(64 days)
The Sysmex UF-50 is a fully automated urine cell analyzer intended for in vitro diagnostic use in urinalysis within the clinical laboratory. The UF-50 replaces microscopic review of normal/abnormal specimens and flags specimens containing certain abnormalities which indicate the need for further testing. Laboratorians are responsible for final microscopic review of flagged abnormalities.
The UF-50 is a fully automated urine cell analyzer for urinalysis in clinical laboratories. It analyzes formed elements in urine using flow cytometry technology.
The provided document describes the Sysmex UF-50, a fully automated urine cell analyzer. The document focuses on its substantial equivalence to a predicate device, the Sysmex UF-100.
Here's an analysis of the acceptance criteria and the study that proves the device meets them:
1. A table of acceptance criteria and the reported device performance
The document does not explicitly state numerical acceptance criteria for the UF-50. Instead, the primary "acceptance criterion" is demonstrated by establishing substantial equivalence to the predicate device, the Sysmex UF-100. The reported device performance is that the "correlation results performance correlated to those of the two analyzers, therefore supporting the claim of substantial equivalence."
Since exact numerical criteria are not given, a table would not be directly comparable. However, the qualitative performance is aligned with the predicate device.
2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)
- Sample Size for Test Set: The document does not specify the exact sample size for the correlation studies. It generally refers to "correlation studies."
- Data Provenance: Not specified. It's unclear if the data was retrospective or prospective, or the country of origin.
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)
This information is not explicitly provided. Given that the study focuses on correlating the UF-50 to a predicate device (UF-100) and that the UF-50 flags specimens for "final microscopic review of abnormalities" by laboratorians, it's highly probable that a "ground truth" might have involved microscopic review by qualified laboratorians, but the number and qualifications are not detailed.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set
The document does not describe any specific adjudication method for establishing ground truth, as the primary comparison is between the UF-50 and UF-100.
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
- MRMC Study: No, a multi-reader multi-case comparative effectiveness study was not done. The study compares the UF-50 (an automated analyzer) to another automated analyzer (UF-100), not human readers with and without AI assistance.
- Effect Size: Not applicable, as no such study was performed. The device is intended to flag specimens for human review, not to improve human reader performance directly.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
Yes, a standalone study of the algorithm's performance was done. The "Clinical Performance Data" section describes "Correlation studies were performed to evaluate the equivalency of the UF-50 performance compared to the predicate device, the UF-100." This indicates that the UF-50's performance was evaluated independently, comparing its output directly to the predicate device's output. The device itself is described as a "fully automated urine cell analyzer" performing "in vitro diagnostic use," implying standalone operation to generate results that are then compared.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)
The ground truth for the correlation studies appears to be the results obtained from the predicate device, the Sysmex UF-100. The study "evaluate[d] the equivalency of the UF-50 performance compared to the predicate device, the UF-100." Therefore, the UF-100's performance served as the reference for determining the UF-50's equivalency.
8. The sample size for the training set
The document does not describe a "training set" in the context of machine learning. The UF-50 is described as using "flow cytometry technology," which is a well-established analytical method, not typically associated with machine learning training sets in the same way an AI algorithm might be.
9. How the ground truth for the training set was established
Not applicable, as no machine learning training set is described.
§ 864.5200 Automated cell counter.
(a)
Identification. An automated cell counter is a fully-automated or semi-automated device used to count red blood cells, white blood cells, or blood platelets using a sample of the patient's peripheral blood (blood circulating in one of the body's extremities, such as the arm). These devices may also measure hemoglobin or hematocrit and may also calculate or measure one or more of the red cell indices (the erythrocyte mean corpuscular volume, the mean corpuscular hemoglobin, or the mean corpuscular hemoglobin concentration). These devices may use either an electronic particle counting method or an optical counting method.(b)
Classification. Class II (performance standards).