CT angiography of the pulmonary arteries of a 58-year-old patient with acute respiratory distress and right-sided chest pain (left). There is a slight contrast sparing in the peripheral branching of the right lower lobe artery, affecting the lateral and dorsal segments. CINA-IPE correctly detects the embolization (right, red box).
Time is Health: Our motto is particularly understandable when it comes to acute pulmonary embolism. Diagnosis with CT angiography of the pulmonary arteries is therefore urgent. The automated early detection of positive findings helps every medical acute care unit that knows what it means to take care of a large number of patients around the clock.
It is with great pleasure that we present CINA-IPE, an AI assistant for the detection of acute pulmonary artery embolism in CT angiographies.
Why CINA-IPE matters and how it works
As a triage tool for emergency radiology, CINA-IPE reports suspected thrombosis of the pulmonary arteries in CT angiographies and enables these patients to be prioritized. The AI assistant provides radiologists and emergency physicians with a clear visual presentation of findings.
Any medical professional can use CINA-IPE for each individual emergency patient by quickly and easily uploading CT angiographies of the pulmonary arteries to Radailogy. In medical institutions, this AI assistant can also do its work automatically in the background in order to fully benefit from its triage potential.
Who benefits
The triage of acute pulmonary embolism is essential for everyone involved, i.e., patients, clinicians and radiologists.
Our own experience at Radailogy
CINA-IPE detects emboli of the central and paracentral pulmonary arteries with excellent certainty. The suspected arteries are labeled in axial MIPs.
We were able to confirm the statistical data given by the developer of sensitivity of 86.6%, specificity of 92.7% and accuracy of 90.0% in our clinical tests. We recognized uncertainties in the detection of peripheral thrombosis. This correlates with the developer’s statement that subsegmental artery emboli are not detected.
The scientific evidence
Grenier PA, Ayobi A, Quenet S, Tassy M, Marx M, Chow DS, Weinberg BD, Chang PD, Chaibi Y. Deep Learning-Based Algorithm for Automatic Detection of Pulmonary Embolism in Chest CT Angiograms. Diagnostics. 2023;13(7), 1324
Schlossman J, Salehi S, Weinberg B, Chow D, Tassy M, Quenet S, Ayobi A, Chaibi Y, Chang P. Validation of a Deep Learning Tool for Automatic Pulmonary Embolism Detection. Am J Respir Crit Care Med. 2023;207:A2607
Data to upload to Radailogy
CT angiographies of the pulmonary arteries of any CT scanner; axial reformations; minimal matrix size 512 x 512; maximal slice thickness 2.5 mm; contrast enhancement of the central pulmonary arteries minimum 100 HU (recommended minimum 130 HU); soft tissue reconstruction kernel