No abstract

BackgroundQuantizing the Breast Imaging Reporting and Data System (BI-RADS) criteria into different categories with the single ultrasound modality has always been a challenge. To achieve this, we proposed a two-stage grading system to automatically evaluate breast tumors from ultrasound images into five categories based on convolutional neural networks (CNNs).MethodsThis new developed automatic grading system was consisted of two stages, including the tumor identification and the tumor grading. The constructed network for tumor identification, denoted as ROI-CNN, can identify the region contained the tumor from the original breast ultrasound images. The following tumor categorization network, denoted as G-CNN, can generate effective features for differentiating the identified regions of interest (ROIs) into five categories: Category “3”, Category “4A”, Category “4B”, Category “4C”, and Category “5”. Particularly, to promote the predictions identified by the ROI-CNN better tailor to the tumor, refinement procedure based on Level-set was leveraged as a joint between the stage and grading stage.ResultsWe tested the proposed two-stage grading system against 2238 cases with breast tumors in ultrasound images. With the accuracy as an indicator, our automatic computerized evaluation for grading breast tumors exhibited a performance comparable to that of subjective categories determined by physicians. Experimental results show that our two-stage framework can achieve the accuracy of 0.998 on Category “3”, 0.940 on Category “4A”, 0.734 on Category “4B”, 0.922 on Category “4C”, and 0.876 on Category “5”.ConclusionThe proposed scheme can extract effective features from the breast ultrasound images for the final classification of breast tumors by decoupling the identification features and classification features with different CNNs. Besides, the proposed scheme can extend the diagnosing of breast tumors in ultrasound images to five sub-categories according to BI-RADS rather than merely distinguishing the breast tumor malignant from benign.

Objective: This retrospective study aimed to analyze the ultrasound (US) imaging features of solitary papillary thyroid carcinoma (PTC) located in the isthmus and to assess the risk factors for lymph node metastasis (LNM) and tumor capsular invasion. Methods: We included a total of 135 patients with solitary PTC located in the isthmus. All the cases underwent US, total thyroidectomy, and prophylactic central lymph node dissection. Patients' demographic and thyroid isthmus nodules' US characteristics, as well as risk factors associated with LNM and tumor capsular invasion, were analyzed. Results: It was revealed that the occurrence of LNM was higher in male patients than in female patients (P < 0.001). As risk factors, the size of PTC in the isthmus was found to be associated with LNM and tumor capsular invasion (P = 0.005 and 0.000, respectively). The area under the receiver operating characteristic curve (AUC) of the size of the isthmus PTC was 0.64 [95% confidence interval (CI) = 0.55-0.72], indicating a probability for LNM. The AUC value for tumor capsular invasion was 0.77 (95% CI: 0.68-0.83). When the threshold was set to 1.1 cm, the larger size indicated that there was a probability of occurrence of LNM with sensitivity and specificity of 47.4 and 73.7%, respectively. When the threshold was set to 0.7 cm, the larger size indicated that there was potentially a tumor capsular invasion, with sensitivity and specificity of 80.6 and 56.3%, respectively. Wider-than-tall nodules were found to be significantly different from those in LNM and tumor capsular invasion (P = 0.038 and 0.030, respectively). There were significant differences in tumor capsular invasion in extrathyroidal extension (ETE) compared with smooth or ill-defined and lobulated or irregular nodules (P = 0.017). Conclusions: This study showed that the incidence of LNM in male patients was higher than that in female ones. When a US image shows a thyroid isthmus nodule with a wider-than-tall shape, LNM and tumor capsular invasion were likely to occur. When a US image shows a thyroid isthmus nodule with an ETE, tumor capsular invasion was likely to occur. ETE and wider-than-tall may be indicators of FNA under US guidance, even though the size of thyroid isthmus nodule may be <1 cm.

