(60 days)
CorScreen is an active medical device and is intended for measuring a 3/6- or 12-channel surface electrocardiogram (ECG) of a patient. The acquired ECG can be recorded and displayed on a screen or printed on paper. 12-channel ECGs are analysed automatically and suggestions for the interpretation of the 12-channel ECG can be made by the software.
CorScreen can be used for non interpretive applications for patients with an age of 4 years and older and a weight of 20 kg or higher. CorScreen is intended for use in routine ECG recording by trained physicians in the office or hospital. CorScreen is not intended for intracardial use. Automatic interpretation of the ECG is not possible for pediatric patients with an age below 16 years and for pacemaker patients.
The interpretation software is intended to support the physician in evaluating the ECG in terms of morphology and rhythm.
Intended use is equivalent to the intended use of the predicate device AB Cardiette Daedalus View Hes.
A qualified physician has to reassess all CorScreen measurements. An interpretation by CorScreen is only significant if it is considered in connection with other clinical findings. ECG interpretation statements made by the CorScreen represent partial qualitative information on the patient's cardiovascular condition and no therapy or drugs can be administered based solely on the interpretation statements.
CorScreen is an electrocardiograph providing the following characteristics:
- Mains operation .
- Simultaneous acquisition of the 12 standard leads ◆
- Colour LCD display for user interface and ECG visualisation .
- Alphanumerical keyboard .
- Colour ink-printer for printouts of ECG and interpretation reports in US-letter and DIN A4 size .
- Digital filters for base-line drift and mains interference suppression .
- Interpretation program Hanover ECG System (HES) providing the following additional . information:
- Representatives templates of each lead including markers on fiducial points .
- Summary of mean measurements .
- Summary of measurements performed on each lead .
- Rhythm Analysis Statements .
- . Signal noise detection and information
- . Specific findings on QRS complex
- Conduction statements .
- QRS T Diagnostic statements .
- . Arrhythmia monitoring detection
- Heart Rate Variability .
- . Storage of 10 seconds of acquired ECG signal
- . Patient information and ECGs are stored in an internal database
- Data can be stored on an SD memory card
The provided document is a 510(k) premarket notification for the CorScreen electrocardiograph. It focuses on demonstrating substantial equivalence to a predicate device rather than presenting a standalone study with specific acceptance criteria and detailed performance metrics as one might find for a novel AI/ML device.
Here's an analysis based on the provided text, addressing your points:
1. A table of acceptance criteria and the reported device performance
The document does not explicitly state "acceptance criteria" in terms of performance metrics for the interpretive software, nor does it report specific performance outcomes like sensitivity, specificity, or accuracy against a ground truth. Instead, the submission focuses on demonstrating that the CorScreen, including its interpretive software (Hanover ECG System - HES), is substantially equivalent to a legally marketed predicate device (AB CARDIETTE Daedalus View Hes) which also uses HES.
The "performance" is implicitly demonstrated through the comparison table (Section 5.12) which shows that the CorScreen either matches the predicate device's specifications or has differences that are deemed "not relevant for safety and effectiveness."
| Parameter | CorScreen Specification / Performance | Predicate Device Specification (AB CARDIETTE Daedalus View Hes) | Implied Acceptance / Justification |
|---|---|---|---|
| Intended Use | Measuring 3/6- or 12-channel surface ECG, automatic analysis, suggestions for interpretation for patients 4+ years, 20+ kg (non-interpretive), 16+ years (interpretive), not for intracardial use or pacemakers. Interpretation software supports physician in evaluating morphology and rhythm. Qualified physician reassesses all measurements; interpretation is only significant with other clinical findings. Not for sole basis of therapy. | Equivalent to predicate (AB Cardiette Daedalus View Hes), which provides automated ECG analysis to provide diagnostic support to the physician for ECG evaluation on rhythm and morphology. | Deemed substantially equivalent to the predicate device's intended use. The interpretative outputs are clearly stated as supportive and explicitly not for sole diagnosis or treatment. |
| ECG Interpretation Software | Based on Hanover ECG System (HES). Includes representative templates, fiducial points, etc. | Based on Hanover ECG System (HES). Includes representative templates, fiducial points, etc. | Direct Equivalence: Since both devices use the same interpreted software (HES), their interpretation capabilities are considered equivalent. No specific "performance" difference is claimed or tested. |
| Input Dynamic Range | +/- 300mV @ DC | +/-300mV @ DC | Matching specification. |
| Frequency Response Bandwidth | 0.05-150Hz / according to EC11 and IEC 60601-2-51 | 0.05-150Hz | Matching specification. |
