Search

In recent years, deep convolutional neural networks (DCNNs) have shown promising performance in medical image analysis, including breast lesion classification in 2D ultrasound (US) images. Despite the outstanding performance of DCNN solutions, explaining their decisions remains an open investigation. Yet, the explainability of DCNN models has become essential for healthcare systems to accept and trust the models. This paper presents a novel framework for explaining DCNN classification decisions of lesions in ultrasound images using the saliency maps linking the DCNN decisions to known cancer characteristics in the medical domain. The proposed framework consists of three main phases. First, DCNN models for classification in ultrasound images are built. Next, selected methods for visualization are applied to obtain saliency maps on the input images of the DCNN models. In the final phase, the visualization outputs and domain-known cancer characteristics are mapped. The paper then demonstrates the use of the framework for breast lesion classification from ultrasound images. We first follow the transfer learning approach and build two DCNN models. We then analyze the visualization outputs of the trained DCNN models using the EGrad-CAM and Ablation-CAM methods. We map the DCNN model decisions of benign and malignant lesions through the visualization outputs to the characteristics such as echogenicity, calcification, shape, and margin. A retrospective dataset of 1298 US images collected from different hospitals is used to evaluate the effectiveness of the framework. The test results show that these characteristics contribute differently to the benign and malignant lesions' decisions. Our study provides the foundation for other researchers to explain the DCNN classification decisions of other cancer types. [ABSTRACT FROM AUTHOR]

The need for wireless condition monitoring and control of hydraulic systems in an autonomous and battery‐free manner is attracting increasing attention in an effort to provide improved sensing functionality, monitoring of system health, and to avoid catastrophic failures. The potential to harvest energy from hydraulic pressure ripples and noise is particularly attractive since they inherently have a high energy intensity, which is associated with the hydraulic mean pressure and flow rate. This paper presents a comprehensive overview of the state of the art in hydraulic pressure energy harvesting, which includes the fundamentals of pressure ripples in hydraulic systems, the choice of electroactive materials and device structures, and the influence of the fluid–mechanical interface. In addition, novel approaches for improving the harvested energy and potential applications for the technology are discussed, and future research directions are proposed and outlined. [ABSTRACT FROM AUTHOR]

RICHARD C. PASTERNAK GEOFFREY S. GINSBURG For some patients, combining agents is still necessary for optimal benefit ABSTRACT: The primary goal of lipid therapy is to reduce serum levels of [...]

Consider an aggressive strategy for prevention of cardiac events ABSTRACT: In patients with established coronary artery disease (CAD), therapy to lower serum low-density lipoprotein (LDL) levels can halt the progression [...]

Emphysematous gastritis is a rare variant of phlegmonous gastritis due to invasion of stomach wall by gas-forming bacteria. It is characterised by abnormal presence of gas in the stomach by imaging, in association with clinical sepsis. Patients suffering fromthis condition usually present with an underlying pathology. We are reporting a middle-aged Chinesemale with hepatitis B virus related hepatocellular carcinoma. He underwent partial hepatectomy and was diagnosed with emphysematous gastritis 4 days after index operation. Emergency laparotomy, including upper endoscopy, was performed. He was managed with antibiotics and discharged 18 days after second operation. This paper shows a review of the literature about the disease, with particular attention to pathology, clinical features, and management. [ABSTRACT FROM AUTHOR]

Coronary heart disease (CHD) is the major cause of death in the developed world and screening for conventional cardiovascular risk factors fails to identify more than 50% of the individuals who will present with acute coronary syndromes. Chronic inflammation appears to play a significant role in the initiation and development of atherosclerosis. Recent investigations have shown an association between inflammatory markers such as C-reactive protein (CRP) and CHD. These markers have proven useful as prognostic indicators in acute coronary syndromes and in predicting future coronary events in apparently healthy men and women. The availability of high sensitivity CRP (hs-CRP) assays has been crucial in exploring the role of this acute phase reactant in primary prevention settings. In this review, we discuss the evidence associating these inflammatory markers, especially CRP, with the pathogenesis of atherosclerosis and acute coronary syndromes, and we address the mechanism of risk as well as the clinical utility of this marker. [ABSTRACT FROM AUTHOR]

This study examines the performance and clinical use of a commercial immunoseparation assay for low-density lipoprotein (LDL) cholesterol in a sample population of normolipidemic and hyperlipidemic adult volunteers. Using paired fasting and nonfasting samples, we compared the direct LDL assay with the beta quantification method and the Friedewald calculation. Overall, the direct LDL assay correctly classified 82% and 60% of fasting and nonfasting subjects, respectively, into National Cholesterol Education Program risk groups. The Friedewald method correctly classified 84% of subjects. The fasting direct LDL assay has comparable positive and negative predictive values to the Friedewald method, except at an LDL cholesterol of 100 mg/dl. The nonfasting direct LDL assay demonstrates unacceptable positive predictive values when LDL cholesterol decreases to the 130 to 159 and > or = 160 mg/dl categories. Overall, direct LDL assay demonstrates limitations in the nonfasting state and at the LDL cholesterol level of 100 mg/dl used for patients with established coronary heart disease. [ABSTRACT FROM AUTHOR]