Time-Series MR Images Identifying Complete Response to Neoadjuvant Chemotherapy in Breast Cancer Using a Deep Learning Approach.

BACKGROUND: Pathological complete response (pCR) is an essential criterion for adjusting follow-up treatment plans for patients with breast cancer (BC). The value of the visual geometry group and long short-term memory (VGG-LSTM) network using time-series dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) for pCR identification in BC is unclear. PURPOSE: To identify pCR to neoadjuvant chemotherapy (NAC) using deep learning (DL) models based on the VGG-LSTM network. STUDY TYPE: Retrospective. POPULATION: Center A: 235 patients (47.7 ± 10.0 years) were divided 7:3 into training (n = 164) and validation set (n = 71). Center B: 150 patients (48.5 ± 10.4 years) were used as test set. FIELD STRENGTH/SEQUENCE: 3-T, T2-weighted spin-echo sequence imaging, and gradient echo DCE sequence imaging. ASSESSMENT: Patients underwent MRI examinations at three sequential time points: pretreatment, after three cycles of treatment, and prior to surgery, with tumor regions of interest manually delineated. Histopathology was the gold standard. We used VGG-LSTM network to establish seven DL models using time-series DCE-MR images: pre-NAC images (t0 model), early NAC images (t1 model), post-NAC images (t2 model), pre-NAC and early NAC images (t0 + t1 model), pre-NAC and post-NAC images (t0 + t2 model), pre-NAC, early NAC and post-NAC images (t0 + t1 + t2 model), and the optimal model combined with the clinical features and imaging features (combined model). The models were trained and optimized on the training and validation set, and tested on the test set. STATISTICAL TESTS: The DeLong, Student's t-test, Mann-Whitney U, Chi-squared, Fisher's exact, Hosmer-Lemeshow tests, decision curve analysis, and receiver operating characteristics analysis were performed. P < 0.05 was considered significant. RESULTS: Compared with the other six models, the combined model achieved the best performance in the test set yielding an AUC of 0.927. DATA CONCLUSION: The combined model that used time-series DCE-MR images, clinical features and imaging features shows promise for identifying pCR in BC. TECHNICAL EFFICACY: Stage 4.

An ultrasound-based ensemble machine learning model for the preoperative classification of pleomorphic adenoma and Warthin tumor in the parotid gland.

OBJECTIVES: The preoperative classification of pleomorphic adenomas (PMA) and Warthin tumors (WT) in the parotid gland plays an essential role in determining therapeutic strategies. This study aims to develop and validate an ultrasound-based ensemble machine learning (USEML) model, employing nonradiative and noninvasive features to differentiate PMA from WT. METHODS: A total of 203 patients with histologically confirmed PMA or WT who underwent parotidectomy from two centers were enrolled. Clinical factors, ultrasound (US) features, and radiomic features were extracted to develop three types of machine learning model: clinical models, US models, and USEML models. The diagnostic performance of the USEML model, as well as that of physicians based on experience, was evaluated and validated using receiver operating characteristic (ROC) curves in internal and external validation cohorts. DeLong's test was used for comparisons of AUCs. SHAP values were also utilized to explain the classification model. RESULTS: The USEML model achieved the highest AUC of 0.891 (95% CI, 0.774-0.961), surpassing the AUCs of both the US (0.847; 95% CI, 0.720-0.932) and clinical (0.814; 95% CI, 0.682-0.908) models. The USEML model also outperformed physicians in both internal and external validation datasets (both p < 0.05). The sensitivity, specificity, negative predictive value, and positive predictive value of the USEML model and physician experience were 89.3%/75.0%, 87.5%/54.2%, 87.5%/65.6%, and 89.3%/65.0%, respectively. CONCLUSIONS: The USEML model, incorporating clinical factors, ultrasound factors, and radiomic features, demonstrated efficient performance in distinguishing PMA from WT in the parotid gland. CLINICAL RELEVANCE STATEMENT: This study developed a machine learning model for preoperative diagnosis of pleomorphic adenoma and Warthin tumor in the parotid gland based on clinical, ultrasound, and radiomic features. Furthermore, it outperformed physicians in an external validation dataset, indicating its potential for clinical application. KEY POINTS: • Differentiating pleomorphic adenoma (PMA) and Warthin tumor (WT) affects management decisions and is currently done by invasive biopsy. • Integration of US-radiomic, clinical, and ultrasound findings in a machine learning model results in improved diagnostic accuracy. • The ultrasound-based ensemble machine learning (USEML) model consistently outperforms physicians, suggesting its potential applicability in clinical settings.

