian pyramid is utilized to preserve the fine structure at each scale. Dynamic UM is presented to adaptively enhance details and suppress noise simultaneously. The proposed method mainly consists of five steps: 1) down-sample images; 2) calculate the weight of dynamic UM; 3) utilize dynamic UM enhancement for each scale image; 4) subtract the enhancement results for each scale image to acquire Laplacian pyramid images; and 5) restore the final enhanced image. To evaluate the proposed method qualitatively and quantitatively, one phantom and three clinical mammography cases were evaluated. Results showed the proposed method can provide much clearer details of the mammary gland. Quantitatively, information entropy and peak signal-to-noise ratio are increased by 0.96 and 3.89 at most compared with the state-of-the-art method. The proposed method has demonstrated that different types of regions are enhanced with the help of a regional adaptive evolution, which has great potential for adaptively enhancing the details and relatively suppressing the noise.

Automatically matching corresponding regions of interest (ROIs) on two-view images is valuable in breast cancer diagnosis, benefiting of saving time and cutting the workload. We propose a method for matching the corresponding ROIs by integrating the geometric model and image similarity searching. The geometric model is implemented by restoring a free breast in the 3D space from two-view preprocessed breast contours. Then, the possible position of the ROI center on cranio-caudal (CC)/medio-lateral oblique (MLO) view image is represented by three feature points in the 3D space. As the view changes, these points can be mapped onto the MLO/CC view image. A matching strip is created later according to the confidence interval, within which the specific position of the ROI can be located by image similarity searching. The experiments were conducted on 273 pairs of mammograms with 400 calcifications and 284 pairs with 300 masses to verify the accuracy of the geometric model and similarity searching. The mean absolute error between the curves and the ROI centers was 3.36 ± 2.90 mm. For 95% detection sensitivity, the confidence interval was ±8.77 mm. For calcifications, the mean distance between the centers of the matched ROIs and the reference was 3.92 ± 4.61 mm. About 93.46% cases had overlap greater than 50%, and 92.46% cases had overlap greater than 75%. For masses, the mean distance was 6.15 ± 7.08 mm. About 88.46% cases had overlap greater than 50%, and 85.58% cases had overlap greater than 75%.

Bone suppression of chest radiographs (CXRs) is potentially useful for diagnosing lung diseases for radiologists and computer–aided diagnosis. This paper presents a cascaded convolutional network model in wavelet domain (Wavelet-CCN) for bone suppression in single conventional CXR. Wavelet coefficients are sparse and suitable as the output of convolutional network. The convolutional networks are trained to predict the wavelet coefficients of bone images from the wavelet coefficients of CXRs, using real two-exposure dual energy subtraction (DES) CXRs as training data. By combining the multi-level wavelet decomposition and a cascaded refinement framework, the Wavelet-CCN model can work automatically with a multi-scale approach and progressively refine the prediction in terms of accuracy and spatial resolution. Compared with previous work of CamsNet model which preforms bone prediction in gradient domain, the Wavelet-CCN model predicts the wavelet coefficients to reconstruct bone images and can avoid the inconsistent background intensity caused by 2D integration of gradients. The predicted bone image is subtracted from the original CXR to produce a soft-tissue image. The Wavelet-CCN model and its variants with different wavelet basis are evaluated on a dataset that consists of 504 cases of real two-exposure DES CXRs (404 cases for training and 100 cases for test). Experimental results show that among all the variants and different wavelet bases, the Wavelet-CCN model with Haar wavelet performs best. The average peak signal-to-noise ratio and structural similarity index of the soft-tissue images produced by the proposed Wavelet-CCN model are both higher than those of the previous CamsNet model in gradient domain, reaching values of 39.4 (±0.94) dB and 0.977 (±0.004), respectively. The results also demonstrate that the Wavelet-CCN model can process the CXRs acquired by four types of X-ray machines.

