Preoperative Prediction of Axillary Lymph Node Metastasis in Patients With Breast Cancer Through Multimodal Deep Learning Based on Ultrasound and Magnetic Resonance Imaging Images.

RATIONALE AND OBJECTIVES: Deep learning can enhance the performance of multimodal image analysis, which is known for its noninvasive attributes and complementary efficacy, in predicting axillary lymph node (ALN) metastasis. Therefore, we established a multimodal deep learning model incorporating ultrasound (US) and magnetic resonance imaging (MRI) images to predict ALN metastasis in patients with breast cancer. MATERIALS AND METHODS: A retrospective cohort of patients with histologically confirmed breast cancer from two hospitals composed of the primary cohort (n = 465) and the external validation cohort (n = 123). All patients had undergone both preoperative US and MRI scans. After data preprocessing, three convolutional neural network models were used to analyze the US and MRI images, respectively. After integrating the US and MRI deep learning prediction results (DLUS and DLMRI, respectively), a multimodal deep learning (DLMRI+US+Clinical parameter) model was constructed. The predictive ability of the proposed model was compared to that of the DLUS, DLMRI, combined bimodal (DLMRI+US), and clinical parameter models. Evaluation was performed using the area under the receiver operating characteristic curves (AUCs) and decision curves. RESULTS: A total of 588 patients with breast cancer participated in this study. The DLMRI+US+Clinical parameter model outperformed the alternative models, achieving the highest AUCs of 0.819 (95% confidence interval [CI] 0.734-0.903) and 0.809 (95% CI 0.723-0.895) on the internal and external validation sets, respectively. The decision curve analysis confirmed its clinical usefulness. CONCLUSION: The DLMRI+US＋Clinical parameter model demonstrates the feasibility and reliability of its performance for ALN metastasis prediction in patients with breast cancer.

Surgical pathway for HIV­infected patients based on the ERAS strategy (Review).

Enhanced recovery after surgery (ERAS), which is based on evidence-based medicine, focuses on patients and aims to reduce the psychological and physiological trauma stress reactions and complications of patients, thus shortening the duration of hospitalization, promoting rapid recovery and reducing medical expenses, readmission rate and mortality rates. Acquired immunodeficiency syndrome (AIDS) is caused by human immunodeficiency virus (HIV) infection. Patients with HIV/AIDS, as with other patient populations, can suffer from several surgical-related diseases. Therefore, the need for surgery in this group of patients exists and the surgical services required by patients with AIDS has gradually become an urgent matter of concern. According to relevant literature and the authors' clinical experience, the present review summarizes the current surgical approaches for patients infected with HIV based on ERAS. In the present review, the related issues observed at different stages of surgery, including pre-operative, intra-operative, post-operative and follow-up stages, are discussed.

Finite element analysis of a novel anatomical plate in posterolateral plateau fractures.

Objective: This study aims to analyze the biomechanical characteristics of posterolateral plateau fractures fixed by a novel anatomical plate using finite element analysis. Methods: A three-dimensional digital model of the full length of right tibiofibula was obtained by CT scanning. A posterolateral tibial plateau fracture model was then created. The acquired fracture model was assembled with 4 groups of internal fixations: Group A, novel anatomical plate; Group B, straight buttress plate; Group C, oblique T-shaped locking plate; Group D, two lag screws. Axial loads of 500, 1,000 and 1,500 N perpendicular to the horizontal plane were used to simulate the stress on the lateral plateau of a 65 kg person standing, walking and fast running. Results: Vertical displacements of the posterolateral fragments in each of the four groups gradually increased under loads from 500 N to 1,500 N. The maximum displacement of the fracture fragment in four groups were all located on the lateral side of the proximal part, and the displacement gradually decreased from the proximal part to the distal end. The maximum displacement values under the axial load of 1,500 N was in the following order: novel anatomical plate (1.2365 mm) < oblique T-shaped locking plate (1.314 mm) < two lag screws (1.3747 mm) < straight buttress plate (1.3932 mm). As the axial load increased, the stress value of the different internal fixation models gradually increased. The stress behavior of the same internal fixation model under different loads was similar. The maximum stress value under the axial load of 1,500 N was in the following order: novel anatomical plate (114.63 MPa) < oblique T-shaped locking plate (277.17 MPa) < two lag screws (236.75 MPa) < straight buttress plate (136.2 MPa). Conclusion: The patients with posterolateral plateau fractures fixed with a novel anatomical plate in standing, walking and fast running can achieve satisfactory biomechanical results, which lays the foundation for future applications. At the same time, clinical fracture types are often diverse and accompanied by damage to the soft tissue. Therefore, the ideal surgical approach and appropriate internal fixation must be selected based on the patient's injury condition.