AMS-U-Net: automatic mass segmentation in digital breast tomosynthesis via U-Net.

Purpose: The objective of this study was to develop a fully automatic mass segmentation method called AMS-U-Net for digital breast tomosynthesis (DBT), a popular breast cancer screening imaging modality. The aim was to address the challenges posed by the increasing number of slices in DBT, which leads to higher mass contouring workload and decreased treatment efficiency. Approach: The study used 50 slices from different DBT volumes for evaluation. The AMS-U-Net approach consisted of four stages: image pre-processing, AMS-U-Net training, image segmentation, and post-processing. The model performance was evaluated by calculating the true positive ratio (TPR), false positive ratio (FPR), F-score, intersection over union (IoU), and 95% Hausdorff distance (pixels) as they are appropriate for datasets with class imbalance. Results: The model achieved 0.911, 0.003, 0.911, 0.900, 5.82 for TPR, FPR, F-score, IoU, and 95% Hausdorff distance, respectively. Conclusions: The AMS-U-Net model demonstrated impressive visual and quantitative results, achieving high accuracy in mass segmentation without the need for human interaction. This capability has the potential to significantly increase clinical efficiency and workflow in DBT for breast cancer screening.

Using time-series chest radiographs and laboratory data by machine learning for identifying pulmonary infection and colonization of Acinetobacter baumannii.

BACKGROUND: Accurately distinguishing between pulmonary infection and colonization in patients with Acinetobacter baumannii is of utmost importance to optimize treatment and prevent antibiotic abuse or inadequate therapy. An efficient automated sorting tool could prompt individualized interventions and enhance overall patient outcomes. This study aims to develop a robust machine learning classification model using a combination of time-series chest radiographs and laboratory data to accurately classify pulmonary status caused by Acinetobacter baumannii. METHODS: We proposed nested logistic regression models based on different time-series data to automatically classify the pulmonary status of patients with Acinetobacter baumannii. Advanced features were extracted from the time-series data of hospitalized patients, encompassing dynamic pneumonia indicators observed on chest radiographs and laboratory indicator values recorded at three specific time points. RESULTS: Data of 152 patients with Acinetobacter baumannii cultured from sputum or alveolar lavage fluid were retrospectively analyzed. Our model with multiple time-series data demonstrated a higher performance of AUC (0.850, with a 95% confidence interval of [0.638-0.873]), an accuracy of 0.761, a sensitivity of 0.833. The model, which only incorporated a single time point feature, achieved an AUC of 0.741. The influential model variables included difference in the chest radiograph pneumonia score. CONCLUSION: Dynamic assessment of time-series chest radiographs and laboratory data using machine learning allowed for accurate classification of colonization and infection with Acinetobacter baumannii. This demonstrates the potential to help clinicians provide individualized treatment through early detection.

Clinical and DCE-CT signs in predicting microvascular invasion in cHCC-ICC.

BACKGROUND: To predict the microvascular invasion (MVI) in patients with cHCC-ICC. METHODS: A retrospective analysis was conducted on 119 patients who underwent CT enhancement scanning (from September 2006 to August 2022). They were divided into MVI-positive and MVI-negative groups. RESULTS: The proportion of patients with CEA elevation was higher in the MVI-positive group than in the MVI-negative group, with a statistically significant difference (P = 0.02). The MVI-positive group had a higher rate of peritumoral enhancement in the arterial phase (P = 0.01) whereas the MVI-negative group had more oval and lobulated masses (P = 0.04). According to the multivariate analysis, the increase in CEA (OR = 10.15, 95% CI: 1.11, 92.48, p = 0.04), hepatic capsular withdrawal (OR = 4.55, 95% CI: 1.44, 14.34, p = 0.01) and peritumoral enhancement (OR = 6.34, 95% CI: 2.18, 18.40, p < 0.01) are independent risk factors for predicting MVI. When these three imaging signs are combined, the specificity of MVI prediction was 70.59% (series connection), and the sensitivity was 100% (parallel connection). CONCLUSIONS: Our multivariate analysis found that CEA elevation, liver capsule depression, and arterial phase peritumoral enhancement were independent risk factors for predicting MVI in cHCC-ICC.

An integrated network based on 2D/3D feature correlations for benign-malignant tumor classification and uncertainty estimation in digital breast tomosynthesis.

