Segmentation of lung nodules based on a refined segmentation network.

BACKGROUND: Accurate segmentation of lung nodules is of great significance for early screening and diagnosis of lung cancer. PURPOSE: However, the heterogeneity of lung nodules and the similarities between them and other lung tissues make it difficult to accurately segment these nodules. As regards the use of deep learning to segment lung nodules, convolutional neural networks would gradually lead to errors accumulating at the network layer due to the presence of multiple upsampling and downsampling layers, resulting in poor segmentation results. METHODS: In this study, we developed a refined segmentation network (RS-Net) for lung nodule segmentation to solve this problem. Accordingly, the proposed RS-Net was first used to locate the core region of the lung nodules and to gradually refine the segmentation results of the core region. In addition, to solve the problem of misdetection of small-sized nodules owing to the imbalance of positive and negative samples, we devised an average dice-loss function computed on nodule level. By calculating the loss of each nodule sample to measure the overall loss, the network can address the misdetection problem of lung nodules with smaller diameters more efficiently. RESULTS: Our method was evaluated based on 1055 lung nodules from Lung Image Database Consortium data and a set of 120 lung nodules collected from Shanghai Chest Hospital for additional validation. The segmentation dice coefficients of RS-Net on these two datasets were 85.90% and 81.13%, respectively. The analysis of the segmentation effect of different properties and sizes of nodules indicates that RS-Net yields a stable segmentation effect. CONCLUSIONS: The results show that the segmentation strategy based on gradual refinement can considerably improve the segmentation of lung nodules.

Correlation between the CEM imaging characteristics and different molecular subtypes of breast cancer.

PURPOSE: To investigate the correlation between the contrast-enhanced mammography (CEM) imaging characteristics and different molecular subtypes of breast cancer (BC). METHODS: We retrospectively included 313 eligible female patients who underwent CEM examination and surgery in our hospital from July 2017 to July 2021. Their lesions were confirmed on histopathological examination and immunohistochemical analysis. BC was divided into luminal A, luminal B, HER2-enriched, and triple-negative BC (TNBC) subtypes according to immunohistochemical markers. Nine features were extracted from CEM images, including tumor shape, margins, spiculated mass, lobulated mass, malignant calcification, lesion conspicuity, internal enhancement pattern, multifocal mass, and swollen axillary lymph nodes. Statistical analysis was performed using SPSS 25.0. Univariate analysis and binomial regression were used to analyze the correlation between CEM imaging features and BC molecular subtypes. RESULTS: There were 184 (58.8 %) Luminal A, 44 (14.1 %) Luminal B, 47 (15.0 %) HER-2-enriched and 38 (12.1 %) TNBC, respectively. Molecular subtypes were significantly related to the tumor shape, margins, spiculated mass, internal enhancement pattern, malignant calcification and swollen axillary lymph nodes. Spiculated and calcified tumors were associated with Luminal subtypes, especially Luminal B (P < 0.05). Irregular tumor shape and malignant calcification were associated with HER-2-enriched subtype (P < 0.05). Oval or round tumor shape, rim enhancement, and swollen axillary lymph nodes were associated with TNBC (P < 0.05). CONCLUSION: CEM imaging features could distinguish BC molecular subtypes. In particular, TNBC showed oval or round tumor shape, rim enhancement, and swollen axillary lymph nodes, providing insights into the diagnosis and prognosis of TNBC.

Predictive value of lesion morphology in rectal cancer based on MRI before surgery.