BACKGROUND: This study evaluated the clinical characteristics and imaging findings of 112 patients with irregular and flat bone osteosarcoma (IFBO). METHODS: The age, gender, location, tumor size, density and signal intensity, osteoid matrix, periosteal reaction, and histological subtypes were analyzed for 112 patients with IFBO. RESULTS: A total of 112 patients with IFBO, including 64 males and 48 females, with a mean age of 34.8 years were enrolled in this study. Over half of the tumors (54.5%) were detected in the craniofacial region and the skull (24 in the maxilla bone, 17 in the mandible bone, 11 in the sphenoid bone, 7 in the temporal bone, 1 in the frontal bone, and 1 in the occipital bone). Other tumor locations included the pelvic region (20.5%; 20 in the ilium and 3 in the pubis), the chest (18.8%; 11 in the scapula, 7 in the ribs, and 3 in the clavicle), and the vertebrae (6.3%; 3 in the thoracic spine, 2 in the lumbar spine, 1 in the sacrum, and 1 in the cervical spine). Transarticular extension occurred in 11 of the 23 pelvic cases (47.8%), primarily involving the sacroiliac joint (90.9%; 10 of 11). Six cases (6/7; 85.7%) of vertebral osteosarcoma arose from the transverse process and the pedicle, and 1 (1/7; 14.3%) arose from the sacral tuberosity and the ala, with partial vertebral body involvement. Additionally, 27 patients (24.1%) presented with secondary osteosarcoma related to prior radiotherapy, and 2 (1.8%) were associated with osteoblastoma and fibrous dysplasia. Histological examination revealed high-grade tumors in 88 (78.6%) cases. The tumors presented as soft-tissue masses with a diameter of 7.5±3.2 cm. A total of 91 patients underwent X-ray examination and/or computed tomography (CT) examinations. The osteoid matrix was detected in 84 patients (84/91;92.3%). A periosteal reaction was detected in 56 cases (56/91; 61.5%), including a lamellar periosteal reaction in 10 patients (11.0%) and a spiculated periosteal reaction in 46 cases (50.5%). All 74 cases who underwent magnetic resonance imaging (MRI) examinations presented with heterogeneous masses in the surrounding soft tissue. Enhancement was homogenous in 12 cases (18.5%) and heterogeneous in 53 cases (81.5%). Peripheral rim enhancement was observed in 10 cases (13.5%). CONCLUSIONS: IFBO should be considered when diagnosing patients over 30 years of age who exhibit osteoid matrix in bone lesions. Maxillofacial osteosarcoma is commonly associated with a history of radiation exposure. Pelvic osteosarcoma is more likely to invade the sacroiliac joint. Vertebral osteosarcoma frequently arises in the transverse process and pedicle, with partial body involvement.

Background: Computer-aided diagnosis (CAD) system can provide reference to radiologists in breast mass classification. This study was to verify if a CAD model, based on perceptive features learned from quantitative BI-RADS descriptions, can help radiologists improve diagnosis performance for breast masses in mammography.Methods: A retrospective multi-reader multi-case (MRMC) study was conducted to evaluate a CAD model established on perceptive features. Digital mammograms of 416 patients with breast masses were collected from 2014 to 2017, including 231 benign and 185 malignant masses. Altogether, 214 of them (109 benign, 105 malignant) were selected randomly to train the CAD model which consisted of perceptive feature extractor and classifier. The other 202 patients were used as the test set for evaluation from which 51 patients (29 benign and 22 malignant) were selected. Six radiologists were divided into three groups (junior, middle-senior, and senior).They evaluated 51 patients without and with support from the CAD model. BI-RADS category, benign or malignant diagnosis, probability of malignancy, and diagnosis time were recorded during two evaluation sessions. Results: In the MRMC evaluation, the average AUC of six radiologists with CAD support was significantly higher than that without support (0.896 vs. 0.850, p=0.02). Both of average sensitivity and average specificity increased (p = 0.0253). More cases were assessed as BI-RADS 4 than BI-RADS 2 or 3. Five radiologists showed comparable diagnosis time per case with and without CAD support, and one radiologist showed a significant decrease when the CAD model was involved. Conclusion: The CAD model could improve radiologists’ diagnostic performance for breast masses without improving the diagnosis time.