Objective.Classification of benign and malignant tumors is important for the early diagnosis of breast cancer. Over the last decade, digital breast tomosynthesis (DBT) has gradually become an effective imaging modality for breast cancer diagnosis due to its ability to generate three-dimensional (3D) visualizations. However, computer-aided diagnosis (CAD) systems based on 3D images require high computational costs and time. Furthermore, there is considerable redundant information in 3D images. Most CAD systems are designed based on 2D images, which may lose the spatial depth information of tumors. In this study, we propose a 2D/3D integrated network for the diagnosis of benign and malignant breast tumors.Approach.We introduce a correlation strategy to describe feature correlations between slices in 3D volumes, corresponding to the tissue relationship and spatial depth features of tumors. The correlation strategy can be used to extract spatial features with little computational cost. In the prediction stage, 3D spatial correlation features and 2D features are both used for classification.Main results.Experimental results demonstrate that our proposed framework achieves higher accuracy and reliability than pure 2D or 3D models. Our framework has a high area under the curve of 0.88 and accuracy of 0.82. The parameter size of the feature extractor in our framework is only 35% of that of the 3D models. In reliability evaluations, our proposed model is more reliable than pure 2D or 3D models because of its effective and nonredundant features.Significance.This study successfully combines 3D spatial correlation features and 2D features for the diagnosis of benign and malignant breast tumors in DBT. In addition to high accuracy and low computational cost, our model is more reliable and can output uncertainty value. From this point of view, the proposed method has the potential to be applied in clinic.

E74­like ETS transcription factor 5 facilitates cell proliferation through regulating the expression of adenomatous polyposis coli 2 in non­small cell lung cancer.

E74­like ETS transcription factor 5 (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 has yet to be fully elucidated. In the present study, the expression of ELF5 was found to be significantly higher in lung adenocarcinoma compared with that in corresponding adjacent normal tissues. Subsequently, cell and animal experiments were performed 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, whereas, by contrast, a reduction in the expression of ELF5 led to a decrease in their proliferation. Mechanistically, the hypothesis is advanced that ELF5 can promote lung cancer cell proliferation through inhibiting adenomatous polyposis coli 2 and increasing the expression of cyclin D1, which is a critical downstream target of the Wnt pathway. Taken together, these findings support the notion that ELF5 exerts an essential role in the proliferation of lung adenocarcinoma cells and may be a therapeutic target for the treatment of lung adenocarcinoma.

Deep learning based de-overlapping correction of projections from a flat-panel micro array X-ray source: Simulation study.

PURPOSE: Flat-panel X-ray source is an experimental X-ray emitter with target application of static computer tomography (CT), which can save imaging space and time. However, the X-ray cone beams emitted by the densely arranged micro-ray sources are overlapped, causing serious structural overlapping and visual blur in the projection results. Traditional deoverlapping methods can hardly solve this problem well. METHOD: We converted the overlapping cone beam projections to parallel beam projections through a U-like neural network and selected structural similarity (SSIM) loss as the loss function. In this study, we converted three kinds of overlapping cone beam projections of the Shepp-Logan, line-pairs, and abdominal data with two overlapping levels to corresponding parallel beam projections. Training completed, we tested the model using the test set data that was not used at the training phase, and evaluated the difference between the test set conversion results and their corresponding parallel beams through three indicators: mean squared error (MSE), peak signal-to-noise ratio (PSNR) and SSIM. In addition, projections from head phantoms were applied for generalization test. RESULT: In the Shepp-Logan low-overlapping task, we obtained a MSE of 1.624×10-5, a PSNR of 47.892 dB, and a SSIM of 0.998 which are the best results of the six experiments. For the most challenging abdominal task, the MSE, PSNR, and SSIM are 1.563×10-3, 28.0586 dB, and 0.983, respectively. In more generalized data, the model also achieved good results. CONCLUSION: This study proves the feasibility of utilizing the end-to-end U-net for deblurring and deoverlapping in the flat-panel X-ray source domain.

Bone suppression of lateral chest x-rays with imperfect and limited dual-energy subtraction images.