Classification and recognition of milk somatic cell images based on PolyLoss and PCAM-Reset50.

Somatic cell count (SCC) is a fundamental approach for determining the quality of cattle and bovine milk. So far, different classification and recognition methods have been proposed, all with certain limitations. In this study, we introduced a new deep learning tool, i.e., an improved ResNet50 model constructed based on the residual network and fused with the position attention module and channel attention module to extract the feature information more effectively. In this paper, macrophages, lymphocytes, epithelial cells, and neutrophils were assessed. An image dataset for milk somatic cells was constructed by preprocessing to increase the diversity of samples. PolyLoss was selected as the loss function to solve the unbalanced category samples and difficult sample mining. The Adam optimization algorithm was used to update the gradient, while Warm-up was used to warm up the learning rate to alleviate the overfitting caused by small sample data sets and improve the model's generalization ability. The experimental results showed that the classification accuracy, precision rate, recall rate, and comprehensive evaluation index F value of the proposed model reached 97%, 94.5%, 90.75%, and 92.25%, respectively, indicating that the proposed model could effectively classify the milk somatic cell images, showing a better classification performance than five previous models (i.e., ResNet50, ResNet18, ResNet34, AlexNet andMobileNetv2). The accuracies of the ResNet18, ResNet34, ResNet50, AlexNet, MobileNetv2, and the new model were 95%, 93%, 93%, 56%, 37%, and 97%, respectively. In addition, the comprehensive evaluation index F1 showed the best effect, fully verifying the effectiveness of the proposed method in this paper. The proposed method overcame the limitations of image preprocessing and manual feature extraction by traditional machine learning methods and the limitations of manual feature selection, improving the classification accuracy and showing a strong generalization ability.

An Online Mammography Database with Biopsy Confirmed Types.

Breast carcinoma is the second largest cancer in the world among women. Early detection of breast cancer has been shown to increase the survival rate, thereby significantly increasing patients' lifespan. Mammography, a noninvasive imaging tool with low cost, is widely used to diagnose breast disease at an early stage due to its high sensitivity. Although some public mammography datasets are useful, there is still a lack of open access datasets that expand beyond the white population as well as missing biopsy confirmation or with unknown molecular subtypes. To fill this gap, we build a database containing two online breast mammographies. The dataset named by Chinese Mammography Database (CMMD) contains 3712 mammographies involved 1775 patients, which is divided into two branches. The first dataset CMMD1 contains 1026 cases (2214 mammographies) with biopsy confirmed type of benign or malignant tumors. The second dataset CMMD2 includes 1498 mammographies for 749 patients with known molecular subtypes. Our database is constructed to enrich the diversity of mammography data and promote the development of relevant fields.

Splenic Artery Steal Syndrome in Patients with Liver Cirrhosis: A Retrospective Clinical Study.

BACKGROUND Splenic artery steal syndrome (SASS) can aggravate liver damage in patients with cirrhosis. This study explored whether SASS could be an effective therapeutic target for improving hepatic artery perfusion and liver function in patients with decompensated cirrhosis. MATERIAL AND METHODS Based on inclusion and exclusion criteria, 87 patients with hepatitis B cirrhosis and portal hypertension hypersplenism admitted to our General Surgery Department for splenectomy and pericardial devascularization surgery were selected. A total of 35 cases met the diagnostic criteria of SASS and were assigned to the SASS group; the remaining 52 cases were assigned to the control group. The indicators before, during, and after surgery were compared between the 2 groups. RESULTS There were no significant differences in preoperative and intraoperative indicators between SASS group and control group (P>0.05). The MELD score 7 days after surgery and the hepatic artery diameter and hepatic artery velocity 14 days after surgery in both groups were significantly better than before surgery. The MELD score 7 days after surgery in the SASS group was significantly better than that in the control group, and the hepatic artery diameter and hepatic artery velocity 14 days after surgery in the SASS group were significantly better than those in the control group (P<0.05). CONCLUSIONS Splenectomy and pericardial devascularization surgery was an effective treatment to redirect blood flow to the hepatic artery for cirrhotic patients diagnosed with SASS. The introduction of cirrhotic SASS into clinical practice may benefit more patients with cirrhotic portal hypertension and hypersplenism.