Objectives. To build and validate random forest (RF) models for the classification of bone tumors based on the conventional radiographic features of the lesion and patients’ clinical characteristics, and identify the most essential features for the classification of bone tumors. Materials and Methods. In this retrospective study, 796 patients (benign bone tumors: 412 cases, malignant bone tumors: 215 cases, intermediate bone tumors: 169 cases) with pathologically confirmed bone tumors from Nanfang Hospital of Southern Medical University, Foshan Hospital of TCM, and University of Hong Kong-Shenzhen Hospital were enrolled. RF models were built to classify tumors as benign, malignant, or intermediate based on conventional radiographic features and potentially relevant clinical characteristics extracted by three musculoskeletal radiologists with ten years of experience. SHapley Additive exPlanations (SHAP) was used to identify the most essential features for the classification of bone tumors. The diagnostic performance of the RF models was quantified using receiver operating characteristic (ROC) curves. Results. The features extracted by the three radiologists had a satisfactory agreement and the minimum intraclass correlation coefficient (ICC) was 0.761 (CI: 0.686-0.824, P < .001 ). The binary and tertiary models were built to classify tumors as benign, malignant, or intermediate based on the imaging and clinical features from 627 and 796 patients. The AUC of the binary (19 variables) and tertiary (22 variables) models were 0.97 and 0.94, respectively. The accuracy of binary and tertiary models were 94.71% and 82.77%, respectively. In descending order, the most important features influencing classification in the binary model were margin, cortex involvement, and the pattern of bone destruction, and the most important features in the tertiary model were margin, high-density components, and cortex involvement. Conclusions. This study developed interpretable models to classify bone tumors with great performance. These should allow radiographers to identify imaging features that are important for the classification of bone tumors in the clinical setting.

Early tuberculosis (TB) screening through chest X-rays (CXRs) is essential for timely detection and treatment of the disease. Previous studies have shown that texture abnormalities in CXRs can be enhanced by bone suppression techniques, which may potentially improve TB diagnosis performance. However, existing TB datasets with CXRs usually lack bone-suppressed images, making it difficult to take advantage of bone suppression for TB recognition. Also, existing bone suppression models are usually trained on a relatively small dual-energy subtraction (DES) dataset with CXRs, without considering the image specificity of TB patients. To this end, we propose a bone-suppressed CXR-based tuberculosis recognition (BCTR) framework, where diagnosis-specific deep features are extracted and used to guide a bone suppression (BS) model to generate bone-suppressed CXRs for TB diagnosis. Specifically, the BCTR consists of a classification model for TB diagnosis and an image synthesis model for bone suppression of CXRs. The classification model is first trained on original CXRs from multiple TB datasets such as the large-scale TBX11K. The image synthesis model is trained on a DES dataset to produce bone-suppressed CXRs. Considering the heterogeneity of CXR images from the TB datasets and the DES dataset, we proposed to extract multi-scale task-specific features from the trained classification model and transfer them (via channel-wise addition) to the corresponding layers in the image synthesis model to explicitly guide the bone suppression process. With the bone-suppressed CXRs as input, the classification model is further trained for multi-class TB diagnosis. Experimental results on five TB databases and a DES dataset suggest that our BCTR outperforms previous state-of-the-arts in automated tuberculosis diagnosis.

