The Koios DS was able to achieve overall accuracy that fell within the 95% confidence interval of the radiologists' performance, rendering them statistically equivalent. The number of US Food and Drug Administration (FDA) approvals of proprietary medical algorithms that are powered by artificial intelligence (AI) for image interpretation is “expanding rapidly,” according to … Ground truth was determined by a review of a single radiologist and the radiology report. It makes case-level output available to a PACS/workstation for worklist prioritization or triage. Various laboratory testing was performed. It makes case-level output available to a PACS/workstation for worklist prioritization or triage. SubtleMR is an image processing software that can be used for image enhancement in MRI images. Broncholab provides physicians with reproducible CT values for pulmonary tissue for providing quantitative support for diagnosis and follow-up examination. The amount of US Food and Drug Administration (FDA) approvals of proprietary medical algorithms powered by artificial intelligence (AI) for image … Safety and performance of the device has been evaluated and verified in accordance with software specifications and applicable performance standards. Design validation was performed in simulated use settings. For malignancy risk classification, bench testing was performed to ascertain the degree of concordance with trained interpreting physicians. LesionQuant is intended for automatic labeling, visualization and volumetric quantification of segmentable brain structures and lesions from a set of MR images. Verification, Validation and Performance Testing was completed. Verification and validation including risk mitigations have been executed with results demonstrating Hepatic VCAR met the design inputs and user needs with no unexpected results or risks. A secondary endpoint measure was Briefcase’s potential clinical benefit of worklist prioritization for true positive ICH cases. QbCheck helps with the diagnosis and treatment of ADHD. There were 74 positive cases and 194 negative cases (images with iPE versus without iPE) included in the analysis. Broncholab can be used to support physicians in the diagnosis and documentation of pulmonary tissues images (e.g., abnormalities) from CT thoracic datasets. EchoGo Core is indicated for use in adult populations. ClearRead CT was found to decrease read times with and without outliers. This system was trained using a dataset of over four million images from 9,000 patients, and aims to assist cardiologists with rapid image analysis. Standalone performance testing of MammoScreen assess the performance of MammoScreen-algorithms in the absence of a clinician and includes mammograms of women. The list simplifies research by radiologists and developers by compiling all AI algorithms cleared for use in radiology in a single location. It can be used to reduce image noise for head, spine, neck and knee MRI, or increase image sharpness for non-contrast enhanced head MRI. A total of 19 readers completed the study 111 breast MR images. Non-Interpretive Panel Call for Volunteers, Hackathon Testing ACR's Platform-Model Communication for AI. MammoScreen is intended for use as a concurrent reading aid for interpreting physicians, to help identify findings on screening FFDM acquired with compatible mammography systems and assess their level of suspicion. These flags are viewed by the clinician in Bone Health and Fracture Liaison Service programs in the medical setting via a worklist application on their Picture Archiving and Communication System (PACS). Critical Care Suite identifies images with critical findings to enable case prioritization or triage in the PACS/workstation. The device is intended to assist hospital networks and trained radiologists in workflow triage by flagging and communication of suspected positive findings of pathologies in head CT images, namely Intracranial Hemorrhage (ICH). QuantX is a computer-aided diagnosis (CADx) software device used to assist radiologists in the assessment and characterization of breast abnormalities using MR image data. Vitrea CT Lung Density Analysis assists in analyzing lung densities and volumes. Dustin Diamond, Known As Screech on ‘Saved By The Bell,’ Starts Stage 4 Lung... Black Women Have Higher Rates of Breast Cancer Recurrence and Death, Myeloma Treatment Improvements Are Less Beneficial to High-Risk Patients. To demonstrate the performance of icobrain, endpoints were validated for accuracy and reproducibility. red dot is not intended to direct attention to specific portions basis for clinical decision-making nor is it intended to rule out Pneumothorax or otherwise preclude clinical assessment of X-Ray cases. DM-Density is a software application intended for use with compatible full field digital mammography systems. Sensitivity was observed to be 93.6% and specificity was observed to be 92.3%. The validation data set was truthed by three US Board-Certified Radiologists. Overall cmTraiage was able to demonstrate an AUC of .951 with a 95% CI of 0.937 to 0.964 on this dataset. RSI-MRI+ is indicated for use as automatic post-acquisition image processing software for analysis of diffusion-weighted and anatomical magnetic resonance imaging data. New