scholarly journals SCU-Net: A deep learning method for segmentation and quantification of breast arterial calcifications on mammograms

2021 ◽  
Author(s):  
Xiaoyuan Guo ◽  
W Charles O’Neill ◽  
Brianna Vey ◽  
Tianen Christopher Yang ◽  
Thomas J Kim ◽  
...  
Keyword(s):  
Clinical Applications ◽  
Arterial Calcification ◽  
Future Research ◽  
Network Architectures ◽  
Image Size ◽  
Simple Method ◽  
Breast Ct ◽  
Area Metric ◽  
Quantitative Metrics ◽  
Index Evaluation

AbstractPurposeMeasurements of breast arterial calcifications (BAC) can offer a personalized, noninvasive approach to risk-stratify women for cardiovascular disease such as heart attack and stroke. We aim to detect and segment breast arterial calcifications in mammograms accurately and suggest novel measurements to quantify detected BAC for future clinical applications.MethodsTo separate BAC in mammograms, we propose a light-weight fine vessel segmentation method Simple Context U-Net (SCU-Net). Due to the large image size of mammograms, we adopt a patch-based way to train SCU-Net and obtain the final whole-image-size results by stitching patch-wise results together. To further quantify calcifications, we test five quantitative metrics to inspect the progression of BAC for subjects: Sum of Mask Probability Metric (𝒫ℳ), Sum of Mask Area Metric (𝒜ℳ), Sum of Mask Intensity Metric (𝒮ℐℳ), Sum of Mask Area with Threshold Intensity Metric (𝒯𝒜ℳX) and Sum of Mask Intensity with Threshold X Metric (𝒯 𝒮ℐℳX). Finally, we demonstrate the ability of the metrics to longitudinally measure calcifications in a group of 26 subjects and evaluate our quantification metrics compared to calcified voxels and calcium mass on breast CT for 10 subjects.ResultsOur segmentation results are compared with state-of-the-art network architectures based on recall, precision, accuracy, F1-score/Dice Score and Jaccard Index evaluation metrics and achieve corresponding values of 0.789, 0.708, 0.997, 0.729, and 0.581 for whole-image-size results. The quantification results all show >95% correlation between quantification measures on predicted masks of SCU-Net as compared to the groundtruth and measurement of calcification on breast CT. For the calcifications quantification measurement, our calcification volume (voxels) results yield R2-correlation values of 0.834, 0.843, 0.832, 0.798, and 0.800 for the 𝒫ℳ, 𝒜ℳ, 𝒮ℐℳ, 𝒯𝒜ℳ100, 𝒯 𝒮ℐℳ100 metrics, respectively; our calcium mass results yield comparable R2-correlation values of 0.866, 0.873, 0.840, 0.774, and 0.798 for the same metrics.ConclusionsSCU-Net is a simple method to accurately segment arterial calcification retrospectively on routine mammograms. Quantification of the calcifications based on this segmentation in the retrospective cohort study has sufficient sensitivity to detect the normal progression over time and should be useful for future research and clinical applications.

A national survey of telemedicine use by US emergency departments

2018 ◽  
Vol 26 (5) ◽  
pp. 278-284 ◽  
Author(s):  
Kori S Zachrison ◽  
Krislyn M Boggs ◽  
Emily M Hayden ◽  
Janice A Espinola ◽  
Carlos A Camargo
Keyword(s):  
Emergency Department ◽  
Emergency Departments ◽  
Multivariable Analysis ◽  
Clinical Applications ◽  
Future Research ◽  
Emergency Medical Care ◽  
Adults And Children ◽  
Annual Total ◽  
National Emergency ◽  
Total Visit