BACKGROUND: Contrast-enhanced mammography (CEM) is an imaging tool for breast cancer detection. Most quantitative analyses of CEM involve two phases, and it is unknown whether an added delayed phase can improve its diagnostic performance compared to dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). This study aimed to evaluate whether the delayed phase improves the diagnostic performance of CEM in distinguishing malignant and benign masses. METHODS: This prospective study enrolled 111 women with 111 pathologically confirmed breast masses. CEM was performed after the injection of contrast agent between 2–3 minutes (T1, early phase), 4–5 minutes (T2, second phase), and 7–9 minutes (T3, delayed phase). The quantitative enhanced gray value of lesions (LGV) and the lesion to background grey value ratio (LBR) were measured within each phase’s corresponding region of interest (ROI). Based on their changes, the kinetic enhancement pattern was assessed among the three phases, and the diagnostic performance was subsequently measured. RESULTS: The LGV and LBR of malignant masses were significantly greater than those of benign lesions. The diagnostic performance of LGV and LBR at the delayed phase was consistent with that of the second phase but poorer than that of the early phase. The sensitivity of LGV(T1) + LGV(T2) + LGV(T3) was less than that of LGV(T1) + LGV(T2) (86.5% vs. 95.1%) with a similar area under the curve (AUC), specificity, positive-predictive value (PPV), negative-predictive value (NPV), and accuracy. The sensitivity of LBR(T1) + LBR(T2) + LBR(T3) increased by 19.6%, and specificity decreased by 20.7% compared with LBR(T1) + LBR(T2). The LGV(T1) + LGV(T2) + LGV(T3) + kinetic enhancement (T1–T3) had the lowest sensitivity (67.0%), but the highest specificity (75.8%), and the sensitivity of LBR(T1) + LBR(T2) + LBR(T3) + kinetic enhancement (T1–T3) was higher than that of LBR(T1) + LBR(T2) + kinetic enhancement (T1–T2) (90.2% vs. 63.4%, respectively). CONCLUSIONS: The addition of a delayed CEM phase for breast cancer diagnosis yielded limited performance improvement. The quantitative analysis combined with enhancement patterns between the two consecutive phases has great potential to distinguish between malignant and benign lesions.

This study established an interpretable machine learning model to predict the severity of coronavirus disease 2019 (COVID-19) and output the most crucial deterioration factors. Clinical information, laboratory tests, and chest computed tomography (CT) scans at admission were collected. Two experienced radiologists reviewed the scans for the patterns, distribution, and CT scores of lung abnormalities. Six machine learning models were established to predict the severity of COVID-19. After parameter tuning and performance comparison, the optimal model was explained using Shapley Additive explanations to output the crucial factors. This study enrolled and classified 198 patients into mild ( n = 162 ; 46.93 ± 14.49 years old) and severe ( n = 36 ; 60.97 ± 15.91 years old) groups. The severe group had a higher temperature ( 37.42 ± 0.99 °C vs. 36.75 ± 0.66 °C), CT score at admission, neutrophil count, and neutrophil-to-lymphocyte ratio than the mild group. The XGBoost model ranked first among all models, with an AUC, sensitivity, and specificity of 0.924, 90.91%, and 97.96%, respectively. The early stage of chest CT, total CT score of the percentage of lung involvement, and age were the top three contributors to the prediction of the deterioration of XGBoost. A higher total score on chest CT had a more significant impact on the prediction. In conclusion, the XGBoost model to predict the severity of COVID-19 achieved excellent performance and output the essential factors in the deterioration process, which may help with early clinical intervention, improve prognosis, and reduce mortality.

To investigate the effect of different breast lesions on exposure parameters in digital mammography and to determine whether the exposure parameters can additively improve diagnostic efficiency. Craniocaudal view and mediolateral view full-field digital mammography images from 982 women with unilateral lesions (341 with malignant lesions, 189 with benign lesions, and 452 healthy women) obtained at Nanfang Hospital were reviewed. Differences in exposure parameters (tube voltage and load, breast thickness (BT), and average glandular dose (AGD)) between breasts were calculated. The relationships between parameter differences and lesion size were explored. A logistic regression model was used based on the AGD and BT differences, and the area under the receiver operating characteristic curve (AUC) was used to assess the performance of these parameters in differentiating malignant from benign and healthy subjects. Independently, data from 129 women (82 with malignant and 47 with benign lesions) treated at Sun Yat-sen Memorial Hospital were collected to validate the model. Differences in tube voltage and load, BT, and AGD between breasts were significantly greater in the malignant subjects than benign (p < 0.05) and healthy subjects (p < 0.05). The AUCs for the comparisons of malignant vs. healthy subjects, malignant vs. benign subjects, and benign vs. healthy subjects were 0.77 ± 0.02, 0.72 ± 0.02, and 0.57 ± 0.02, respectively. The model combining the exposure parameters with the BI-RADS category resulted in a higher AUC (0.910 ± 0.03) compared with physician diagnosis alone (0.820 ± 0.04) for differentiating between malignant and benign lesions. Exposure parameters additively improved diagnostic accuracy for breast cancer and yielded more reliable results. • Differences in kVp, mAs, BT, and AGD between breasts were significantly greater in the malignant subjects than benign and healthy subjects. • The model combining exposure parameters with the BI-RADS category resulted in a higher AUC compared with the physician’s diagnosis for differentiating between malignant and benign lesions. • Exposure parameters additively improved diagnostic accuracy for breast cancer.