A radiomics-based model can predict recurrence-free survival of hepatocellular carcinoma after curative ablation.

BACKGROUND: Prediction of early recurrence (ER) of HCC after radical treatment is of great significance for follow-up and subsequent treatment, and there is a lot of unmet needs. Here, our goal is to develop and validate a radiomics nomogram that can predict ER after curative ablation. OBJECTIVE: The aim of this study was to evaluate the efficacy and safety of regorafenib after disease progression with sorafenib in Chinese patients with advanced HCC through this retrospective analysis. METHODS: 149 HCC patients treated between November 2008 and February 2018 were enrolled and randomly divided into training cohort (n = 105) and validation cohort (n = 44). The survival endpoint was recurrence-free survival (RFS). A total of 16908 radiomics features were extracted from the contrast-enhanced MR images of each patient. The minimum redundancy maximum relevance algorithm (mRMR) and random survival forest (RSF) were used for feature selection. Twelve kinds of support vector machine (SVM) models, a Cox regression model (Cox PH), a random survival forest (RSF) model and a gradient boosting model (GBoost) were used to build a radiomics signature. These models were trained after adjusting the model parameters using 5-fold cross-validation. The best models were selected according to the C-index. RESULTS: Using the machine learning (ML) framework, 40 features were identified that demonstrated good prediction of HCC recurrence across all cohorts. The random survival forest (RSF) model showed higher prognostic value, with a C-index of 0.733-0.801 and an integrated Brier score of 0.147-0.165, compared with other SVM models, Cox regression models, etc. (all P < 0.05). Time-dependent receiver operating characteristic (ROC) curve analysis, survival analysis, and decision curve analysis (DCA) were used to verify the performance of the RSF model in predicting tumor recurrence. CONCLUSION: We successfully built a radiomics-based RSF model with integrated radiomics and clinicopathological features that can potentially be used to predict ER after curative ablation in HCC patients.

MRI-Based Radiomics Features to Predict Treatment Response to Neoadjuvant Chemotherapy in Locally Advanced Rectal Cancer: A Single Center, Prospective Study.

This is a prospective, single center study aimed to evaluate the predictive power of peritumor and intratumor radiomics features assessed using T2 weight image (T2WI) of baseline magnetic resonance imaging (MRI) in evaluating pathological good response to NAC in patients with LARC (including Tany N+ or T3/4a Nany but not T4b). In total, 137 patients with LARC received NAC between April 2014 and August 2020. All patients were undergoing contrast-enhanced MRI and 129 patients contained small field of view (sFOV) sequence which were performed prior to treatment. The tumor regression grade standard was based on pathological response. The training and validation sets (n=91 vs. n=46) were established by random allocation of the patients. Receiver operating characteristic curve (ROC) analysis was applied to estimate the performance of different models based on clinical characteristics and radiomics features obtained from MRI, including peritumor and intratumor features, in predicting treatment response; these effects were calculated using the area under the curve (AUC). The performance and agreement of the nomogram were estimated using calibration plots. In total, 24 patients (17.52%) achieved a complete or near-complete response. For the individual radiomics model in the validation set, the performance of peritumor radiomics model in predicting treatment response yield an AUC of 0.838, while that of intratumor radiomics model is 0.805, which show no statically significant difference between then(P>0.05). The traditional and selective clinical features model shows a poor predictive ability in treatment response (AUC=0.596 and 0.521) in validation set. The AUC of combined radiomics model was improved compared to that of the individual radiomics models in the validation sets (AUC=0.844). The combined clinic-radiomics model yield the highest AUC (0.871) in the validation set, although it did not improve the performance of the radiomics model for predicting treatment response statically (P>0.05). Good agreement and discrimination were observed in the nomogram predictions. Both peritumor and intratumor radiomics features performed similarly in predicting a good response to NAC in patients with LARC. The clinic-radiomics model showed the best performance in predicting treatment response.