OBJECTIVE: To explore the relationship of MRI morphology of primary rectal cancer with extramural vascular invasion (EMVI), metastasis and local recurrence. MATERIALS AND METHODS: This retrospective study included 153 patients with rectal cancer. Imaging factors and histopathological index including nodular projection (NP), cord sign (CS) at primary tumor margin, irregular nodules (IN) of mesorectum, MRI-detected peritoneal reflection invasion (PRI), range of rectal wall invasion (RRWI), patterns and length of tumor growth, maximal extramural depth (EMD), histologically confirmed local node involvement (hLN), MRI T stage, MRI N stage, MRI-detected extramural vascular invasion (mEMVI) and histologically confirmed extramural vascular invasion (hEMVI) were evaluated. Determining the relationship between imaging factors and hEMVI, synchronous metastasis and local recurrence by univariate analysis and multivariable logistic regression, and a nomogram validated internally via Bootstrap self-sampling was constructed based on the latter. RESULTS: Thirty-eight cases of hEMVI, fourteen cases of synchronous metastasis and ten cases of local recurrence were observed among 52 NP cases. There were 50 cases of mEMVI with moderate consistency with hEMVI (Kappa = 0.614). NP, CS, EMD and mEMVI showed statistically significant differences in the negative and positive groups of hEMVI, synchronous metastasis, and local recurrence. Compared to patients with local mass growth, the rectal tumor with circular infiltration had been found to be at higher risk of synchronous metastasis and local recurrence (P < 0.05). NP and IN remained as significant predictors for hEMVI, and mEMVI was a predictor for synchronous metastasis, while PRI and mEMVI were predictors for local recurrences. The nomogram for predicting hEMVI demonstrated a C-index of 0.868, sensitivity of 86.0%, specificity of 79.6%, and accuracy of 81.7%. CONCLUSION: NP, CS, IN, large EMD, mEMVI, and circular infiltration are significantly associated with several adverse prognostic indicators. The nomogram based on NP has good predictive performance for preoperative EMVI. mEMVI is a risk factor for synchronous metastasis. PRI and mEMVI are risk factors for local recurrence.

Molecular-subtype guided automatic invasive breast cancer grading using dynamic contrast-enhanced MRI.

BACKGROUND AND OBJECTIVES: Histological grade and molecular subtype have presented valuable references in assigning personalized or precision medicine as the significant prognostic indicators representing biological behaviors of invasive breast cancer (IBC). To evaluate a two-stage deep learning framework for IBC grading that incorporates with molecular-subtype (MS) information using DCE-MRI. METHODS: In Stage I, an innovative neural network called IOS2-DA is developed, which includes a dense atrous-spatial pyramid pooling block with a pooling layer (DA) and inception-octconved blocks with double kernel squeeze-and-excitations (IOS2). This method focuses on the imaging manifestation of IBC grades and performs preliminary prediction using a novel class F1-score loss function. In Stage II, a MS attention branch is introduced to fine-tune the integrated deep vectors from IOS2-DA via Kullback-Leibler divergence. The MS-guided information is weighted with preliminary results to obtain classification values, which are analyzed by ensemble learning for tumor grade prediction on three MRI post-contrast series. Objective assessment is quantitatively evaluated by receiver operating characteristic curve analysis. DeLong test is applied to measure statistical significance (P < 0.05). RESULTS: The molecular-subtype guided IOS2-DA performs significantly better than the single IOS2-DA in terms of accuracy (0.927), precision (0.942), AUC (0.927, 95% CI: [0.908, 0.946]), and F1-score (0.930). The gradient-weighted class activation maps show that the feature representations extracted from IOS2-DA are consistent with tumor areas. CONCLUSIONS: IOS2-DA elucidates its potential in non-invasive tumor grade prediction. With respect to the correlation between MS and histological grade, it exhibits remarkable clinical prospects in the application of relevant clinical biomarkers to enhance the diagnostic effectiveness of IBC grading. Therefore, DCE-MRI tends to be a feasible imaging modality for the thorough preoperative assessment of breast biological behavior and carcinoma prognosis.

Weakly Supervised Breast Lesion Detection in Dynamic Contrast-Enhanced MRI.