Objective Telemedicine has the potential to improve the delivery of emergency medical care: however, the extent of its adoption in United States (US) emergency departments is not known. Our objective was to characterise the prevalence of telemedicine use among all US emergency departments, describe clinical applications for which it is most commonly used, and identify emergency department characteristics associated with its use. Methods As part of the National Emergency Department Inventory-USA survey, we queried all 5375 US emergency departments open in 2016. Multivariable logistic regression analyses identified characteristics associated with emergency department receipt of telemedicine services. Results Overall, 4507 emergency departments (84%) responded to our survey, with 4031 responding to both telemedicine questions (75%). Although 1694 emergency departments (42%) reported no telemedicine in 2016, most did: 1923 (48%) emergency departments received telemedicine services, 149 (4%) emergency departments received telemedicine services and were in hospitals that provided telemedicine, and 265 emergency departments (7%) did not receive telemedicine but were in hospitals that provided telemedicine services. Among emergency departments receiving telemedicine, the most common applications were stroke/neurology (76%), psychiatry (38%), and paediatrics (15%). In multivariable analysis, telemedicine-receiving emergency departments had higher annual total visit volume for adults and lower annual total visit volume by children; were less likely to be academic or freestanding; and varied by region. In multivariable analysis, emergency departments in telemedicine-providing hospitals had higher annual total visit volume for adults and children, were more likely to be academic and were less likely to be freestanding. Conclusion In 2016, telemedicine was used in most US emergency departments (58%), especially for stroke/neurology and psychiatry. Future research is needed to understand the value of telemedicine for different clinical applications, and the barriers to its implementation.

scholarly journals “Adherent” versus Other Isolation Strategies for Expanding Purified, Potent, and Activated Human NK Cells for Cancer Immunotherapy

2015 ◽  
Vol 2015 ◽  
pp. 1-14 ◽  
Author(s):  
Senthamil R. Selvan ◽  
John P. Dowling
Keyword(s):  
Nk Cells ◽  
Large Scale ◽  
Mononuclear Cells ◽  
Adoptive Cell Therapy ◽  
Cost Effective ◽  
Future Research ◽  
Peripheral Blood Mononuclear ◽  
Simple Method ◽  
Antitumor Potential ◽  
Healthy Donors

Natural killer (NK) cells have long been hypothesized to play a central role in the development of new immunotherapies to combat a variety of cancers due to their intrinsic ability to lyse tumor cells. For the past several decades, various isolation and expansion methods have been developed to harness the full antitumor potential of NK cells. These protocols have varied greatly between laboratories and several have been optimized for large-scale clinical use despite associated complexity and high cost. Here, we present a simple method of “adherent” enrichment and expansion of NK cells, developed using both healthy donors’ and cancer patients’ peripheral blood mononuclear cells (PBMCs), and compare its effectiveness with various published protocols to highlight the pros and cons of their use in adoptive cell therapy. By building upon the concepts and data presented, future research can be adapted to provide simple, cost-effective, reproducible, and translatable procedures for personalized treatment with NK cells.

A survey of unsupervised learning in medical image registration

2022 ◽  
Vol 2 (1) ◽  
pp. 0-0
Keyword(s):  
Deep Learning ◽  
Image Registration ◽  
Unsupervised Learning ◽  
Medical Image ◽  
Continuous Optimization ◽  
Optimization Method ◽  
Clinical Applications ◽  
Future Research ◽  
Medical Image Registration ◽  
Short Time

Medical image registration has important value in actual clinical applications. From the traditional time-consuming iterative similarity optimization method to the short time-consuming supervised deep learning to today's unsupervised learning, the continuous optimization of the registration strategy makes it more feasible in clinical applications. This survey mainly focuses on unsupervised learning methods and introduces the latest solutions for different registration relationships. The registration for inter-modality is a more challenging topic. The application of unsupervised learning in registration for inter-modality is the focus of this article. In addition, this survey also proposes ideas for future research methods to show directions of the future research.