Bone suppression is to suppress the superimposed bone components over the soft tissues within the lung area of Chest X-ray (CXR), which is potentially useful for the subsequent lung disease diagnosis for radiologists, as well as computer-aided systems. Despite bone suppression methods for frontal CXRs being well studied, it remains challenging for lateral CXRs due to the limited and imperfect DES dataset containing paired lateral CXR and soft-tissue/bone images and more complex anatomical structures in the lateral view. In this work, we propose a bone suppression method for lateral CXRs by leveraging a two-stage distillation learning strategy and a specific data correction method. Specifically, a primary model is first trained on a real DES dataset with limited samples. The bone-suppressed results on a relatively large lateral CXR dataset produced by the primary model are improved by a designed gradient correction method. Secondly, the corrected results serve as training samples to train the distillated model. By automatically learning knowledge from both the primary model and the extra correction procedure, our distillated model is expected to promote the performance of the primary model while omitting the tedious correction procedure. We adopt an ensemble model named MsDd-MAP for the primary and distillated models, which learns the complementary information of Multi-scale and Dual-domain (i.e., intensity and gradient) and fuses them in a maximum-a-posteriori (MAP) framework. Our method is evaluated on a two-exposure lateral DES dataset consisting of 46 subjects and a lateral CXR dataset consisting of 240 subjects. The experimental results suggest that our method is superior to other competing methods regarding the quantitative evaluation metrics. Furthermore, the subjective evaluation by three experienced radiologists also indicates that the distillated model can produce more visually appealing soft-tissue images than the primary model, even comparable to real DES imaging for lateral CXRs.

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

PURPOSE: 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. METHOD: 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. RESULTS: 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. CONCLUSIONS: The proposed method can stabilize the training and generate SDM images with lower intensity distortion and higher perceptual similarity.

Predicting local persistence/recurrence after radiation therapy for head and neck cancer from PET/CT using a multi-objective, multi-classifier radiomics model.

Objectives: Accurate identifying head and neck squamous cell cancer (HNSCC) patients at high risk of local persistence/recurrence (P/R) is of importance for personalized patient management. Here we developed a multi-objective, multi-classifier radiomics model for early HNSCC local P/R prediction based on post-treatment PET/CT scans and clinical data. Materials and methods: We retrospectively identified 328 individuals (69 patients have local P/R) with HNSCC treated with definitive radiation therapy at our institution. The median follow-up from treatment completion to the first surveillance PET/CT imaging was 114 days (range: 82-159 days). Post-treatment PET/CT scans were reviewed and contoured for all patients. For each imaging modality, we extracted 257 radiomic features to build a multi-objective radiomics model with sensitivity, specificity, and feature sparsity as objectives for model training. Multiple representative classifiers were combined to construct the predictive model. The output probabilities of models built with features from various modalities were fused together to make the final prediction. Results: We built and evaluated three single-modality models and two multi-modality models. The combination of PET, CT, and clinical data in the multi-objective, multi-classifier radiomics model trended towards the best prediction performance, with a sensitivity of 93%, specificity of 83%, accuracy of 85%, and AUC of 0.94. Conclusion: Our study demonstrates the feasibility of employing a multi-objective, multi-classifier radiomics model with PET/CT radiomic features and clinical data to predict outcomes for patients with HNSCC after radiation therapy. The proposed prediction model shows the potential to detect cancer local P/R early after radiation therapy.

Preoperative radiomics model using gadobenate dimeglumine-enhanced magnetic resonance imaging for predicting ß-catenin mutation in patients with hepatocellular carcinoma: A retrospective study.

Objective: To compare and evaluate radiomics models to preoperatively predict ß-catenin mutation in patients with hepatocellular carcinoma (HCC). Methods: Ninety-eight patients who underwent preoperative gadobenate dimeglumine (Gd-BOPTA)-enhanced MRI were retrospectively included. Volumes of interest were manually delineated on arterial phase, portal venous phase, delay phase, and hepatobiliary phase (HBP) images. Radiomics features extracted from different combinations of imaging phases were analyzed and validated. A linear support vector classifier was applied to develop different models. Results: Among all 15 types of radiomics models, the model with the best performance was seen in the RHBP radiomics model. The area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity of the RHBP radiomics model in the training and validation cohorts were 0.86 (95% confidence interval [CI], 0.75-0.93), 0.75, 1.0, and 0.65 and 0.82 (95% CI, 0.63-0.93), 0.73, 0.67, and 0.76, respectively. The combined model integrated radiomics features in the RHBP radiomics model, and signatures in the clinical model did not improve further compared to the single HBP radiomics model with AUCs of 0.86 and 0.76. Good calibration for the best RHBP radiomics model was displayed in both cohorts; the decision curve showed that the net benefit could achieve 0.15. The most important radiomics features were low and high gray-level zone emphases based on gray-level size zone matrix with the same Shapley additive explanation values of 0.424. Conclusion: The RHBP radiomics model may be used as an effective model indicative of HCCs with ß-catenin mutation preoperatively and thus could guide personalized medicine.