Currently, obtaining accurate medical annotations requires high labor and time effort, which largely limits the development of supervised learning-based tumor detection tasks. In this work, we investigated a weakly supervised learning model for detecting breast lesions in dynamic contrast-enhanced MRI (DCE-MRI) with only image-level labels. Two hundred fifty-four normal and 398 abnormal cases with pathologically confirmed lesions were retrospectively enrolled into the breast dataset, which was divided into the training set (80%), validation set (10%), and testing set (10%) at the patient level. First, the second image series S2 after the injection of a contrast agent was acquired from the 3.0-T, T1-weighted dynamic enhanced MR imaging sequences. Second, a feature pyramid network (FPN) with convolutional block attention module (CBAM) was proposed to extract multi-scale feature maps of the modified classification network VGG16. Then, initial location information was obtained from the heatmaps generated using the layer class activation mapping algorithm (Layer-CAM). Finally, the detection results of breast lesion were refined by the conditional random field (CRF). Accuracy, sensitivity, specificity, and area under the receiver operating characteristic (ROC) curve (AUC) were utilized for evaluation of image-level classification. Average precision (AP) was estimated for breast lesion localization. Delong's test was used to compare the AUCs of different models for significance. The proposed model was effective with accuracy of 95.2%, sensitivity of 91.6%, specificity of 99.2%, and AUC of 0.986. The AP for breast lesion detection was 84.1% using weakly supervised learning. Weakly supervised learning based on FPN combined with Layer-CAM facilitated automatic detection of breast lesion.

Weakly supervised breast lesion detection in DCE-MRI using self-transfer learning.

BACKGROUND: Breast cancer is a typically diagnosed and life-threatening cancer in women. Thus, dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is increasingly used for breast lesion detection and diagnosis because of the high resolution of soft tissues. Moreover, supervised detection methods have been implemented for breast lesion detection. However, these methods require substantial time and specialized staff to develop the labeled training samples. PURPOSE: To investigate the potential of weakly supervised deep learning models for breast lesion detection. METHODS: A total of 1003 breast DCE-MRI studies were collected, including 603 abnormal cases with 770 breast lesions and 400 normal subjects. The proposed model was trained using breast DCE-MRI considering only the image-level labels (normal and abnormal) and optimized for classification and detection sub-tasks simultaneously. Ablation experiments were performed to evaluate different convolutional neural network (CNN) backbones (VGG19 and ResNet50) as shared convolutional layers, as well as to evaluate the effect of the preprocessing methods. RESULTS: Our weakly supervised model performed better with VGG19 than with ResNet50 (p < 0.05). The average precision (AP) of the classification sub-task was 91.7% for abnormal cases and 88.0% for normal samples. The area under the receiver operating characteristic (ROC) curve (AUC) was 0.939 (95% confidence interval [CI]: 0.920-0.941). The weakly supervised detection task AP was 85.7%, and the correct location (CorLoc) was 90.2%. A sensitivity of 84.0% at two-false positives per image was assessed based on free-response ROC (FROC) curve. CONCLUSIONS: The results confirm that a weakly supervised CNN based on self-transfer learning is an effective and promising auxiliary tool for detecting breast lesions.

STEMI in an adolescent boy due to anomalous left main coronary artery arising from the right sinus of Valsalva: Case report and brief review of the literature.

Congenital anomalous origin of coronary artery is a rare cardiovascular malformation and the most common anomaly is the left circumflex (LCX) arising from the right sinus of Valsalva (RSV). Other forms include both coronary arteries from RSV, the left anterior descending coronary artery from RSV, and a single coronary artery from the left sinus of Valsalva. Despite being rare, anomalous origin of left main coronary artery (LMCA) from RSV carries a high risk of sudden cardiac death. Here, we report a case of 13-year-old boy with chest pain and acute extensive anterior ST-segment elevation myocardial infarction (STEMI) who was initially diagnosed as acute myocarditis in the emergency department. A bedside echocardiogram showed severe global hypokinesia of left ventricle (LV) and normal right ventricle (RV) function. Coronary computed tomography angiography (CCTA) examination showed LMCA originated from the RSV. The patient underwent coronary artery bypass grafting surgery and was discharged without complications. A timely correct diagnosis of an anomalous coronary artery is critical in symptomatic patients, CCTA plays an important role in clinical decision making.