Brain-Like System for Audiovisual Person Authentication Based on Time-to-First Spike Coding

2013 ◽  
pp. 662-689
Author(s):  
Simei Gomes Wysoski ◽  
Lubica Benuskova ◽  
Nikola K. Kasabov
Keyword(s):  
Neural Network ◽  
Pattern Recognition ◽  
Experimental Research ◽  
Auditory Processing ◽  
Neural Code ◽  
Future Research ◽  
Network Architectures ◽  
Spiking Neural Network ◽  
Coding Schemes ◽  
Spike Coding

The question of the neural code, or how neurons code information about stimuli, is not definitively answered. In addition to experimental research, computational modeling can help in getting closer to solving this problem. In this chapter, spiking neural network architectures for visual, auditory and integrated audiovisual pattern recognition and classification are described. The authors’ spiking neural network uses time to first spike as a code for saliency of input features. The system is trained and evaluated on the person authentication task. The chapter concludes that the time-to-first-spike coding scheme may not be suitable for this difficult task, nor for auditory processing. Other coding schemes and extensions of this spiking neural network are discussed as the topics of the future research.

Medical Robotics

2012 ◽  
pp. 253-287 ◽  
Author(s):  
Ahmad Taher Azar ◽  
M. Sam Eljamel
Keyword(s):  
Surgical Procedures ◽  
Historical Review ◽  
Brain Biopsy ◽  
Medical Robotics ◽  
Medical Application ◽  
Clinical Applications ◽  
Future Research ◽  
Comprehensive Overview ◽  
Medical Specialties ◽  
Interdisciplinary Field

Medical robotics is an interdisciplinary field that focuses on developing electromechanical devices for clinical applications. The goal of this field is to enable new medical techniques by providing new capabilities to the physician or by providing assistance during surgical procedures. Medical robotics is a relatively young field, as the first recorded medical application occurred in 1985 for a brain biopsy. It has tremendous potential for improving the precision and capabilities of physicians when performing surgical procedures, and it is believed that the field will continue to grow as improved systems become available. This chapter offers a comprehensive overview about medical robotics field and its applications. It begins with an introduction to robotics, followed by a historical review of their use in medicine. Clinical applications in several different medical specialties are discusssed. The chapter concludes with a discussion of technology challenges and areas for future research.

Forecasting Stock Returns with Artificial Neural Networks

2011 ◽  
pp. 47-79 ◽  
Author(s):  
Suraphan Thawornwong ◽  
David Enke
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Network ◽  
Artificial Neural Networks ◽  
Stock Returns ◽  
Future Research ◽  
Network Architectures ◽  
Network Applications ◽  
Artificial Neural ◽  
Neural Network Applications

During the last few years there has been growing literature on applications of artificial neural networks to business and financial domains. In fact, a great deal of attention has been placed in the area of stock return forecasting. This is due to the fact that once artificial neural network applications are successful, monetary rewards will be substantial. Many studies have reported promising results in successfully applying various types of artificial neural network architectures for predicting stock returns. This chapter reviews and discusses various neural network research methodologies used in 45 journal articles that attempted to forecast stock returns. Modeling techniques and suggestions from the literature are also compiled and addressed. The results show that artificial neural networks are an emerging and promising computational technology that will continue to be a challenging tool for future research.

Clinical Applications and Investigation of the Acceptance of Mental Illness

2016 ◽  
pp. 146-158
Author(s):  
Lauren Mizock ◽  
Zlatka Russinova
Keyword(s):  
Mental Illness ◽  
Serious Mental Illness ◽  
Culturally Diverse ◽  
Clinical Care ◽  
Clinical Applications ◽  
Barriers And Facilitators ◽  
Future Research ◽  
Potential Development ◽  
Peer Specialists ◽  
Diverse Groups

This chapter reviews the 14 key principles of the process of acceptance of mental illness among culturally diverse groups that emerged from the findings in this book. Each principle is accompanied by clinical recommendations for facilitating the process of acceptance of mental illness. Examples are provided as to how clinicians, peer specialists, and researchers might respond to issues of acceptance of mental illness to facilitate hope and recovery. A number of acceptance-related techniques and theories in clinical care are also discussed. To further understanding and promote the process of acceptance of mental illness among persons in recovery, areas of potential development for future research are reviewed. An “Acceptance of Mental Illness Checklist” with scoring information is provided to assess the dimensions of acceptance and barriers and facilitators among people with serious mental illness and to aid further clinical and research examination of this construct.