Predicting futile recanalization, malignant cerebral edema, and cerebral herniation using intelligible ensemble machine learning following mechanical thrombectomy for acute ischemic stroke.

Purpose: To establish an ensemble machine learning (ML) model for predicting the risk of futile recanalization, malignant cerebral edema (MCE), and cerebral herniation (CH) in patients with acute ischemic stroke (AIS) who underwent mechanical thrombectomy (MT) and recanalization. Methods: This prospective study included 110 patients with premorbid mRS ≤ 2 who met the inclusion criteria. Futile recanalization was defined as a 90-day modified Rankin Scale score >2. Clinical and imaging data were used to construct five ML models that were fused into a logistic regression algorithm using the stacking method (LR-Stacking). We added the Shapley Additive Explanation method to display crucial factors and explain the decision process of models for each patient. Prediction performances were compared using area under the receiver operating characteristic curve (AUC), F1-score, and decision curve analysis (DCA). Results: A total of 61 patients (55.5%) experienced futile recanalization, and 34 (30.9%) and 22 (20.0%) patients developed MCE and CH, respectively. In test set, the AUCs for the LR-Stacking model were 0.949, 0.885, and 0.904 for the three outcomes mentioned above. The F1-scores were 0.882, 0.895, and 0.909, respectively. The DCA showed that the LR-Stacking model provided more net benefits for predicting MCE and CH. The most important factors were the hypodensity volume and proportion in the corresponding vascular supply area. Conclusion: Using the ensemble ML model to analyze the clinical and imaging data of AIS patients with successful recanalization at admission and within 24 h after MT allowed for accurately predicting the risks of futile recanalization, MCE, and CH.

Clinical and imaging features of 112 patients with irregular and flat bone osteosarcoma.

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.

Predicting the molecular subtype of breast cancer and identifying interpretable imaging features using machine learning algorithms.

OBJECTIVES: To evaluate the performance of interpretable machine learning models in predicting breast cancer molecular subtypes. METHODS: We retrospectively enrolled 600 patients with invasive breast carcinoma between 2012 and 2019. The patients were randomly divided into a training (n = 450) and a testing (n = 150) set. The five constructed models were trained based on clinical characteristics and imaging features (mammography and ultrasonography). The model classification performances were evaluated using the area under the receiver operating characteristic (ROC) curve (AUC), accuracy, sensitivity, and specificity. Shapley additive explanation (SHAP) technique was used to interpret the optimal model output. Then we choose the optimal model as the assisted model to evaluate the performance of another four radiologists in predicting the molecular subtype of breast cancer with or without model assistance, according to mammography and ultrasound images. RESULTS: The decision tree (DT) model performed the best in distinguishing triple-negative breast cancer (TNBC) from other breast cancer subtypes, yielding an AUC of 0.971; accuracy, 0.947; sensitivity, 0.905; and specificity, 0.941. The accuracy, sensitivity, and specificity of all radiologists in distinguishing TNBC from other molecular subtypes and Luminal breast cancer from other molecular subtypes have significantly improved with the assistance of DT model. In the diagnosis of TNBC versus other subtypes, the average sensitivity, average specificity, and average accuracy of less experienced and more experienced radiologists increased by 0.090, 0.125, 0.114, and 0.060, 0.090, 0.083, respectively. In the diagnosis of Luminal versus other subtypes, the average sensitivity, average specificity, and average accuracy of less experienced and more experienced radiologists increased by 0.084, 0.152, 0.159, and 0.020, 0.100, 0.048. CONCLUSIONS: This study established an interpretable machine learning model to differentiate between breast cancer molecular subtypes, providing additional values for radiologists. KEY POINTS: • Interpretable machine learning model (MLM) could help clinicians and radiologists differentiate between breast cancer molecular subtypes. • The Shapley additive explanations (SHAP) technique can select important features for predicting the molecular subtypes of breast cancer from a large number of imaging signs. • Machine learning model can assist radiologists to evaluate the molecular subtype of breast cancer to some extent.

A deep learning-machine learning fusion approach for the classification of benign, malignant, and intermediate bone tumors.

OBJECTIVES: 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. METHODS: 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). RESULTS: 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. CONCLUSION: 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. KEY POINTS: • 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.

A System Pharmacology Model for Decoding the Synergistic Mechanisms of Compound Kushen Injection in Treating Breast Cancer.

