Search Results
Found 2 results
510(k) Data Aggregation
(117 days)
Mycardium AI Limited
EchoConfidence is Software as a Medical Device (SaMD) that displays images from a Transthoracic Echocardiogram, and assists the user in reviewing the images, making measurements and writing a report.
The intended medical indication is for patients requiring review or analysis of their echocardiographic images acquired for their cardiac anatomy, structure and function. This includes automatic view classification; segmentation of cardiac structures including the left and right ventricle, chamber walls, left and right atria and great vessels; measures of cardiac function; and Doppler assessments.
The intended patient population is both healthy individuals and patients in whom an underlying cardiac disease is known or suspected; the intended patient age range is for adults (>= 22 years old) and adolescent in the age range 18 – 21 years old.
EchoConfidence is Software as a Medical Device (SaMD) that displays images from a Transthoracic Echocardiogram, and assists the user in reviewing the images, making measurements and writing a report.
Here's an analysis of the provided FDA 510(k) clearance letter for EchoConfidence (USA), incorporating all the requested information:
Acceptance Criteria and Device Performance Study for EchoConfidence (USA)
The EchoConfidence (USA) device, a Software as a Medical Device (SaMD) for reviewing, measuring, and reporting on Transthoracic Echocardiogram images, underwent a clinical evaluation to demonstrate its performance against predefined acceptance criteria.
1. Acceptance Criteria and Reported Device Performance
The primary acceptance criteria for EchoConfidence were based on the "mean absolute error" (MAE) of the AI's measurements compared to three human experts. The reported performance details indicate that the device met these criteria.
| Acceptance Criteria Category | Acceptance Criteria | Reported Device Performance |
|---|---|---|
| Primary Criteria (AI vs. Human Expert MAE) | The upper 95% confidence interval of the difference between the MAE of the AI (against 3 human experts) and the MAE of the 3 human experts (against each other) must be less than +25%. | In the majority of cases, the point estimate (of the difference between AI MAE and human expert MAE) was substantially below 0% (indicating the AI agrees with humans more than they agree with each other). The reporting consistently showed that the upper 95% confidence interval was |
Ask a specific question about this device
(123 days)
Mycardium AI Limited
1CMR Pro is software that displays, analyses and transfers DICOM cardiovascular images acquired in Cardiovascular Magnetic Resonance (CMR) scanners, specifically structure, function and flow in the heart and major vessels using multi-slice, multi-parametric and velocity encoded CMR images. It is compatible with 1.5T and 3T CMR acquisitions.
The intended patient population is both known healthy patients in whom an underlying cardiac disease is suspected. The standard viewing tools are indicated for all patients. The Al analysis components are not intended for use in patients with a known congenital cardiac abnormality, children (Age
1CMR Pro is software that displays, analyses and transfers DICOM cardiovascular images acquired in Cardiovascular Magnetic Resonance (CMR) scanners, specifically structure, function and flow in the heart and major vessels using multi-slice, multi-phase, multi-parametric and velocity encoded CMR images. It is compatible with 1.5T and 3T CMR acquisitions.
Here's a breakdown of the acceptance criteria and the study details for the 1CMR Pro device, based on the provided FDA 510(k) summary:
Acceptance Criteria and Reported Device Performance
The document doesn't explicitly state 'acceptance criteria' in a formal table with pass/fail thresholds. Instead, it describes performance benchmarks which the device met or exceeded. The performance for 1CMR Pro was assessed against two primary benchmarks: human 'truthers' and other FDA-cleared software.
| Metric / Assessment Type | Acceptance Criteria (Implied) | Reported Device Performance (1CMR Pro) |
|---|---|---|
| DICE Scores (LV short axis contours) | Expected to be comparable to or superior to human truthers. | Overall DICE scores averaged 0.90, which was superior to Truther average of 0.89. |
| Accuracy (14 variables: volumes, function, LV mass) | Expected to pass all assessments and demonstrate accuracy comparable to or exceeding human truthers. | Passed all assessments. Accuracy exceeded that of the truthers. |
| Precision (LV variables - LVEF, LVmass, LVEDV, LVESV) | Expected to be superior to human clinicians and prior FDA cleared software (lower Coefficient of Variation). | Superior to humans for all measurements. |
| LVEF Coefficient of Variation (CoV) | Lower CoV than Clinician CoV. | 4.3±0.3% vs 7.0±0.6% (Clinician), p5 years experience." (Specific medical specialty e.g., radiologist, cardiologist, or CMR specialist is not stated, but implied to be relevant to cardiac imaging). |
-
Adjudication method for the test set: Not explicitly stated. The document mentions "3 independent US based truthers" indicating multiple expert readings but does not specify how discrepancies were resolved or if a consensus method (e.g., 2+1, 3+1) was used to establish the final ground truth. It seems each truther's reading contributed to a 'Truther DICE score average' and 'Clinician coefficient of variation' rather than establishing a single adjudicated ground truth for direct comparison.
-
If a multi-reader multi-case (MRMC) comparative effectiveness study was done: Yes, a form of MRMC study was implicitly conducted for precision and DICE score comparison.
- Effect size of how much human readers improve with AI vs without AI assistance: The study focused on the AI's standalone performance compared to human readers, rather than human readers with AI assistance. The results show 1CMR Pro's performance (DICE scores, accuracy, precision/CoV) was superior to human truthers/clinicians, suggesting that if AI assistance leads to similar performance as 1CMR Pro standalone, it would represent an improvement over unassisted human performance. For example, for LVEF CoV, the AI achieved 4.3% while clinicians achieved 7.0%.
-
If a standalone (i.e. algorithm only without human-in-the-loop performance) was done: Yes, the entire validation described for accuracy and precision was for the 1CMR Pro algorithm in a standalone capacity ("no human editing" for the comparison with other cleared software). The comparison "Al versus Clinician" also reflects standalone AI performance against human performance.
-
The type of ground truth used (expert consensus, pathology, outcomes data, etc.): The ground truth was established by "3 independent US based truthers" with >5 years experience. This indicates an expert consensus or expert-generated truth, specifically based on their interpretation and measurements of the CMR images. It is not based on pathology or clinical outcomes data.
-
The sample size for the training set: Not provided in the document. The document only discusses "independent validation datasets."
-
How the ground truth for the training set was established: Not provided in the document, as the training set details are omitted.
Ask a specific question about this device
Page 1 of 1