Table of Contents
Cross-checked and validated medical devices and algorithms
Cross-checking and validation of all announcements resulted in a database with 64 AI/ML based, FDA-approved medical devices and algorithms. We decided to include only those 29 devices in our further analysis that met the criteria of being considered an AI/ML-based technology in the related official FDA announcements (Table 2). For the other 35 devices, online sources other than the FDA marked them as AI/ML-based technologies. A comprehensive overview is provided in the online open access database.
A short overview of these medical devices and algorithms can be found in the infographic (Fig. 1) and a detailed overview, with subsequent directions to the official FDA announcements, is provided in the online open access database.
Of these medical devices and algorithms, the vast majority (n = 23, 79.3%) was approved by the FDA with a 510(k) clearance, while 5 (17.2%) received de novo pathway clearance and one (3.4%) received PMA clearance.
The first FDA approval was granted in the year 2016, with three approvals at the end of the year 2017. Most FDA approvals were granted in the year 2018, with 13 (44.8%) approvals, while 10 (34.4%) and 2 (6.9%) approvals were granted in 2019 and 2020 up until February, respectively.
The two main medical specialties with AI/ML-based medical innovations are Radiology and Cardiology, with 21 (72.4%) and 4 (13.8%) FDA approved medical devices and algorithms respectively. The remaining medical devices and algorithms can be grouped as focusing on internal medicine/endocrinology, neurology, ophthalmology, emergency medicine, and oncology.
The medical field of radiology is the trendsetter regarding FDA-approved medical devices and algorithms, with the introduction of AI/ML-based solutions for worldwide applied image reading software. Examples are the three algorithms for Arterys Inc., Arterys Cardio DL, Arterys Oncology DL and Arterys MICA, which are connected to the workflow Picture Archiving and Communication Systems from main vendors as Siemens Healthineers AG (Germany) and GE Healthcare (USA)20. Six out of these 21 algorithms can be applied in the field of oncology, with three focusing on mammography analyses (ProFound™ AI Software V2.1, cmTriage and TransparaTM) and three others on CT-based lesion detection (Arterys Oncology DL, Arterys MICA and QuantX). This is followed by two algorithms focusing on brain image analyses, with innovations for stroke and hemorrhage detection (ContaCT, Accipiolx, and icobrain), and six algorithms to improve image processing, with noise and radiation dosage reduction (SubtlePET, Deep Learning Image Reconstruction, Advanced Intelligent Clear-IQ Engine, SubtleMR, AI-Rad Companion (Pulmonary) and AI-Rad Companion (Cardiovascular)). Another four algorithms focusing on acute care, with two algorithms for the assessment of pneumothorax (HealthPNX and Critical Care Suite), one focusing on wrist fracture diagnosis (OsteoDetect) and the Aidoc Medical BriefCase system for triage of head, spine, and chest injuries. The final two algorithms in this specialty can be applied for cardiovascular assessments, focusing on the assessment of the heart ejection fraction (EchoMD AEF Software and EchoGo Core).
Cardiology is another category with major advancements, resulting in four FDA-approved medical devices and algorithms. Most investment goes to innovations for the detection of cardiac rhythm abnormalities, with FDA approval for the AI-ECG Platform and Eko Analysis Software. The other two algorithms overlap with the field of Radiology, being EchoMD AEF software and EchoGo Core.
With diabetes affecting a significant part of society, innovations to manage blood glucose levels were highly warranted. The first steps were made with the introduction of the Guardian Connect System by Medtronic and the DreaMed Diabetes system (DreaMed Diabetes Ltd)21. AI/ML-based interpretation of laboratory results was also introduced for the field of Internal Medicine, with the FerriSmart Analysis System (Resonance Health Analysis Service Pty Ltd) for liver iron concentration assessment.
To increase access to early eye disease detection, one company introduced an AI/ML-based algorithm for the interpretation of ophthalmology tests, being Idx (IDx LLC) for detection of diabetic retinopathy22.
There are also some devices and algorithms related to neurology, with broad overlap with the field of Radiology. In addition to these overlapping algorithms, EnsoSleep was introduced for the diagnosis of sleep disorders.
Additional medical devices and algorithms
For these 35 medical devices, the application of AI/ML has not been confirmed by the official FDA announcements but by other online sources.
With the introduction of the BodyGuardian Remote Monitoring System from Preventice Solutions in 2012, the first FDA-approved AI/ML-based medical device was introduced23. This initiated further investments in innovations for the detection of cardiac rhythm abnormalities, resulting in 14 medical devices and algorithms for this purpose. The other two algorithms in this field focus on the detection of cardiac murmurs (eMurmer ID, CSD Labs GmbH). Interest from multinational technology companies is evident, with two FDA-approved algorithms from Apple Inc, being the ECG App and Apple Irregular Rhythm Notification Feature.
Application of AI/ML-based algorithms for rapid interpretation of the most general values in medical care, being the vital signs, was achieved by Excel Medical Electronics, Spry Health and Current Health. To further assist medical personnel in general, Stratoscientific, Inc. introduced the Steth IO device to analyse heart and lung sounds.
With two AI/ML-based algorithms, BrainScope Company Inc. has introduced AI/ML for the evaluation of brain injuries. At first, this company introduced Ahead 100, an electroencephalograph-based algorithm to evaluate patients after a mild traumatic brain injury. This algorithm was further developed, resulting in the introduction of BrainScope TBI, an algorithm which can be used for a broader scope of traumatic brain injuries—from functional abnormality (concussions) to structural injury (brain bleeds)24. Further attention goes to a gamified neurorehabilitation program, MindMotion GO (MindMaze SA), which is introduced to support rehabilitation for the elderly25. Using motion capture technology and an AI/ML-based algorithm, this invention promotes functional improvements. Other areas of interest are the assessment of memory loss in the eldery (Cantab Mobile, Cambridge Cognition Ltd) and seizure monitoring (Embrace, Empatica Srl.)26,27.
With a high disease burden and a shortage of care providers, the medical field of psychiatry is in need of AI/ML-based support28. Research efforts focus on the diagnosis and stratification of psychological disorders, followed by subsequent treatment support strategies. Two of these AI/ML-based algorithms reached the stage of FDA approval, QbCheck (QbTech AB) and ReSET-O (Pear Therapeutics Inc.). With QbCheck, healthcare workers can substantiate their diagnosis or rule out attention deficit hyperactivity disorder (ADHD), enhancing objective medical decisions in psychiatry29, whereas ReSET-O can be applied for patients with Opioid Use Disorder, providing cognitive behavioral therapy as a mobile medical application for prescription use only. As a next step, the ReSET-O algorithm will be used in a randomized controlled trial, which is scheduled to start this year (April, 2020)30.