A deep learning image prediction model, developed using Indian data combined with datasets from other countries, achieves higher accuracy compared to conventional risk scores.
Google Research & Apollo Hospital Collaboration Initiative
Feedback Loop from the prospective use in patients
Validated using various National & International Data sets
QMS: ISO 13485:2016 Certified, MDSAP Certified

Interpretation & Adoption Message

AI Algorithm + Clinicians: The chest X-ray anomaly detection AI model has been built as an adjunct tool for physicians to detect the chest radiograph as "normal" or "abnormal." When simulating the Model, the system reduced the turnaround time for abnormal cases by up to 28%.
Risk Identification and Prevention: This AI model for chest X-ray anomaly detection enhances the availability, accuracy, and consistency of chest X-ray interpretation, aiding in case prioritization and rapid triaging of anomalies while streamlining clinical workflows. However, it is intended to complement clinical judgment and should not be relied upon as the sole method for detecting anomalies in chest radiographs.
Where to use: The Chest X-ray Anomaly Detection AI tool can be utilized by physicians across various scenarios requiring quick and accurate chest X-ray evaluations. These include preventive health screenings, emergency rooms for case prioritization and rapid triaging, and settings with limited access to trained radiologists. Additionally, it facilitates the efficient review process by enabling the batching of negative chest X-rays for streamlined analysis.

Apollo Hospitals CXR Anomaly Detection AIApollo Hospitals Enterprise Ltd

Apollo Hospitals CXR Anomaly Detection AI