Real-Life Experience of Regorafenib in Patients With Advanced Hepatocellular Carcinoma.

Background: The RESORCE trial reported that regorafenib was effective as the second-line treatment for patients with hepatocellular carcinoma (HCC) after progression on sorafenib. Real-world data are needed to assess clinical outcomes and adverse events in the setting of daily practice. Objective: We aimed to evaluate the efficacy and safety of regorafenib after disease progression with sorafenib in Chinese patients with advanced HCC. Patients and Methods: A total of 41 patients with advanced HCC who did not respond to sorafenib and followed a regorafenib regimen were enrolled in this retrospective study. Overall survival (OS), progression-free survival (PFS), radiological responses, and adverse events (AEs) were evaluated. Survival curves were compared by using the log-rank test and constructed with the Kaplan-Meier method. Results: The median PFS with regorafenib was 6.6 months (range: 5.0-8.2 months), and the median OS with regorafenib was not reached. The 1-year OS rate of regorafenib was 66.4%. The median OS of sequential sorafenib to regorafenib treatment was 35.3 months [95% confidence interval (CI), 24.3-46.3], and the 2-year OS rate of sequential sorafenib to regorafenib treatment was 74.4%. The most common AEs of regorafenib treatment were elevated aspartate aminotransferase [17/41 patients (41.5%)], elevated alanine aminotransferase [16/41 patients (39%)] and hand-foot syndrome [14/41 patients (34.1%)]. Conclusion: Regorafenib appears to be safe and clinically effective in patients with advanced HCC who progressed on first-line sorafenib.

Ultrasound-Based Radiomics Analysis for Predicting Disease-Free Survival of Invasive Breast Cancer.

BACKGROUND: Accurate prediction of recurrence is crucial for personalized treatment in breast cancer, and whether the radiomics features of ultrasound (US) could be used to predict recurrence of breast cancer is still uncertain. Here, we developed a radiomics signature based on preoperative US to predict disease-free survival (DFS) in patients with invasive breast cancer and assess its additional value to the clinicopathological predictors for individualized DFS prediction. METHODS: We identified 620 patients with invasive breast cancer and randomly divided them into the training (n = 372) and validation (n = 248) cohorts. A radiomics signature was constructed using least absolute shrinkage and selection operator (LASSO) Cox regression in the training cohort and validated in the validation cohort. Univariate and multivariate Cox proportional hazards model and Kaplan-Meier survival analysis were used to determine the association of the radiomics signature and clinicopathological variables with DFS. To evaluate the additional value of the radiomics signature for DFS prediction, a radiomics nomogram combining the radiomics signature and clinicopathological predictors was constructed and assessed in terms of discrimination, calibration, reclassification, and clinical usefulness. RESULTS: The radiomics signature was significantly associated with DFS, independent of the clinicopathological predictors. The radiomics nomogram performed better than the clinicopathological nomogram (C-index, 0.796 vs. 0.761) and provided better calibration and positive net reclassification improvement (0.147, P = 0.035) in the validation cohort. Decision curve analysis also demonstrated that the radiomics nomogram was clinically useful. CONCLUSION: US radiomics signature is a potential imaging biomarker for risk stratification of DFS in invasive breast cancer, and US-based radiomics nomogram improved accuracy of DFS prediction.

A practical study of the hepatic vascular system anatomy of the caudate lobe.

BACKGROUND: This study aimed to examine the inflow and outflow vascular system of the caudate lobe and determine its relevance to hepatobiliary surgery. METHODS: A total of 41 cadaveric liver specimens were dissected in 2019 to evaluate the inflow and outflow vascular system of the caudate lobe. RESULTS: The Glisson's pedicles of the paracaval portion were mainly from the right pedicle in 14 cases (34.15%), mainly from the left pedicle in 22 cases (53.66%), and equally from the left and right pedicle in 5 cases (12.19%). Many thick branches of the portal vein were found behind the plane consisting of the hilar plate and Arantius ligament, but none of them were thicker than 1 mm in front of the plane. All of the veins of the caudate lobe drained into the inferior vena cava (IVC) via the anterior face. There was an avascular zone without short hepatic veins (SHVs) consisting of loose connective tissue between the retrohepatic IVC and caudate lobe, with its length and width being 45-97 mm and 6-15 mm, respectively. CONCLUSIONS: The plane consisting of the hilar plate and Arantius ligament can be regarded as the boundary between the caudate lobe and the other lobes. There is an avascular zone without SHVs consisting of loose connective tissue between the retrohepatic IVC and caudate lobe.

