FDA Radiology AI Devices

DentalMonitoring is a medical software device that uses AI to analyze pictures of a patient's mouth taken via smartphone and proprietary hardware. It helps dental professionals remotely monitor orthodontic treatments by assessing oral hygiene, soft tissue, dental conditions, alignment, and occlusion. It provides 2D and 3D monitoring capabilities to track treatment progress safely and non-invasively.

Rho is a machine learning software that analyzes standard frontal x-rays of various body parts in patients aged 50 and older to identify possible low bone mineral density. It helps radiologists by generating reports suggesting patients who may benefit from further clinical bone health assessment, thereby assisting in early detection and management of bone health issues.

DermaSensor is a handheld device combined with a base unit that uses optical spectroscopy and an AI/machine learning algorithm to analyze suspicious skin lesions and assist physicians, especially those not trained as dermatologists, in deciding whether a patient should be referred to a dermatologist. It provides a classification indicating whether to monitor the lesion or investigate further, helping clinicians detect common skin cancers earlier and more accurately.

Fibresolve is a software device that analyzes lung CT images using deep learning to identify patterns suggestive of idiopathic pulmonary fibrosis. It provides a binary output to assist clinicians in diagnosing fibrotic lung disease, serving as an adjunct tool in the diagnostic workflow especially before invasive testing, helping to improve early detection and referral for appropriate clinical discussion.

BrainSee is software that helps clinicians predict whether patients aged 55 to 95 with amnestic mild cognitive impairment (aMCI) will progress to clinical Alzheimer's disease dementia within 5 years. It uses standard brain MRI scans along with patient demographics and cognitive test scores to generate a similarity score indicating the likelihood of progression. This tool provides supplemental prognostic information to aid clinical decision-making but is not a standalone diagnostic device.

Viz HCM is a machine learning-based software designed to analyze 12-lead ECG recordings to detect signs of hypertrophic cardiomyopathy (HCM). It helps clinicians identify patients who may have HCM and need further clinical follow-up, supporting early detection though not replacing full diagnostic evaluation. The system includes an ECG analysis algorithm and a mobile app to present results, improving screening in routine cardiology care.

Caption Interpretation Automated Ejection Fraction Software is an AI-powered tool that processes cardiac ultrasound images to automatically estimate the left ventricular ejection fraction, a critical measure of heart function. It helps clinicians quickly and accurately evaluate cardiac health by analyzing ultrasound video clips, selecting the best quality views, and calculating ejection fraction with confidence metrics. This supports clinical decision-making in adult cardiac evaluations.

The Analytic for Hemodynamic Instability (AHI) is a software device that analyzes ECG signals to monitor patients' hemodynamic status, detecting signs of instability defined by low blood pressure combined with high heart rate. It provides clinicians with frequent, updated alerts to increase vigilance for patients at risk of hemodynamic deterioration, serving as an adjunctive monitoring tool alongside other patient data to improve early detection and patient management in clinical settings.

EyeBOX is a device that uses an eye-tracking camera and software to measure and analyze eye movements as an aid in assessing concussion within one week of head injury for patients aged 5 to 67. It provides clinicians with a score indicating possible brain injury based on abnormal eye movement patterns, helping supplement traditional neurological assessment.

IDx-DR is an AI-based retinal diagnostic software designed to automatically detect more than mild diabetic retinopathy in adults with diabetes using retinal fundus images acquired with a specific fundus camera. It helps clinicians by providing screening results that indicate the presence or absence of disease, enabling timely referral and treatment to prevent vision loss.