The AI-powered medical imaging company received FDA approval for CINA-iPE and CINA-ASPECTS algorithms, which enable incidental detection of pulmonary embolism and automatic assessment of stroke severity


US FDA approved the CINA-iPE and CINA-ASPECTS solutions. (Credit: National Cancer Institute on Unsplash)

The French medical imaging company Avicenna.AI has received the US Food and Drug Administration (FDA) 510(k) approval for its AI-powered solutions, CINA-iPE and CINA-ASPECTS.

Avicenna.AI has developed the solutions using a combination of deep learning and machine learning technologies, to automatically detect and prioritise life-threatening conditions.

CINA-iPE is an AI-powered solution, designed to detect incidental pulmonary embolisms during routine CT scans, a serious issue causing delays and missed findings in diagnostic imaging.

Its algorithm identifies lung blood clots detected during routine CT scans for different health conditions.

The tool can be used for different scan types, including full-body scans, scans of the chest, abdomen, and pelvis, and scans of the thoracic area along with the abdomen and pelvis.

According to the company, unsuspected pulmonary embolism is a common finding in routine CT scans of the chest, but only 25% of emboli are reported during the initial interpretation.

The CINA-iPE tool has been validated on 381 CT scans, performed for clinical indications other than pulmonary embolism, and acquired on 39 different scanner models from five manufacturers.

Avicenna.AI clinical affairs manager Yasmina Chaibi said: “From day one, we have been committed to validating our AI tools on every type of CT scanner.

“In our validation studies, CINA-iPE achieved excellent sensitivity and specificity, demonstrating its ability to provide effective prioritisation and triage on routine CT scans performed for other clinical indications than pulmonary embolism suspicion.

“The validations and multi-reader-multi-case studies we conducted highlighted that CINA-ASPECTS not only obtained outstanding standalone performance, but also demonstrated that its adjunctive use significantly improved clinicians’ accuracy in the assessment of ASPECTS regions, compared to the conventional use of NCCT images alone.”

CINA-ASPECTS is an AI tool designed for the assessment of stroke severity, which automatically processes non-contrast CT scans and calculates the ASPECT score.

ASPECT score is a topographic scoring system that differentiates the severity of a stroke from a CT scan of the brain.

The CINA-ASPECTS tool computes a heat map indicating the probability of hypodensity and sulcal effacement in the brain.

Also, the system displays a list of infarcted regions and provides CT images corrected from tilt to compare the right and left hemispheres.

In addition, CINA-ASPECTS also helps improve physicians’ reproducibility in ASPECT scoring, which often varies depending on the radiologist reading the scan.

It was validated on 200 scans acquired on 27 scanner models from four leading manufacturers.

Avicenna.AI said that CINA-ASPECTS is its first FDA-approved computer-aided diagnosis (CADx) tool, which goes beyond identifying abnormalities to assess the severity of a condition.

Avicenna.AI regulatory and quality manager Stéphane Berger said: “We take pride in being at the forefront of CADx solutions, being the first AI company supported by thorough clinical validation and a rich repository of real-world clinical data, which ensures compatibility across all manufacturers and platforms.”

Avicenna.AI co-founder and CEO Cyril Di Grandi said: “After a long journey of dedication and perseverance, we are thrilled to announce the FDA clearance of not one, but two of our groundbreaking products.

“The clearance of CINA-iPE and CINA-ASPECTS marks a significant milestone in our mission to strive for excellence in advancing patient care. These achievements stand as a testament to the unwavering commitment of our team.”