Radiomics Model Based on Gadoxetic Acid Disodium-Enhanced MR Imaging to Predict Hepatocellular Carcinoma Recurrence After Curative Ablation.

BACKGROUND: A practical prognostic prediction model is absent for hepatocellular carcinoma (HCC) patients after curative ablation. We aimed to develop a radiomics model based on gadoxetic acid disodium-enhanced magnetic resonance (MR) images to predict HCC recurrence after curative ablation. METHODS: We retrospectively enrolled 132 patients with HCC who underwent curative ablation. Patients were randomly divided into the training (n = 92) and validation (n = 40) cohorts. Radiomic features were extracted from gadoxetic acid disodium-enhanced MR images of the liver before curative ablation, and various baseline clinical characteristics were collected. Cox regression and random survival forests were used to construct models that incorporated radiomic features and/or clinical characteristics. The predictive performance of the different models was compared using the concordance index (C-index) and decision curves analysis (DCA). A cutoff derived from the combined model was used for risk categorization, and recurrence-free survival (RFS) was compared between groups using the Kaplan-Meier survival curve analysis. RESULTS: Twenty radiomic features and four clinical characteristics were identified and used for model construction. The radiomics model constructed by tumoral and peritumoral radiomic features had better predictive performance (C-index 0.698, 95% confidence interval [CI] 0.640-0.755) compared with the clinical model (C-index 0.614, 95% CI 0.499-0.695), while the combined model had the best predictive performance (C-index 0.706, 95% CI 0.638-0.763). A better net benefit was observed with the combined model compared with the other two models according to the DCA. Distinct RFS distributions were observed when patients were categorized based on the cutoff derived from the combined model (Log rank test, p = 0.007). CONCLUSION: The radiomics model which combined radiomic features extracted from gadoxetic acid disodium-enhanced MR images with clinical characteristics could predict HCC recurrence after curative ablation.

Radiomics-based model using gadoxetic acid disodium-enhanced MR images: associations with recurrence-free survival of patients with hepatocellular carcinoma treated by surgical resection.

PURPOSE: To develop a prediction model that combined magnetic resonance images (MRI)-based radiomics features with clinical factors to predict recurrence-free survival (RFS) of hepatocellular carcinoma (HCC) patients treated with surgical resection. METHODS: HCC patients treated with surgical resection (n = 153) were randomly divided into training (n = 107) and validation (n = 46) datasets. The volumes of interest were manually outlined around the lesion and additional 2 mm and 5 mm peritumoral areas were created with automated dilatation in MRI to extract tumoral (T) and peritumoral (PT) radiomics features. The radiomics models were constructed using least absolute shrinkage and selection operator Cox regression. The combined model incorporated clinical factors and radiomics features using multivariable Cox regression based on the Akaike information criterion principle. Predictive performance of different models were evaluated by receiver operating characteristic (ROC) curves, decision curves, and calibration curves. RESULTS: Among the radiomics models, similar performance was observed in the 2 mm and 5 mm PT models (C-index both 0.657), which were better than the T model or T + PT model (C-index 0.607 and 0.641, respectively) in the validation dataset, whereas the model combined with the three identified clinical risk factors showed the best performance (C-index 0.725). Results of the ROC curves, decision curves, and the calibration curves indicated that the combined model and the derived nomogram had better prediction performance, greater clinical benefits, and fair calibration efficiency. CONCLUSION: The prediction model that combined MRI radiomics signatures with clinical factors can effectively predict the prognosis of patients with HCC treated with surgical resection.

Differentiating benign from malignant pancreatic cysts on computed tomography.