A method for distinguishing benign and malignant pulmonary nodules based on 3D dual path network aided by K-means clustering analysis.

In the USA, there were about 1 |806 |590 new cancer cases in 2020, and 606 520 cancer deaths are expected to have occurred in 2021. Lung cancer has become the leading cause of death from cancer in both men and women (Siegel et al., 2020). Clinical studies show that the five-year survival rate of lung cancer patients after early diagnosis and treatment intervention can reach 80%, compared with that of patients having advanced lung cancer. Thus, the early diagnosis of lung cancer is a key factor to reduce mortality.

Comparison of Contrast-Enhanced Spectral Mammography and Contrast-Enhanced MRI in Screening Multifocal and Multicentric Lesions in Breast Cancer Patients.

Objectives: We aimed to determine the difference between contrast-enhanced spectral mammography (CESM) and contrast-enhanced magnetic resonance imaging (CE-MRI) in detecting multifocal and multicentric breast cancer (MMBC). Methods: : This study was conducted among breast cancer patients between July 1, 2017, and May 30, 2021. The sensitivity, specificity, and accuracy of CESM and CE-MRI in the diagnosis of MMBC were evaluated with pathological results as the gold standard. Results: A total of 188 lesions were detected in 54 patients with MMBC, including 177 breast cancer and 11 benign lesions. Based on CESM and CE-MRI, 4 false-positive cases and 3 false-negative cases and 7 false-positive cases and 1 false-negative case, respectively, were found. The accuracy of CESM was higher than that of MRI (96.3% vs 95.7%), and the specificity was higher than that of MRI (63.6% vs 36.4%). There were no significant differences in the sensitivity, specificity, and accuracy for the detection of MMBC between CESM and CE-MRI (p = 0.500; p = 0.250; p = 0.792). Conclusion: CESM is an effective method for the detection of MMBC, which is consistent with the sensitivity and accuracy of CE-MRI.

Transfer Learning Strategy Based on Unsupervised Learning and Ensemble Learning for Breast Cancer Molecular Subtype Prediction Using Dynamic Contrast-Enhanced MRI.

BACKGROUND: Imaging-driven deep learning strategies focus on training from scratch and transfer learning. However, the performance of training from scratch is often impeded by the lack of large-scale labeled training data. Additionally, owing to the differences between source and target domains, analyzing medical image tasks satisfactorily via transfer learning based on ImageNet is difficult. PURPOSE: To investigate two transfer learning algorithms for breast cancer molecular subtype prediction (luminal and non-luminal) based on unsupervised pre-training and ensemble learning: M_EL and B_EL, using malignant and benign datasets as the source domain, respectively. STUDY TYPE: Retrospective. POPULATION: Eight hundred and thirty-three female patients with histologically confirmed breast lesions (567 benign and 266 malignant cases) were selected. In the 5-fold cross-validation, the malignant cohort was randomly divided into 5 subsets to form a training set (80%) and a validation set (20%). FIELD STRENGTH/SEQUENCE: 3.0 T, dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) using T1-weighted high-resolution isotropic volume examination. ASSESSMENT: First, three datasets acquired at different times post-contrast were preprocessed as unlabeled source domains. Second, three baseline networks corresponding to the different MRI post-contrast phases were built, optimized by a combination of mutual information maximization between high- and low-level representations and prior distribution constraints. Next, the pre-trained networks were fine-tuned on the labeled target domain. Finally, prediction results were integrated using weighted voting-based ensemble learning. STATISTICAL TESTS: Mean accuracy, precision, specificity, and area under receiver operating characteristic curve (AUC) were obtained with 5-fold cross-validation. P < 0.05 was considered to be statistically significant. RESULTS: Compared with a convolutional long short-term memory network, pre-trained VGG-16, VGG-19, and DenseNet-121 from ImageNet, M_EL and B_EL exhibited significantly more optimized prediction performance (specificity: 90.5% and 89.9%; accuracy: 82.6% and 81.1%; precision: 91.2% and 90.9%; AUC: 0.836 and 0.823, respectively). DATA CONCLUSION: Transfer learning based on unsupervised pre-training may facilitate automatic prediction of breast cancer molecular subtypes. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY: Stage 2.