artificial intelligence-powered technology may soon help reduce delays in diagnosing a collapsed lung, or pneumothorax, which is a serious condition that can be treated with quick intervention. This analysis was performed for GM, WM, CSF, ICV, and WMHs. MaxQ-AI conducted a retrospective study to test the clinical performance of Accipiolx in processing non-contrast head CT cases with a high or low probability of ICH. The goal of this study was primarily to demonstrate that the BriefCase software can identify PE with high accuracy. This technology is the first consumer-available product that allows users to take an ECG from their own wrist, and can provide critical data to physicians. By using this FDA-approved diagnostic tool, radiologists can better diagnose the severity of the arterial defect to improve patient outcomes. Another radiologist was used to break ties between the report and the reviewer. Software Development and Validation & Verification Processes have been implemented to ensure performance according to specifications, User Requirements and Federal Regulations and Guidance document. The sensitivity and specificity of the HealthCXR was reported for two operating points, the first for equal sensitivity and specificity and the second operating point for a higher specificity. To demonstrate the performance of icobrain ctp, the measurements are validated for accuracy and reproducibility. The reader study measured the reduction in time required by the radiologists to localize the tips of tubes, lines, and electrical cardiac wires (TLW) when using ClearRead +Confirm., The study also measured the radiologists' accuracy in localizing the TLW when using ClearRead +Confirm. Software can be used to mark calcified coronary lesions and allocate each lesion to one of several coronary arteries. The device leverages the power of AI algorithms to analyze medical images of eyes to determine if treatment is required. We also suggest a threshold and de ﬁnition of what de nes a medical technology as AI/ML based. CINA uses an artificial intelligence algorithm to analyze images and highlight cases with detected (1) ICH or (2) LVO on a standalone Web application in parallel to the ongoing standard of care image interpretation. They included 29 medical devices and algorithms that met the criteria of being considered an AI/ML-based technology in the FDA's official announcements. OsteoDetect analyzes wrist radiographs using machine learning techniques to identify and highlight distal radius fractures during the review of PA and LAT radiographs of adult wrists. For example, an EF of 50 percent would mean the left ventricle pumps out half of the blood within it each contraction. Convenient summaries for each algorithm include model manufacturer, FDA … The dataset contained a total of 6597 radiographs representing 1149 individuals for which ground truth grading for Kellgren Lawrence grades, osteophyte, sclerosis and joint space narrowing grades, was established by three physicians. It was tested for compliance with the DICOM standard and passed the six DICOM specific test cases provided in section 184.108.40.206 of the Validation Test Report. qER uses an artificial intelligence algorithm to analyze images on a standalone cloud-based application in parallel to the ongoing standard of care image interpretation. Sensitivity and specificity exceeded the 80% performance goal. The reader study measured the AUC of the localization ROV response when using ClearRead CT relative to the unaided read. DeepCT uses an artificial intelligence algorithm to analyze images for findings suggestive of a pre-specified clinical condition and to notify an appropriate medical specialist of these findings in parallel to standard of care image interpretation. The AIMI-Triage CXR PTX Application is a notification-only triage workflow toll for use by hospital networks and clinics to identify and help prioritize chest X-rays acquired in the acute setting for review by hospital radiologists. The standard of care time-to-exam-open is 72.6 minutes. For other details, clicking on the model will take you directly to the FDA summary. FerriSmart is intended to use for measurement of R2 and iron concentration in the liver from MRI scans, The Imbio CT Lung Density Analysis Software provides reproducible CT values for pulmonary tissue, which is essential for providing quantitative support for diagnosis and follow up examinations. The stand-alone detection accuracy was measured on this cohort respective to ground truth. The first operating point showed a sensitivity of 96.74% and specificity of 93.17%. To date, FDA has cleared or approved several AI/ML-based SaMD. Imagen conducted a fully-crossed multiple reader, multiple case (MRMC) retrospective reader study to determine the impact of FX on reader performance in diagnosing fractures. The device is intended to assist hospital networks and trained radiologists in workflow triage by flagging and communication of suspected positive findings of linear lucencies in the cervical spine bone in patterns compatible with fractures. The sensitivity is 93.15% and specificity is 92.99%. Bench testing was performed to ascertain the degree of concordance with skilled physicians. Design verification testing which included