Researchers from the Massachusetts Institute of Technology, Cambridge, have developed an artificial intelligence (AI) tool that can detect COVID-19 by listening to patients' coughs, regardless if they are symptomatic or not.

To build the tool, the researchers solicited audio recordings of patients coughing and accompanying information about their conditions via an online website. They collected a dataset of more than 70,000 recordings containing an average of three coughs per patient and an estimated 2,600 patients with a positive case, to date.

Using the COVID-19 cough recordings and an equal number of COVID-19 negative recordings randomly selected from the dataset of 5,320, the researchers developed, trained, and validated a model that listens for specific acoustic biomarkers related to muscular degradation, vocal cord changes, sentiment or mood changes, and changes in the lungs or respiratory tract.

The tool discriminated COVID-19 positive patients with 97.1% accuracy, 98.5% sensitivity, and 94.2% specificity. The model performed at 100% accuracy when detecting coughs from asymptomatic positive cases.

This tool can provide a free, non-invasive, real-time large-scale COVID-10 asymptomatic screening test to augment current approaches, the researchers say. Practical uses include daily screening of students, workers, and the public, or for pool testing to quickly alert groups to outbreaks.

The post AI tool detects COVID-19 by listening to patients’ coughs appeared first on IIOT Connection.

Researchers from the University of Minnesota and M Health Fairview are using artificial intelligence (AI) to detect COVID-19 in chest x-rays of hospitalized patients.

The researchers used 18,000 x-rays from COVID-19 patients and 100,000 x-rays from patients without the disease to develop and train an AI program to discover COVID-19-related patterns in the images.

When patients arrive in the emergency department with suspected COVID-19, it is standard protocol to take a chest x-ray. This program can help physicians learn if a patient has COVID-19 much quicker than traditional testing.

The researchers now plan to tweak the AI program and use it to predict the progression of the disease in different patients.

The post Researchers use AI to diagnose COVID-19 in chest x-rays appeared first on IIOT Connection.

Leandro Pecchia, PhD, associate professor of biomedical engineering at the University of Warwick, Coventry, UK, has been awarded £13 million (nearly $17 million) for the ODIN project.

The project will explore the use of robots and artificial intelligence (AI) to ease pressures on hospitals during recovery from COVID-19 and future disasters.

Dr Pecchia has identified 11 hospital challenges the project will take on, using robotics, Internet of Things (IoT), and artificial intelligence (AI).

The interventions will be piloted in six top hospitals in Europe and will span from clinical to logistic procedures, including patient management, medical device and PPE management, disaster preparedness, and hospital resiliency.

Dr Pecchia and his team also will work with three medical device manufacturers–Samsung, Philips, and Medtronic–and seven other enterprises.

The post AI, robots to transform hospitals’ resilience to COVID-19, future disasters appeared first on IIOT Connection.

The National Institutes of Health (NIH) on August 5 announced the launch of the Medical Imaging and Data Resource Center, which will create new tools using artificial intelligence (AI) and medical imaging to diagnose and personalize therapies for COVID-19 patients.

The Center will build a large repository of COVID-19 chest images so that researchers can evaluate lung and cardiac tissue data, ask critical research questions, and develop predictive COVID-19 imaging signatures.

A major challenge is to rapidly and accurately identify these signatures and evaluate this information in combination with other clinical symptoms and tests.

The goals of the Center are to lead the development and implementation of new diagnostics, including machine learning algorithms, that will allow rapid and accurate assessment of the patient's COVID-19 disease status and help physicians optimize treatment.

The post NIH harnesses power of AI for COVID-19 diagnosis, treatment appeared first on IIOT Connection.

The use of an automated web-based symptom monitoring tool as part of Maine’s COVID-19 contact tracing program made the state’s contact tracing and monitoring faster and more efficient.

The tool was well received, with the majority of monitored contacts (96.4%) agreeing to the automated system. Contacts automatically received daily symptom questionnaires via their choice of e-mailed weblink, test message, texted weblink, or telephone call until they completed their quarantine.

The system promptly identified COVID-19 in the monitored contacts, with 190 (11.7%) of 1,622 developing COVID-19.

Contacts who refused automated monitoring or could not be enrolled because of language barriers were monitored using direct monitoring by the contact tracing team.

Automated monitoring tools can augment traditional contact tracing, but they can't take the place of a large, trained public health workforce needed for a COVID-19 response, the authors say.

