KEROS
AI solution for diagnostic assistance for knee MRI
KEROS is an Artificial Intelligence Augmented Radiology solution for knee MRI. It automatically analyzes the main anatomical structures of the knee and detects lesions of ligaments, menisci and cartilages. It is compatible with all types of machines and automatically integrates into the radiologist’s usual reading environments, PACS or review workstations
Detection and characterization of lesions
KEROS detects and characterizes lesions of ligaments, menisci and cartilages.
Automatic generation of a pre-filled report
KEROS generates automatically a pre-filled report, either a detailed report in PDF format, or a synthetic report in “Secondary Capture” format. This report is available on your PACS or review workstation as an additional series.
- MRI
- KNEE
- LIGAMENTS
- MÉNISQUES
- CARTILAGES
- PACS
- AI
Save time, secure your diagnosis and optimize your workflow with Incepto
Testimony
“We have been using KEROS at the clinic for over a year for the interpretation of our knee MRIs. Our experience with the solution is that it reinforces the reliability of the diagnosis made by the radiologist and allows him to go straight to the point in his exploration, by identifying beforehand the lesions potentially present in the analyzed structures. A tool that becomes indispensable, especially for examinations of patients managed in a traumatic context!”
Dr Stephen Binsse,
Radiologist
APHP and Soissons imaging center
Publications
- Meniscal lesion detection and characterization in adult knee MRI: A deep learning model approach with external validation
B.Rizk H.Brat P.Zille R.Guillin C.Pouchy C.Adam R.Ardon G.d’Assignies Physica Medica Volume 83, March 2021, Pages 64-71 - Tran A, Lassalle L, Zille P, Guillin R, Pluot E, Adam C, Charachon M, Brat H, Wallaert M, d’Assignies G, Rizk B. Deep learning to detect anterior cruciate ligament tear on knee MRI: Multi-continental external validation. European Radiology. 2022 Jun 21:1-0.
- Deep learning to detect anterior cruciate ligament tear on knee MRI: first multi-continental external validation (JFR 2021)
- Comparative performance of an artificial intelligence algorithm and human annotators for meniscal crack detection (JFR 2022)
- Performance comparison of a deep learning algorithm and human readers on anterior cruciate ligament tear detection (ESSR 2022)
- Evaluation of Visual Explanation for Knee MRI pathologies classifiers (ECR 2022)
- Impact of AI for pathology detection from knee MRI (RSNA 2022)
- Performance comparison of a deep learning algorithm and human readers on anterior cruciate ligament tear detection (ECR 2023)