confirming all labeling complies with 21CFR801 and all software requirements work as expected. NeuroQuant is intended for automatic labeling, visualization and volumetric quantification of segmentable brain structures and lesions from a set of MR images. It performs quality control checks and generates parametric perfusion maps such as rCBV, CBF, MTT, and TTP and sends the maps to a PACS for subsequent viewing. A secondary endpoint was established to determine time performance of cmTriage to ensure that mammograms can be processed, and notification results returned for use by radiologists within minutes which is clinically acceptable. Approved last summer, the system was the first deep learning technology designed to assist radiologists in assigning urgency to patients’ injuries. CuraRad-ICH analyzes cases using deep learning algorithms to identify suspected ICH findings. Three-D segmentation and isolation of subcompartments, volumetric analysis, density evaluations, low density cluster analysis, fissure evaluation and reporting tools are combined with a dedicated workflow. OsteoDetect-Unaided AUC .840, sensitivity .747, and specificity .889. The reader study design was a concurrent read, cross-over study design with 50 cases with each of 10 regions scored independently. External validation evaluated if the Brain Perfusion with Bayesian algorithm (subject device) was substantially equivalent with the Brain Perfusion with SVD+ algorithm (predicate device). For Brain Volumetry, the test set included 33 3DT1w MR images. For detection accuracy, the AUC of the ROC is .965, sensitivity is 92.1% and specificity is 90.2%. This indicates how well the heart pumps blood and can indicate heart failure. BriefCase is a radiological computer aided triage and notification software indicated foruse in the analysis of CTPA images. Icobrain cross is intended to provide volumes from images acquired at a single time point. This performance was observed for both PA and LAT views across elderly status and post-surgical status, and it was observed on radiographs representative of OsteoDetect’s intended use. Koios DS for Breast software applies proprietary artificial intelligence and machine learning algorithms to analyze user-identified regions of interest (ROIs) in ultrasound images and returns an assessment of risk along with specific lesion characteristics associated with reporting and BI-RADS classification. Read on for a summary of the most recent approvals to see the versatility of AI in the medical space. With BriefCase software the time-to-notification is 4.46 minutes. AliveCor supports the early detection of atrial fibrillation, developed an ECG analytics platform – just as PhysiQ Heart Rhythm Module, Apple, and Cardiologs– and a six-lead smartphone ECG. PowerLook® Tomo Detection V2 Software is a computer-assisted detection and diagnosis (CAD) software device intended to be used concurrently by interpreting physicians while reading digital breast tomosynthesis (DBT) exams from compatible DBT systems. CureMetrix conducted a retrospective, blinded, multi-center study of the cmTriage software. 3. The U.S. Food and Drug Administration has approved a new artificial intelligence (AI) algorithm … All Agatston equivalent scores per study were found identical over all three readings. Each user felt that the Vitrea CT Lung Density Analysis software enables the user to assess and quantify lung density. The device is intended to visualize and quantify MRI data in DICOM format. The triage effectiveness was evaluated by the standalone per-case processing time of the device. The technology leverages machine learning and image analysis with a large database of documented cases to help radiologists analyze MRIs. Ground truth for malignancy risk classification was determined by pathology or 1- year follow-up for cases that were not biopsied. ClearRead CT™ is comprised of computer assisted reading tools designed to aid the radiologist in the detection of pulmonary nodules during review of CT examinations of the chest on an asymptomatic population. Being that some of these hemorrhages can be life-threatening, this is an imperative task that the radiologist is required to fulfill. Ground truth for shape and orientation supplied by three MQSA certified skilled physicians. Results from the standalone study showed that Case-level sensitivity, lesion-level sensitivity, FP Rate in Non-Cancer Cases, and specificity met design specifications. “The FDA has only just begun to develop procedures to evaluate the safety of AI-based medical imaging algorithms. The software development process complies with FDA Guidance documents related to software in Medical Devices. Verification testing was conducted against predetermined acceptance criteria to show that the subject device performs the functions that are equivalent to the predicate device. A primary endpoint goal established to validate that cmTriage operates at a 95% CI for both sensitivity and specificity above the 80% CI reported in the BCSC. Ground truth for each case was determined by three US board certified orthopedic hand surgeons. NeuroQuant performance was evaluated by comparing segmentation accuracy with expert manual segmentations and by measuring segmentation reproducibility between same subject