Artificial Intelligence in Healthcare: 10 Algorithms Cleared by the FDA

Artificial intelligence is becoming more and more integrated into healthcare by the day. In the past year, there’s been an abundance of new healthcare-focused software all geared towards reducing burden and improving patient results.

Silhouette of human head, brain, in blue electronic lines.

Despite the debate around how the FDA is reviewing this technology, 10 algorithms have been cleared for general use.

Contact (Viz.ai): A clinical decisions support (CDS) tool that analyzes CT results and highlights patients that may have experienced a stroke. Relying on an algorithm to scan CT images for stroke-related indicators, the software then sends a text notification to a neurovascular specialist if a potential large vessel blockage is found. This helps patients receive earlier attention from a specialist, lowering the risk of a potentially fatal event.

IDx-RD (IDx): This AI software system helps to detect diabetic retinopathy in adults with diabetes. Images taken with the Topcon NW4000 retinal camera are uploaded to a cloud server, where they are quickly analyzed and then passed to doctors for results.

Accipio IX (MaxQ AI): Is a workflow tool designed to aid clinicians prioritize adult patients likely presenting with acute intracranial hemorrhage. The algorithm automatically retrieves and processes non-contrast CT images to provide a case-level indicator, which is then used to triage cases that are in most need of expert review and diagnosis.

OsteoDetect (Imagen): Intended for use by primary, emergency, urgent and specialty care practitioners, this AI algorithm scans X-ray images for distal radius fractures—a common one for the wrist bone. After the software has been fed images of adult wrists in various positions, it can highlight regions with potential fracture.

Advisor Pro (DreaMed): This software is a diabetes treatment decision support product that analyzes data from continuous glucose monitors, insulin pumps, and self-monitoring to choose an insulin delivery recommendation. Incorporating factors like basal rate, carbohydrate ration, and a correction factor, dosage recommendations are then delivered directly to the monitoring provider, who can then adjust a patient’s treatment with the click of a button.

RELATED: To Infinity and Beyond: How Artificial Intelligence is Taking Healthcare to the Future

AF Algorithm (AliveCor): This app-based algorithm works with a smartphone ECG device to detect atrial fibrillation. Users who receive a positive result are encouraged to print out and confirm their results with a certified cardiologist.

AppleWatch (Apple): The latest ECG-equipped smartwatch also comes with an algorithm to detect irregular heart rates. It monitors users’ heart rates behind its other functions and alerts them if it notices cause for concern.

EchoMD AutoEF (Bay Labs): Designed to aid cardiologists by automatically review relevant digital video clips collected from an echocardiography study, then rates the quality and selects the best to calculate ejection fraction—a key measure of cardiac function. This algorithm can be integrated into a cardiologist’s routine diagnostic workflow.

Coronary Calcium Scoring (Zebra Medical Vision): Provides a coronary artery calcification score based on a patient’s ECG-gated CT scan. Providers can then use this score to flag patients at high risk of cardiovascular disease sooner, allowing for quicker and more effective care.

Artyrys Oncology AI (Arterys Inc.): This web-based platform helps clinicians analyze ARIs and CT scans for signs of liver and lung cancer. Deep learning algorithms helps speed the process of interpreting the images collected.

As AI’s reach and influence spreads across all corners of healthcare, improved patient care and innovation are to be expected.

5WaysTelemedicineEnhancesPatientCare-Blog