Objectives. To compare the differences in normalized average glandular dose (NAGD) between the breasts of healthy subjects and those of cancer patients and to determine if the NAGD difference is associated with breast cancer risk and improves breast cancer classification. Materials and Methods. Craniocaudal view and mediolateral view full-field digital mammography (FFDM) images were obtained from 1682 healthy subjects whose breasts were categorized as Breast Imaging-Reporting and Data System (BI-RADS) I or II and from 811 biopsy-confirmed unilateral breast cancer patients whose breasts on the contralateral side were category I or II. Both populations were randomized into training and test sets. Multivariate logistic regression analysis was used to build the breast cancer risk assessment model, and the area under the receiver operating characteristic curve ( A z ) was used to evaluate the model. Twenty-two breast cancer patients who were originally categorized as BI-RADS I or II for both breasts, but were diagnosed with unilateral biopsy-confirmed breast cancer subsequently, were included to validate the model. Results. The NAGD differences in both FFDM images between tumor-bearing breasts and the healthy breasts of patients were significantly higher than those in healthy subjects ( P < 0.001 ). The model with NAGD differences had a higher A z value than the model without NAGD differences. While there was no NAGD differences between originally healthy breasts of breast cancer patients, significant NAGD differences between now tumor-bearing breasts and the then previously healthy breasts were found in both FFDM images. Conclusions. NAGD differences between both breasts can be included in the breast cancer risk assessment model to evaluate breast cancer risk.

Generating dual-energy subtraction (DES) soft-tissue images from chest radiographs (also called bone suppression) is an important task, as it improves the detection rates for lung nodules. Previous studies focus on generating DES-like soft-tissue images from CXRs through machine/deep learning techniques. However, they usually require tedious image processing steps for bone segmentation/delineation or ignore anatomical structure information (e.g., edges of ribs and clavicles) in CXRs. In this work, we propose a bone Edge-guided Generative Adversarial Network (EGAN) to generate DES-like soft-tissue images from conventional CXRs, which does not require human intervention and can explicitly use anatomical structure information of bones in CXRs. Specifically, the edges of ribs and clavicles in an input CXR were first detected by a trained fully convolutional network. Then, the edge probability map, as well as the original CXR image, are fed into a GAN model to generate the DES-like soft-tissue image, where the detected edge information is used as the prior knowledge to directly and specifically guide the image generation process. Experimental results on 504 subjects (each equipped with a CXR, a DES bone image, and a DES soft-tissue image) demonstrate that EGAN can produce DES-like soft-tissue images with high-quality and high-resolution, compared with classic deep learning methods. We further apply the trained EGAN to CXRs acquired by different types of X-ray machines in the public JSRT and NIH ChestXray 14 datasets, and our method can also produce visually appealing DES-like soft-tissue images.