scans. Quantib ND is a non-invasive medical imaging processing application that is intended for automatic labeling, visualization, and volumetric quantification of segmentable brain structures from a set of MR images. Deep01 conducted a retrospective, multicenter, multinational study with the DeepCT software with the primary endpoint to evaluate the software's performance in identifying non-contrast CT head images containing ICH findings in 260 cases from 5 clinical sites. The software is intended to automate the current manual process of identifying, labeling and quantifying the segmentable brain structures identified on MR images. iSchemaView conducted a retrospective, blinded, multicenter, multinational study with the RAPID ICH software with the primary endpoint to evaluate the software's performance in identifying noncontrast CT head images containing ICH findings in 336 cases from US and OUS sites. The final standalone dataset was composed of 888 CXR images from 2 clinical sites from the US and 2 clinical sites from the UK. 2. Nines performed software verification and validation testing that covers the performance of the algorithms, as well as the performance of the software and its components. The company performed standalone clinical performance validation on a dataset of images from large longitudinal US study, Osteoarthritis initiative (OAI) study. So far, only 26 have received FDA approval as of May 2019. Clinical validation was conducted in a MRMC study to validate that the device conformed to the defined user needs and intended uses. HealthVCF uses an artificial intelligence algorithm to analyze chest and abdominal CT scans and flags those that are suggestive of the presence of at least one vertebral compression at the exam level. CINA is a radiological computer aided triage and notification software indicated for use in the analysis of (1) nonenhanced head CT images and (2) CT angiographies of the head. Software validation and bench testing have been conducted. For BI-RADS decriptors, bench testing was performed to ascertain the degree of concordance with trained interpreting physicians. The software provides segmentation and hounsfield numerical analysis values which are indicative of various substances. Imagen performed a MRMC retrospective study to assess performance of OsteoDetect when used with concurrent reading to assist clinicians in detecting distal radius fractures during interpretation of PA and LAT radiographs of adult wrists. The device is intended to assist hospital networks and trained radiologists in workflow triage by flagging and communicating suspected positive findings of (1) head CT images for Intracranial Hemorrhage (ICH) and (2) CT angiographies of the head for large vessel occlusion (LVO). Clinical testing was a retrospective multi reader multi case study meant to compare cancer detection performance of radiologists reading with the aid of MammoScreen to the reading results of the same cohort of radiologists without any decision support. Safety and performance of Arterys Cardio DL has been evaluated and verified in accordance with software specifications and applicable performance standards through software verification and validation testing. Arterys Software v2.0 , Medis Imaging QMass. The study also measured the radiologists’ interpretation time when using ClearRead CT relative to unaided interpretations. ... we need to make the function and application of algorithms transparent, and the FDA provides a useful model. This solution reviews images immediately after the patient is scanned, and informs the radiologists of potentially dangerous cases to aid in prioritizing cases. Rapid LVO is a radiological computer aided triage and notification software indicated for use in the analysis of CTA head images. All performance results were superior to the ones achieved using the predicate device supporting substantial equivalence. NinesAI uses artificial intelligence algorithms to analyze head CT images for findings suggestive of a pre-specified emergent clinical condition. The study objective was to determine the impact on reader performance as defined by the AUC when Koios DS for Breast and an ultrasound examination are combined, compared to ultrasound examination alone in patients that present with a soft tissue breast lesion. The safety and performance of Arterys Oncology DL has been evaluated and verified in accordance with software specifications and applicable performance standards through software verification and validation testing. Analysis of 360 newly tested cases collected from multiple sites across 17 US states. A pivotal reader study was conducted. The user is presented with notifications for cases with suspected ICH or LVO findings. This study was designed to provide evidence for safety and effectiveness of the device. Non-clinical verification and validation test results established that the device meets its design requirements and intended use, that it is as safe and as effective as the predicate devices, and that no new issues of safety and effectiveness were raised. Sensitivity and specificity exceeded the 80% performance goal. The dataset consists of 1255 mammographic studies. The reproducibility was assessed on 150 studies that were read three times. VIDA|vision