PURPOSE: It is important to identify features on computed tomography (CT) that can distinguish between benign and premalignant or malignant pancreatic cysts to avoid unnecessary surgeries. This study investigated the preoperative diagnostic evaluation of cystic pancreatic lesions to determine how advanced imaging and clinical factors should guide management. METHODS: In total, 53 patients with 27 benign and 26 premalignant or malignant cysts were enrolled. CT features of the cysts were compared using univariate and multivariate analyses. RESULTS: On univariate analysis, a solid component (p < 0.01), septation (p < 0.01), location (p < 0.01), border (p < 0.01), wall enhancement (p = 0.01), lesion margins (p < 0.01), pancreatic atrophy (p = 0.04), and a cystic wall (p < 0.01) were all significantly different between benign and premalignant or malignant cysts. On multivariate analysis, only a solid component (p < 0.01) and septation (p < 0.01) were significant. CONCLUSION: A thin cystic wall, uniform homogeneity, a clear border, the presence of septation, pancreatic atrophy, and the absence of both wall enhancements and solid components were more frequently seen in benign cysts. A thick wall, lack of homogeneity, the presence of wall enhancements and solid components, absence of septation, only a small degree of pancreatic atrophy, and unclear borders were more frequent among premalignant or malignant cysts. The only CT features to differentiate benign from premalignant or malignant cysts were a solid component and septation.

Modified Anatomic Locking Plate for the Treatment of Posteromedial Tibial Plateau Fractures.

OBJECTIVE: To evaluate the safety and clinical efficacy of a modified anatomic locking plate for the treatment of posteromedial tibial plateau fractures. METHODS: A retrospective study was performed in our department. Between January 2014 and February 2017, 11 patients with posteromedial tibial plateau fractures underwent surgery with the new anatomic locking plate for the posteromedial tibial plateau via the posteromedial approach. The study included 7 male and 4 female patients, with a mean age at the time of the operation of 39 years. During surgery, operation time and blood loss were recorded. Clinical evaluation was performed using the Tegner-Lysholm functional score, the Rasmussen functional score, and the Rasmussen anatomical score. RESULTS: The mean follow-up time of the study was 35 months. The mean interval between the time of injury and the surgery was 7.4 days. Radiological fracture union was evident in all patients at 14 weeks. During surgery, the blood loss ranged from 50 to 150 mL, and the duration ranged from 55 to 90 min. The Tegner-Lysholm functional score ranged from 80 to 96 at the final follow up. Moreover, the final Rasmussen functional score ranged from 25 to 28, and the Rasmussen anatomical score ranged from 15 to 18. The mean knee arc of motion was 137° (range, 122°-153°). Symptoms of knee instability or severe pain were not found in any cases. No flexion contractures or extensor lag was seen. No infection, deep vein thrombosis, or graft site morbidity was seen at the follow up. No case of reduction loss or internal fixation failure was reported during the follow-up. CONCLUSION: With the clinical data of the small-sample-size population (11 patients) during a 19 to 60-month follow-up, the modified anatomic locking plate for the posteromedial tibial plateau proved to be safe and effective and is an adequate fixation method for the treatment of posteromedial tibial plateau fractures.

Effect of lesser trochanter posteromedial wall defect on the stability of femoral intertrochanteric fracture using 3D simulation.

BACKGROUND: This study investigated the effects of posteromedial fracture fragments on the postoperative stability of intertrochanteric fractures of the femur by analyzing the quantity and range of fragments in CT 3D reconstruction. MATERIALS AND METHODS: Patients diagnosed with femoral lesser trochanter fractures were collected from September 2015 to February 2018. CT 3D reconstruction was applied to evaluate the quantity and extension of posteromedial fragments and the presence of isolated medial fragments. The stability of postoperative fracture was evaluated by comparing the changes of "neck-shaft angle" and "telescoping" from 1 week to 1 year after operation. RESULTS: A total of 143 patients were finally confirmed, in which 63 patients contained isolated fragments on the medial side, and the average number of fragments in the posteromedial side was 1.93 ± 0.34, which accounted for an average of about 86.11% ± 8.20% in the whole posteromedial wall. When the number of posteromedial fragments was > 2 and the range of posteromedial fragments was > 75%, then the changes in the neck-shaft angle and "telescoping" showed statistical significance (12.27 ± 4.18 mm and 10.13 ± 6.17°, respectively), and when there were isolated medial isolated fragments, then the change in the neck-shaft angle was 10.66 ± 4.27°, showing statistical significance. CONCLUSIONS: These findings revealed a certain correlation between the quantity and the range of posteromedial fragments and the postoperative "shortening" and "collapse" of femoral intertrochanteric fractures.