No abstract

BackgroundGlioblastoma multiforme (GBM) is the most common cancer of the central nervous system, while Parkinson’s disease (PD) is a degenerative neurological condition frequently affecting the elderly. Neurotrophic factors are key factors associated with the progression of degenerative neuropathies and gliomas.MethodsThe 2601 neurotrophic factor-related genes (NFRGs) available in the Genecards portal were analyzed and 12 NFRGs with potential roles in the pathogenesis of Parkinson’s disease and the prognosis of GBM were identified. LASSO regression and random forest algorithms were then used to screen the key NFRGs. The correlation of the key NFRGs with immune pathways was verified using GSEA (Gene Set Enrichment Analysis). A prognostic risk scoring system was constructed using LASSO (Least absolute shrinkage and selection operator) and multivariate Cox risk regression based on the expression of the 12 NFRGs in the GBM cohort from The Cancer Genome Atlas (TCGA) database. We also investigated differences in clinical characteristics, mutational landscape, immune cell infiltration, and predicted efficacy of immunotherapy between risk groups. Finally, the accuracy of the model genes was validated using multi-omics mutation analysis, single-cell sequencing, QT-PCR, and HPA.ResultsWe found that 4 NFRGs were more reliable for the diagnosis of Parkinson’s disease through the use of machine learning techniques. These results were validated using two external cohorts. We also identified 7 NFRGs that were highly associated with the prognosis and diagnosis of GBM. Patients in the low-risk group had a greater overall survival (OS) than those in the high-risk group. The nomogram generated based on clinical characteristics and risk scores showed strong prognostic prediction ability. The NFRG signature was an independent prognostic predictor for GBM. The low-risk group was more likely to benefit from immunotherapy based on the degree of immune cell infiltration, expression of immune checkpoints (ICs), and predicted response to immunotherapy. In the end, 2 NFRGs (EN1 and LOXL1) were identified as crucial for the development of Parkinson’s disease and the outcome of GBM.ConclusionsOur study revealed that 4 NFRGs are involved in the progression of PD. The 7-NFRGs risk score model can predict the prognosis of GBM patients and help clinicians to classify the GBM patients into high and low risk groups. EN1, and LOXL1 can be used as therapeutic targets for personalized immunotherapy for patients with PD and GBM.

Integrated sensing and communication (ISAC) is emerging as a key enabler to address the growing spectrum congestion problem and satisfy increasing demands for ubiquitous sensing and communication. By sharing various resources and information, ISAC achieves much higher spectral, energy, hardware, and economic efficiencies. Concurrently, reconfigurable intelligent surface (RIS) technology has been deemed as a promising approach due to its capability of intelligently manipulating the wireless propagation environment in an energy and hardware efficient manner. In this article, we analyze the potential of deploying RIS to improve communication and sensing performance in ISAC systems. We first describe the fundamentals of RIS and its applications in traditional communication and sensing systems, then introduce the principles of ISAC and overview existing explorations on RIS-assisted ISAC, followed by one case study to verify the advantages of deploying RIS in ISAC systems. Finally, open challenges and research directions are discussed to stimulate this line of research and pave the way for practical applications.

No abstract

Abstract. BACKGROUND: Static shear wave elastography (SWE) is used to detect breast lesions, but slice and plane selections result in discrepancies. OBJECTIVE: To evaluate the intraobserver reproducibility of continuous SWE, and whether quantitative elasticities in orthogonal planes perform better in the differential diagnosis of breast lesions. METHOD: One hundred and twenty-two breast lesions scheduled for ultrasound-guided biopsy were recruited. Continuous SWE scans were conducted in orthogonal planes separately. Quantitative elasticities and histopathology results were collected. Reproducibility in the same plane and diagnostic performance in different planes were evaluated. RESULTS: The maximum and mean elasticities of the hardest portion, and standard deviation of whole lesion, had high inter-class correlation coefficients (0.87 to 0.95) and large areas under receiver operation characteristic curve (0.887 to 0.899). Without loss of accuracy, sensitivities had increased in orthogonal planes compared with single plane (from 73.17% up to 82.93% at most). Mean elasticity of whole lesion and lesion-to-parenchyma ratio were significantly less reproducible and less accurate. CONCLUSION: Continuous SWE is highly reproducible for the same observer. The maximum and mean elasticities of the hardest portion and standard deviation of whole lesion are most reliable. Furthermore, the sensitivities of the three parameters are improved in orthogonal planes without loss of accuracies.