Objective: This study aims to assess the CT and MRI features of calvarium and skull base osteosarcoma (CSBO). Methods: The CT and MRI features and pathological characteristics of 12 cases of pathologically confirmed CSBO were analyzed retrospectively. Results: 12 patients (age range 9–67 years; 3 male, 9 female) were included in the study. Tumours occurred in skull base (7, 58.3%), temporal (4, 33.3%) and frontal (1, 8.3%). Among all, six patients received radiotherapy for nasopharyngeal carcinoma. According to pathology, 11 out of 12 tumours were high-grade (91.7%). On CT, all the tumours had soft tissue mass penetrated into cortical bone with invasion of surrounding soft tissue. Six tumours were shown to have lytic density and six were mixed density. Matrix mineralization was present in 10 cases (83.3%). On MRI, tumours presented as soft-tissue masses measuring 5.9 ± 2.4 (3.9–8.0) cm. Five tumours showed low signal intensities on T1 weighted imaging with seven having heterogeneous signal intensities. One showed low signal intensity on T2 weighted imaging, two showed high signal intensities and nine heterogeneous signal intensities. All the tumours showed low signal intensities on diffusion-weighted imaging. On contrast enhanced images, seven cases showed heterogeneous enhancement, three showed peripheral enhancementand and two showed homogeneous enhancement. Dural tail sign were detected in nine cases. Conclusion: CSBO is rare, and is commonly associated with previous radiation exposure. A presumptive diagnosis for osteosarcoma should be considered when calvarium and skull base tumours with osteoid matrix and duraltail sign are found. Advances in knowledge: CT and MR features of CSBO have not been reported. The study helps to identify CSBO and other sarcomas.

Digital breast tomosynthesis (DBT) with improved lesion conspicuity and characterization has been adopted in screening practice. DBT-based diagnosis strongly depends on physicians' experience, so an automatic lesion malignancy classification model using DBT could improve the consistency of diagnosis among different physicians. Tensor-based approaches that use the original imaging data as input have shown promising results for many classification tasks. However, DBT data are pseudo-3D volumetric imaging as the slice spacing of DBT is much coarser than that of the in-plane resolution. Thus, directly constructing DBT as the third-order tensor in a conventional tensor-based classifier with introducing additional information to the original DBT data along the slice-spacing dimension will lead to inconsistency across all three dimensions. To avoid such inconsistency, we introduce a collection input based support tensor machine (CISTM)-based classifier that uses the tensor collection as input for classifying lesion malignancy in DBT. In CISTM, instead of introducing the third dimension directly into the geometry construction, the third-dimension structural relationship is related by weight parameters in the decision function, which is dynamically and automatically constructed during the classifier training process and is more consistent with the pseudo-3D nature of DBT. We tested our method on a DBT dataset of 926 images among which 262 were malignant and 664 were benign. We compared our method with the latest tensor-based method, KSTM (kernelled support tensor machine), which does not consider the unique non-uniform resolution property of DBT. Experimental results illustrate that the CISTM-based classifier is effective for classifying breast lesion malignancy in DBT and that it outperforms the KSTM-based classifier.

Digital breast tomosynthesis (DBT) mammography is a promising imaging technique for detecting early-stage breast cancers. However, DBT imaging usually contains truncation artifacts around the chest wall area, where many recurring breast tumors or masses are located. Extrapolation or interpolation and weighting techniques are used to suppress artifacts in existing methods, but these are inappropriate for DBT due to its high spatial resolution requirement. To solve this problem, we propose a practical truncation correction method, which can realize simultaneous truncation artifact reduction and image reconstruction (STARIR) for DBT. In contrast to extrapolation or interpolation estimations and weighting strategy, this integrated method renews each truncated voxel by considering the information of all measured projections rather than that of one projection at each reconstruction iteration. Qualitative and quantitative studies were performed on simulated and realistic DBT data to validate the proposed method. Results showed that STARIR reduces truncation artifacts and preserves tissue details in reconstructed images more effectively than do the current methods. The proposed method can be used for simultaneous and accurate truncation artifact reduction and image reconstruction in DBT.

Multifaceted features decoded from mammographic images may describe various perspectives of the breast mass heterogeneity, in this study, we aimed to explore a methodology to effectively integrate multifaceted mass representations extracted from the digital breast tomosynthesis (DBT) and full-field digital mammography (FFDM) to enhance breast cancer discrimination. A novel multi-criterion decision making-based multi-channel fusion (MDMF) framework was proposed to fuse different breast mass representations processed in multi-channels built on the deep convolutional neural network (DCNN) and the multilayer perceptron (MLP) at the decision level. A hierarchical framework (HFMM) was also developed for multi-modality images and multi-channel fusion to integrate multimodality information from DBT and FFDM. We retrospectively collected 441 patients with both DBT and FFDM, and the regions of interest (ROIs) covering the malignant, benign, and normal tissues were extracted for validation. The MDMF achieved the area under the receiver operating characteristic curve (AUC) of 93.14%, 91.30%, 97.35% (FFDM) and 93.79%, 95.16%, 99.31% (DBT) respectively for the malignant, benign and normal mass. While the HFMM further boosted the performance to AUC of malignant 94.14%, benign 95.42% and normal mass 99.56% The matthews correlation coefficient (MCC) were 73.15% and 81.02% for FFDM and DBT accomplished by MDMF, and enhanced to 81.72% when integrating the multimodality information from DBT and FFDM via the proposed HFMM. The experimental results suggested that the proposed HFMM achieved superior discriminative performance when compared with the benchmark classification algorithms and fusion architectures, rendering it a practical tool for breast mass discrimination in breast cancer screening.