Breast cancer (BC) is one of the most common malignant tumors among women worldwide and can be treated using various methods; however, side effects of these treatments cannot be ignored. Increasing evidence indicates that compound kushen injection (CKI) can be used to treat BC. However, traditional Chinese medicine (TCM) is characterized by "multi-components" and "multi-targets", which make it challenging to clarify the potential therapeutic mechanisms of CKI on BC. Herein, we designed a novel system pharmacology strategy using differentially expressed gene analysis, pharmacokinetics synthesis screening, target identification, network analysis, and docking validation to construct the synergy contribution degree (SCD) and therapeutic response index (TRI) model to capture the critical components responding to synergistic mechanisms of CKI in BC. Through our designed mathematical models, we defined 24 components as a high contribution group of synergistic components (HCGSC) from 113 potentially active components of CKI based on ADME parameters. Pathway enrichment analysis of HCGSC targets indicated that Rhizoma Heterosmilacis and Radix Sophorae Flavescentis could synergistically target the PI3K-Akt signaling pathway and the cAMP signaling pathway to treat BC. Additionally, TRI analysis showed that the average affinity of HCGSC and targets involved in the key pathways reached -6.47 kcal/mmol, while in vitro experiments proved that two of the three high TRI-scored components in the HCGSC showed significant inhibitory effects on breast cancer cell proliferation and migration. These results demonstrate the accuracy and reliability of the proposed strategy.

Pure and Mixed Tubular Carcinoma of the Breast: Mammographic Features, Clinicopathological Characteristics and Prognostic Analysis.

Objective: To evaluate the mammographic features, clinicopathological characteristics, treatments, and prognosis of pure and mixed tubular carcinomas of the breast. Materials and methods: Twenty-five tubular carcinomas were pathologically confirmed at our hospital from January 2011 to May 2019. Twenty-one patients underwent preoperative mammography. A retrospective analysis of mammographic features, clinicopathological characteristics, treatment, and outcomes was performed. Results: Altogether, 95% of the pure tubular carcinomas (PTCs) and mixed tubular carcinomas (MTCs) showed the presence of a mass or structural distortions on mammography and the difference was not statistically significant (P = .373). MTCs exhibited a larger tumor size than PTCs (P = .033). Lymph node metastasis was more common (P = .005) in MTCs. Patients in our study showed high estrogen receptor and progesterone receptor positivity rates, but low human epidermal growth factor receptor 2 positivity rate. The overall survival rate was 100% in both PTC and MTC groups and the 5-year disease-free survival rates were 100% and 75%, respectively with no significant difference between the groups (P = .264). Conclusion: Tubular carcinoma of the breast is potentially malignant and has a favorable prognosis. Digital breast tomosynthesis may improve its detection. For patients with PTC, breast-conserving surgery and sentinel lymph node biopsy are recommended based on the low rate of lymph node metastasis and good prognosis. MTC has a relatively high rate of lymph node metastasis and a particular risk of metastasis. Axillary lymph node dissection should be performed for MTC even if the tumor is smaller than 2 cm.

Improved detection of solitary pulmonary nodules on radiographs compared with deep bone suppression imaging.

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<0.01), 0.834 with CCR plus DES radiography (P<0.01), and 0.939 with DTS (P<0.01). Comparisons between CCR and CCR plus deep BSI found that the sensitivities of the assessments by the 3 residents increased from 53.2% to 69.5% (P=0.014) for nodules located in the upper lung field, from 30.6% to 44.6% (P=0.015) for nodules that were partially/completely obscured by the bone, and from 33.2% to 45.8% (P=0.006) for nodules <10 mm. CONCLUSIONS: The deep BSI technique can significantly increase the sensitivity of radiology residents for solitary pulmonary nodules compared with CCR. Increased detection was seen mainly for smaller nodules, nodules with partial/complete obscuration, and nodules located in the upper lung field.

Can the delayed phase of quantitative contrast-enhanced mammography improve the diagnostic performance on breast masses?

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 LGVT1 + LGVT2 + LGVT3 was less than that of LGVT1 + LGVT2 (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 LBRT1 + LBRT2 + LBRT3 increased by 19.6%, and specificity decreased by 20.7% compared with LBRT1 + LBRT2. The LGVT1 + LGVT2 + LGVT3 + kinetic enhancement (T1-T3) had the lowest sensitivity (67.0%), but the highest specificity (75.8%), and the sensitivity of LBRT1 + LBRT2 + LBRT3 + kinetic enhancement (T1-T3) was higher than that of LBRT1 + LBRT2 + 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.