icometrix announced today that the FDA has allowed the clinical use of their icolung AI algorithms for fast and objective quantification of lung pathology on chest CT scans in admitted COVID-19 patients.
The AI algorithm, icolung, quantifies the degree of lung involvement in COVID-19 patients on CT in different lung regions and per type of CT findings (Ground Glass Opacity, Consolidation, and Crazy Paving Pattern). Quantification can help radiologists to assess the pulmonary status, as recommended by the Fleischner Society Consensus Statement. Automated quantification introduces consistency and objectivity in the assessment of regional lung pathology and can assist with the risk-based triage of COVID-19 patients with moderate to severe symptoms.
“AI should complement radiologists, extending beyond detection of lung abnormalities to identifying COVID-suggestive patterns and quantifying lung involvement to support risk assessment,” says Dr. Lawrence Tanenbaum, VP and CTO at RadNet.
“We have been using icolung in clinical practice since its CE-marking and have seen that quantitative CT data is important in the triage of patients. The icolung AI tool can provide a clear percentage of affected tissue, which radiologists can’t,” says Prof. Johan de Mey, Head of Radiology at UZ Brussel. “Of course, we still look at the scans and the radiology expertise is very important, but it’s great that this can go hand in hand with an accurate quantitative evaluation by AI.”
The COVID-19 pandemic puts continuous pressure on the health care system. The potential waves of infection, together with the efforts to resume non-essential imaging and procedures, will prolong the high-pressure situation health care is currently facing. The need for triage and resource-allocation will remain vital throughout the next year.