Suppression of bony structures in chest radiographs (CXRs) is potentially useful for radiologists and computer-aided diagnostic schemes. In this paper, we present an effective deep learning method for bone suppression in single conventional CXR using deep convolutional neural networks (ConvNets) as basic prediction units. The deep ConvNets were adapted to learn the mapping between the gradients of the CXRs and the corresponding bone images. We propose a cascade architecture of ConvNets (called CamsNet) to refine progressively the predicted bone gradients in which the ConvNets work at successively increased resolutions. The predicted bone gradients at different scales from the CamsNet are fused in a maximum-a-posteriori framework to produce the final estimation of a bone image. This estimation of a bone image is subtracted from the original CXR to produce a soft-tissue image in which the bone components are eliminated. Our method was evaluated on a dataset that consisted of 504 cases of real two-exposure dual-energy subtraction chest radiographs (404 cases for training and 100 cases for test). The results demonstrate that our method can produce high-quality and high-resolution bone and soft-tissue images. The average relative mean absolute error of the produced bone images and peak signal-to-noise ratio of the produced soft-tissue images were 3.83% and 38.7dB, respectively. The average bone suppression ratio of our method was 83.8% for the CXRs with pixel sizes of nearly 0.194mm. Furthermore, we apply the trained CamsNet model on the CXRs acquired by various types of X-ray machines, including scanned films, and our method can also produce visually appealing bone and soft-tissue images.

To evaluate the impact of utilizing digital breast tomosynthesis (DBT) or/and full-field digital mammography (FFDM), and different transfer learning strategies on deep convolutional neural network (DCNN)-based mass classification for breast cancer. We retrospectively collected 441 patients with both DBT and FFDM on which regions of interest (ROIs) covering the malignant, benign and normal tissues were extracted for DCNN training and validation. Experiments were conducted for tasks in distinguishing malignant/benign/normal: (1) classification capabilities of DBT vs FFDM and the role of transfer learning were validated on 2D-DCNN; (2) different strategies of combining DBT and FFDM and the associated impacts on classification were explored; (3) 2D-DCNN and 3D-DCNN trained from scratch with volumetric DBT were compared. 2D-DCNN with transfer learning outperformed that without for DBT in distinguishing malignant (ΔAUC = 0.059 ± 0.009, p

Predicting lesion malignancy accurately and reliably in digital breast tomosynthesis is critically important for breast cancer screening. Tumor shape and interactive effect between the tumor and surrounding normal tissue are two of the most important indicators in radiologists’ reading. On the other hand, the density and texture of region within the tumor also play an important role in malignancy classification. Inspired by the above observations, shell and kernel descriptors were proposed in this work for breast lesion malignancy prediction, in which the shell descriptor is used for describing the tumor shape and surrounding normal tissue while the kernel descriptor is used to describe the internal tumor region. A joint deep learning model based on the AlexNet was designed to learn and fuse features from shell and kernel. Additionally, to obtain more reliable predictive results, a multi-objective optimization algorithm and a reliable classifier fusion strategy were used to train the predictive model and optimally combine outputs from both shell and kernel descriptors. In this study, 278 malignant and 685 benign cases were used through 2-fold cross validation. Compared with the single descriptor based models using either shell or kernel, the experimental results demonstrated that the combined shell and kernel descriptors can capture the most important features and the corresponding predictive model achieved the best performance as well.