A Review of Recent Research on the Manipulation of Response Effort in Applied Behavior Analysis

2020 ◽  
pp. 014544552090850 ◽  
Author(s):  
David A. Wilder ◽  
Hallie M. Ertel ◽  
Daniel J. Cymbal
Keyword(s):  
Behavior Analysis ◽  
Applied Behavior Analysis ◽  
Clinical Applications ◽  
Future Research ◽  
Organizational Studies ◽  
Response Effort ◽  
Independent Variable ◽  
Behavioral Mechanisms ◽  
Desirable Behavior ◽  
Force Pressure

Response effort refers to the distance, force/pressure, or number of discrete behaviors required to engage in a response. In applied behavior analysis, response effort has been used as an independent variable to address a variety of target responses. In this manuscript, we summarize recent clinical and organizational studies in which response effort was manipulated to increase a desirable behavior or decrease a problematic behavior. Recent clinical applications include the manipulation of response effort to decrease self-injurious behavior and pica and increase appropriate eating, compliance, and manding. Recent organizational applications include the manipulation of response effort to increase safety and recycling. We also review the collection of data on treatment integrity, social validity, and maintenance in response effort research and analyze the effectiveness of response effort manipulations. We conclude by discussing the putative behavioral mechanisms responsible for the effects of response effort manipulations and by providing some directions for future research.

Clinical applications and future research horizons

1995 ◽  
Vol 3 (4) ◽  
pp. 270-271
Author(s):  
D.H. Sutherland
Keyword(s):  
Clinical Applications ◽  
Future Research

scholarly journals Establishment of a simpler method for measuring HDL-microRNAs

2018 ◽  
Vol 56 (1) ◽  
pp. 49-55 ◽  
Author(s):  
Hiroaki Ishikawa ◽  
Hiroya Yamada ◽  
Kanako Kondo ◽  
Takeru Ota ◽  
Mirai Yamazaki ◽  
...  
Keyword(s):  
Real Time ◽  
Real Time Pcr ◽  
Phosphotungstic Acid ◽  
Density Lipoprotein ◽  
Clinical Applications ◽  
Processing Capacity ◽  
Quantitative Real Time Pcr ◽  
Simple Method ◽  
Standard Curve ◽  
Purification Methods

Background MicroRNAs are present not only in exosomes but also in high-density lipoprotein (HDL) and have the potential as biomarkers for various diseases. Various purification methods have been developed to quantify HDL-miRNAs; however, they are unsuitable for clinical applications. Therefore, we aimed to establish a simpler analytical method to quantify HDL-miRNAs for clinical applications. Methods We purified HDL fraction from pooled plasma using a three-step protocol consisting of ultracentrifugation, phosphotungstic acid/MgCl2 precipitation and desalting/buffer exchange followed by the quantification of HDL-miRNAs by quantitative real-time PCR. In order to establish a method to quantify HDL-miRNAs by quantitative real-time PCR, we prepared standard curves for miR-223 and miR-92. The HDL-miRNAs of 10 volunteers were assessed. Results Exosomes and LDL were not detected in the purified HDL fraction. Furthermore, we confirmed that only HDL was purified and that the HDL recovery rate of our method was at least approximately 50%. The threshold cycle values of miR-223, miR-92, miR-146a and miR-150 in the same subject were 32.11 ± 0.58, 32.50 ± 0.35, 34.30 ± 0.70 and 34.91 ± 0.77, respectively ( n = 10). The coefficient of variation values for these miRNAs were 1.08–2.21%. In addition, the standard curve for the quantitative analysis of miRNAs showed high linearity (30–30,000 copies/ μL) with a correlation coefficient of >0.99. The concentrations of HDL-miR-223 and HDL-miR-92 in the plasma of 10 subjects were 1.98 ± 0.32 and 0.90 ± 0.14 copies/mL (×104). Conclusions We established a simple method for quantifying HDL-miRNAs and improved the sample processing capacity compared with conventional methods.

Sign in / Sign up

Export Citation Format

Share Document