Deep Convolutional Neural Networks-Based Automatic Breast Segmentation and Mass Detection in DCE-MRI.

Breast segmentation and mass detection in medical images are important for diagnosis and treatment follow-up. Automation of these challenging tasks can assist radiologists by reducing the high manual workload of breast cancer analysis. In this paper, deep convolutional neural networks (DCNN) were employed for breast segmentation and mass detection in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). First, the region of the breasts was segmented from the remaining body parts by building a fully convolutional neural network based on U-Net++. Using the method of deep learning to extract the target area can help to reduce the interference external to the breast. Second, a faster region with convolutional neural network (Faster RCNN) was used for mass detection on segmented breast images. The dataset of DCE-MRI used in this study was obtained from 75 patients, and a 5-fold cross validation method was adopted. The statistical analysis of breast region segmentation was carried out by computing the Dice similarity coefficient (DSC), Jaccard coefficient, and segmentation sensitivity. For validation of breast mass detection, the sensitivity with the number of false positives per case was computed and analyzed. The Dice and Jaccard coefficients and the segmentation sensitivity value for breast region segmentation were 0.951, 0.908, and 0.948, respectively, which were better than those of the original U-Net algorithm, and the average sensitivity for mass detection achieved 0.874 with 3.4 false positives per case.

A comparison of deltopectoral versus Judet approach for glenoid exposure.

BACKGROUND: Open reduction-internal fixation via an anterior or posterior approach is a widely used method for treating displaced glenoid fractures. This study aimed to identify the exposure range of the glenoid rim by these 2 approaches (deltopectoral and Judet approaches) and provide reference data for the choice of surgical approach. METHODS: Twelve cadaveric shoulders were dissected. Both deltopectoral and Judet approaches were performed on each shoulder to mark the glenoid fracture. In addition, the shoulder was disarticulated to record the exposure range of the glenoid rim. RESULTS: For the deltopectoral approach, the range of the exposed glenoid rim was from 5:50 to 11:30, which accounted for about 47.2% of the clock face. For the Judet approach, the range of the exposed glenoid rim was from 1:30 to 6:20, which accounted for about 40.3% of the clock face. Along the inferior glenoid, there was an area of partial overlap for the 2 approaches. The superior glenoid rim located from 11:30 to 1:30 was considered inaccessible, as it could not be exposed by the 2 approaches. CONCLUSION: Less than 50% of the glenoid rim can be exposed by the deltopectoral or Judet approach. With a single approach, it may be difficult to expose and fix some complex glenoid fractures. The superior part of the glenoid fracture is the non-access area via the deltopectoral or Judet approach.

The Impact of Preoperative Radiomics Signature on the Survival of Breast Cancer Patients With Residual Tumors After NAC.

BACKGROUND: Residual cancer cells remaining after chemotherapy may have more aggressive behavior that promotes recurrence or metastasis, and which patients would benefit from subsequent additional treatment is controversial. The purpose of our study was to evaluate the prognostic value of the preoperative radiomics features of computed tomography (CT) imaging in breast cancer (BC) patients with residual tumors after neoadjuvant chemotherapy (NAC). METHODS: Post-NAC CT images were reviewed from 114 patients who had received breast surgery and had residual breast tumors. The association of the 110 radiomics features derived from CT images with 5-year disease-free survival (DFS) was assessed by log-rank test in the training cohort, resulting in 13 prognostic radiomics features. RESULTS: We constructed a radiomics signature consisting of four selected features by using least absolute shrinkage and selection operator (LASSO) Cox regression analysis, which performed well in the discrimination with an area under the curve (AUC) of 0.78 (95% CI, 0.67-0.89) and 0.73 (95% CI, 0.59-0.87) in the training and validation cohorts, respectively. Radiomics nomogram, incorporating the radiomics signature with the conventional clinical variables, also performed well in the two cohorts (training cohort: AUC, 0.84; validation cohort: AUC, 0.82). Moreover, we found that the high-risk patients determined by our radiomics nomogram could benefit from postoperative adjuvant chemotherapy, while the low-risk and total patient groups could not. CONCLUSIONS: Our novel radiomics nomogram is a promising and favorable prognostic biomarker for preoperatively predicting survival outcomes and may aid in clinical decision-making in BC patients with residual tumors after NAC.