| A/D Conversion | 24 bits | 14 bits | Better Performance: 24 bits is superior to 14 bits, indicating better signal resolution, which is considered a positive difference that doesn't raise new questions of safety or effectiveness. |
| Leads | 12 Standard | 12 Standard / 12 Cabrera | Similar capability. |
| Recorder Sensitivity | 5, 10, 20 mm/mV (According to EC11) | 1.25, 2.5, 5, 10, 20, 40 mm/mV | Acceptable Difference: Fewer low sensitivities. Justified as "not relevant for safety and effectiveness" because missing low sensitivities would lead to compressed/difficult-to-read ECG signals. |
| Trace Speeds | 5, 10, 25, 50 mm/s | 1.25, 2.5, 5, 10, 12.5, 25, 50 mm/s | Acceptable Difference: Fewer low speeds. Justified as "not relevant for safety and effectiveness" because missing low speeds lead to highly compressed/difficult-to-read ECG recordings. |
| Display Size | 320 x 240 pixels (Color LCD) | VGA 640 x 480 pixels | Acceptable Difference: Smaller display. Justified as "not relevant for safety and effectiveness" because diagnosis is intended from paper printouts; the display is for controlling the measurement process. |
| Overall Safety and Effectiveness | CorScreen has been tested according to EN 60601-1, IEC 601-2-25, and other standards listed in chapter 9. | Predicate device also tested to EN 60601-1 and IEC 601-2-25. | Compliance: The device meets relevant electrical safety and performance standards for an electrocardiograph, similar to the predicate. |
2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)
The document does not describe a clinical study with a "test set" in the context of evaluating interpretive software performance. There's no mention of a separate dataset of ECGs analyzed by the CorScreen and compared against a ground truth.
The "testing" mentioned in Section 5.13 ("Summary of Device Testing") refers to compliance with electrical safety and performance standards (EN 60601-1, IEC 601-2-25), not clinical performance for the interpretive software itself.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience)
Since no clinical "test set" or performance validation study for the interpretive software is described, there's no mention of experts establishing a ground truth. The justification for the interpretive software's performance relies on it being the same software (Hanover ECG System - HES) as used in the legally marketed predicate device.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set
Not applicable, as no clinical performance study for the interpretive software with a test set is described.
5. If a multi reader multi case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance
No such MRMC study is mentioned. This type of study is more common for AI/ML diagnostic aids where the objective is to show the clinical benefit of the AI "assistance" to human readers. Here, the interpretive software is presented as a component of the ECG device, equivalent to that of the predicate.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
No standalone algorithm performance study for the HES interpretive software is presented in this document. The submission hinges on the fact that the HES software is already commercially available and integrated into an equivalent predicate device. The device is explicitly designed to support physicians, stating "A qualified physician has to reassess all CorScreen measurements" and "An interpretation by CorScreen is only significant if it is considered in connection with other clinical findings." This indicates an "AI-in-the-loop" design where the software provides suggestions rather than making final diagnoses independently.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)
Not discussed, as no specific clinical performance study requiring a ground truth for the interpretive software is detailed.
8. The sample size for the training set
The document does not provide any information about a training set for the Hanover ECG System (HES). This is likely because the HES software itself is a pre-existing, established technology, not a novel algorithm being developed and trained for this particular 510(k) submission.
9. How the ground truth for the training set was established
Not discussed, as no training set information is provided.
In summary:
This 510(k) submission for the CorScreen relies heavily on demonstrating substantial equivalence to a predicate device (AB CARDIETTE Daedalus View Hes). The key argument regarding the interpretive software is that it uses the same underlying software (Hanover ECG System - HES) as the predicate. Therefore, the submission does not present new clinical performance studies or specific acceptance criteria for the interpretive algorithm's accuracy, sensitivity, or specificity. Instead, it focuses on general device specifications and compliance with electrical safety standards, asserting that any differences are not relevant to safety or effectiveness.
§ 870.2340 Electrocardiograph.
(a)
Identification. An electrocardiograph is a device used to process the electrical signal transmitted through two or more electrocardiograph electrodes and to produce a visual display of the electrical signal produced by the heart.(b)
Classification. Class II (performance standards).