AI-Based Detection and the COVID-19 Pandemic

Alleviating the burden on health systems

COVID-19, or coronavirus, continues to rapidly spread around the world, and healthcare systems and providers—including radiology teams—may soon become overwhelmed with the volume of patients who will require testing, imaging, and treatment for this disease.

As a healthcare artificial intelligence (AI) software company, RADLogics recognizes the power of AI image analysis for making radiology image interpretation faster and more effective. To alleviate the inevitable burden on health systems, the RADLogics has rapidly applied machine learning image analysis and interpretation to automatically detect, quantify, and track COVID-19 via thoracic CT exams.

Analyzing large numbers of CT studies for COVID-19

RADLogics has rapidly developed its AI-based CT image analysis tools to automatically and accurately detect the COVID-19 / coronavirus in large numbers of CT studies. For patients with COVID-19, the RADLogics solution classifies results per thoracic CT studies utilizing deep-learning image analysis. Our AI-based image analysis further outputs a suggested “corona score” to measure the progression of patients’ disease and/or recovery over time.

Our solution at work

This solution has been studied via multiple international datasets and a range of retrospective experiments to analyze the performance of the system in the detection of suspected COVID-19 thoracic CT images.

Read the Article

Load More

RADLogics, founded in 2010, and based in Boston and Tel-Aviv, is one of the pioneers in using AI & machine learning for medical image analysis for 10x improvement in radiologists’ productivity while also improving diagnostics accuracy. The company was founded by radiologists, big data entrepreneurs, and medical imaging AI experts. The company’s AI medical image analysis platform (US patent 8953858, Japan patent 6585772, other patents pending) enables rapid development of AI algorithms, and provides seamless integration into existing radiology workflow. The company cleared its first FDA 510(k) in 2012.