can be used to support the physician in the diagnosis and documentation of pulmonary tissue images from CT thoracic datasets. BriefCase is a radiological computer aided triage and notification software indicated for use in the analysis of non-enhanced head CT images. The data included a retrospective cohort of 427 anonymized head CT cases from USE and Israel. The device is intended to assist hospital networks and trained medical specialists in workflow triage by flagging the following suspected positive findings of pathologies in head CT images: intracranial hemorrhage, mass effect, midline shift and cranial fracture. Based on the scores provided by the physicians, Vital concluded the Brain Perfusion with Bayesian algorithm is as safe and effective as the already cleared Brain Perfusion with SVD+ algorithm and fulfills its intended use. NinesAI is a parallel workflow tool indicated for use by hospital networks and trained clinicians to identify images of specific patients to a radiologist, independent of standard of care workflow, to aid in prioritizing and performing the radiological review. The Vitrea CT Lung Density Analysis software was designed, developed, and tested according to written procedures that included risk management. However, 14 of those 26 have been approved in 2019 and the list … It can detect atrial fibrillation, a dangerous arrhythmia that can result in stroke if left untreated. The NinesAI device underwent performance testing to verify the efficacy and safety of the machine learning algorithms. The study evaluated the Viz ICH image analysis algorithm in terms of sensitivity and specificity with respect to a ground truth, as established by trained neuro-radiologists, in the detection of ICH in the brain. The device is intended to be used on neonates and infants, aged 0-12 months. It flags those that are suggestive of the presence of at least one suspicious finding at the exam level. The device has been designed, developed, tested, verified and validated according to documented procedures and specific protocols in line with the FDA guidance documents. Specifically, the device uses an artificial intelligence algorithm to analyze images for features suggestive of moderate to large sized pneumothorax; it makes case-level output available to a PACS/workstation for worklist prioritization or triage. Aidoc conducted a retrospective, blinded, multicenter, multinational study with the BriefCase software with the primary endpoint to evaluate the software's performance in identifying CTPAs containing pulmonary embolism in 184 cases from 3 clinical sites. For each scan, 6 slices were selected for comparison. Testing was conducting at the Mammogram-level, Breast-level and Finding-level. The Medical Futurist Magazine The validation set was truthed by two US board certified neuro-radiologists. EchoMD automated ejection fraction software is used to process previously acquired transthoracic cardiac ultrasound images, to store images, and to manipulate and make measurements on images using a personal computer or a compatible DICOM-compliant PACS system in order to provide automated estimation of left ventricular ejection fraction. Performance was consistent for all critical subgroups, such as vendors or reconstruction parameters and patient age. Viz CTP is a standalone software package that is comprised of several modules including DICOM receiving and sending modules, a study processor, image analysis algorithm, as well as software system components including a DICOM storage database and system health-monitoring. Check out more including a list of the approved AI platforms from Dr. Topol and Nature Journal below. Clearview cCAD system was able to achieve overall accuracy that fell within the 95% confidence interval of the radiologist performance, rendering them statistically equivalent. The software outputs parametric maps related to tissue blood flow (perfusion) and tissue blood volume that are written back to the source DICOM. The Quantib™ Brain output consists of segmentations, visualizations and volumetric measurements of grey matter (GM), white matter (WM), and cerebrospinal fluid (CSF). Over 20 years of scientific research from the lab of Maryellen Giger, PhD, went into the creation of this powerful tool. With BriefCase software the time-to-notification is 4.46 minutes. ClearRead Bone Suppression (SoftView) (K092363) , DRX-Revolution (K120062). The system detects soft tissue densities (masses, architectural distortions and asymmetries) and calcifications in the 3D DBT slices. Software verification and validation testing were conducted. The algorithm prediction is compared to the ground truth established by 3 radiologists. Results from internal verification and validation testing performed in accordance with Densitas' design control processes confirm that DM-Density product specifications have been met. Arterys Cardio DL is a web-accessible image post-processing analysis software device used for viewing and quantifying cardiovascular MR images. Convenient summaries for each algorithm include model manufacturer, FDA product code, body area, modality, predicate devices, product testing and evaluation related to product performance, and clinical validation.
list of fda approved ai algorithms 2021