Effects of the NF­κB/p53 signaling pathway on intervertebral disc nucleus pulposus degeneration.

The incidence of intervertebral disc degeneration (IDD) is increasing, especially among elderly individuals. The present study aimed to investigate the effects of the NF­κB/p53 signaling pathway on IDD and its regulatory effect on associated cytokines. In the present study, human nucleus pulposus cells were isolated from patients with thoracic­lumbar fractures and patients with IDD to observe cellular morphology and detect phosphorylated (p)­p65/p53 expression levels. The locality and expression levels of p65 in interleukin (IL)­1ß­stimulated nucleus pulposus cells, with or without the addition of ammonium pyrrolidinedithiocarbamate (PDTC; a NF­κB signaling pathway­specific blocker), were measured. Furthermore, the effects of IL­1ß stimulation on the protein and gene expression levels of IDD­related cytokines were determined following p53 knockdown and inhibition of the NF­κB signaling pathway. The results suggested that p­p65 and p53 expression was significantly increased in IDD cells compared with normal nucleus pulposus cells. Moreover, nucleus pulposus cells isolated from patients with IDD contained less cytoplasm compared with normal nucleus pulposus cells, and p65 expression levels were higher in the cytoplasm than the nucleus of IL­1ß­stimulated PDTC­treated healthy nucleus pulposus cells. Moreover, the p53 expression levels were significantly decreased following transfection with sip53. PDTC treatment and p53 knockdown significantly decreased matrix metallopeptidase (MMP)­3, MMP­13, metallopeptidases with thrombospondin type 1 motif (ADAMTS)­4 and ADAMTS­5 expression levels, and increased aggrecan and collagen type II expression levels in IL­1ß­stimulated cells. The present study indicated that activation of the NF­κB/p53 signaling pathway might be related to the occurrence of IDD; therefore, the NF­κB/p53 signaling pathway may serve as a therapeutic target for IDD.

Preoperative Computed Tomography (CT) Evaluation of Anatomical Abnormalities in Endonasal Transsphenoidal Approach in Pituitary Adenoma.

BACKGROUND This study aimed to retrospectively analyze patient clinical data to investigate the effects of computed tomography (CT) reconstruction and the measurement of abnormal structures in the endonasal sphenoidal sinus approach on the operative effects in patients undergoing pituitary adenoma resection. MATERIAL AND METHODS The records of 53 patients who underwent pituitary adenoma resection via the endonasal transsphenoidal approach in the Neurosurgery Department of Tai'an City Central Hospital from December 2010 to June 2016 were analyzed retrospectively. All cases showed anatomical abnormalities in the endonasal transsphenoidal approach that were detected by conventional CT scans. The clinical data of the patients were reviewed. After review, 26 patients who underwent preoperative CT reconstruction and measurement of abnormal structures before surgery were included in the observation group (CT reconstruction group), and 27 patients who did not undergo CT reconstruction and measurement of abnormal structures were included in the control group. Data on intraoperative blood loss, surgical time, hospital stay, and postoperative complications were collected to assess the quality of the surgery. RESULTS Compared with the control group, the observation group showed less blood loss (p<0.001), a shorter operation time (p<0.001), fewer postoperative complications (p<0.001), and a shorter hospital stay (p<0.001). CONCLUSIONS Preoperative CT reconstruction and measurement of abnormal structures in patients undergoing pituitary adenoma resection by the endonasal transsphenoidal approach can improve operative quality and reduce complications.