The post Automated monitoring tool facilitates contact tracing of COVID-19 patients appeared first on IIOT Connection.

The Curial AI test, an artificial intelligence (AI) program developed at the University of Oxford, UK, can identify COVID-19 within 1 hour of a patient arriving at an emergency department (ED), the July 31 Belfast Telegraph reports.

The AI test analyzes patient data, such as blood tests and vital signs gathered in the ED, to determine the chance of a patient testing positive for COVID-19.

The AI program has been tested on data from 115,000 ED patient visits and has a more than 90% accuracy rate.

The post AI program can detect COVID-19 in 1 hour appeared first on IIOT Connection.

Researchers at Binghamton University, State University of New York, are using data collected from around the world by Johns Hopkins University to build prediction models that take advantage of artificial intelligence (AI) to track COVID-19 and predict where it might surge next.

The models examine trends and patterns from 50 countries where COVID-19 infections are the highest and can predict within a 10% margin of error what will happen for the next 3 days based on the data for the past 14 days.

The main use of this study is to help hospitals and healthcare workers prepare for a surge with proper equipment and hospital beds, the researchers say.

The post Researchers use AI to build COVID-19 predictions appeared first on IIOT Connection.

Researchers from Florida Atlantic University's College of Engineering and Computer Science and Schmidt College of Medicine (all in Boca Raton) announced on June 24 that they have received a 1-year, $90,000 National Science Foundation RAPID project grant to design a COVID-19 knowledge base and risk assessment tool powered by artificial intelligence (AI).

The research will use social networks and AI, facilitated by molecular genetics and viral infection for COVID-19 modeling.

The researchers note that as the fight against COVID-19 continues, a wealth of information is being produced. As a result, differences in outbreaks with respect to diverse geographies, regional policies, and cultural groups is raising confusion, contradictions, and inconsistencies in disease outbreak modeling.

This research project, "RAPID: COVID-19 Coronavirus Testbed and Knowledge Base Construction and Personalized Risk Evaluation," will address these discrepancies and help understand the correlations and roles different factors play in predicting the spread of the virus. It also will provide individual information for risk evaluation.

The post Researchers design COVID-19 knowledge base, risk assessment tool powered by AI appeared first on IIOT Connection.

The Cleveland Clinic announced June 15 that its researchers have developed the world's first model to predict the likelihood of patients testing positive for COVID-19 and their outcomes from the disease.

The model was based on data from nearly 12,000 patients in the Cleveland Clinic's COVID-19 Registry, including those who tested positive and negative. It also isolated 400 variables and included the data in statistical algorithms to develop the risk calculator.

The model has been shown to be 85% accurate. It takes only minutes to complete the questions in the risk calculator, and the percentage risk of developing COVID-19 is immediately available.

Among the findings are that patients are less likely to test positive for COVID-19 if they:

• received the pneumococcal polysaccharide vaccine and the flu vaccine
• are taking melatonin (sleep aid), carvedilol (high blood pressure and heart failure medication), or paroxetine (anti-depressent)
• are of Asian descent.

This nomogram will bring precision medicine to the COVID-19 pandemic, enabling researchers and physicians to predict a patient's risk of testing positive, the researchers say.

The post Cleveland Clinic develops COVID-19 risk calculator appeared first on IIOT Connection.

In this study, researchers at the Massachusetts General Hospital, Boston; King's College, London; and health science company Zoe Global, Accra, Ghana; developed an artificial intelligence (AI) diagnostic that can predict whether someone is likely to have COVID-19 based on their symptoms.

The researchers analyzed data from 2,618,862 people (2,450,569 UK, 168,293 US) who had regularly logged their health status into a smartphone-based app, around a third of whom had logged symptoms linked to COVID-19. An AI mathematical model was applied to the data, and it predicted that 140,312 (17.42%) were likely to have COVID-19.

The AI model predicted with nearly 80% accuracy whether an individual was likely to have COVID-19 based on their age, gender, and a combination of four key symptoms:

• loss of smell or taste
• severe or persistent cough
• fatigue
• skipping meals.

The researchers suggest that combining the AI prediction model with widespread adoption of the app could help identify those who are likely to be infectious as soon as symptoms start to appear.

The post AI diagnostic predicts COVID-19 without testing appeared first on IIOT Connection.