(22 days)
By evaluating current and cumulative blood glucose levels, EndoTool IV 1.10 - a glycemic management software support program - is designed for use by healthcare professionals in all patient care settings to recommend intravenous and subcutaneous transition dosing, as well as dextrose, to adjust and maintain the blood glucose level within a configurable clinician- determined target range. The EndoTool Drug Delivery Calculator is indicated for use in adult and pediatric patients ages 2 years and above and who weigh 12 kgs. or more. EndoTool IV 1.10 logic is not a substitute for, but rather an adjunct to clinical reasoning. No medical decision should be based solely on the recommended guidance provided by this software program.
EndoTool IV is a software device for blood glucose management which evaluates current and cumulative blood glucose values to recommend intravenous insulin or dextrose/carbohydrate dose, to adjust and maintain the blood glucose level within a provider-ordered target range. In addition, the application will recommend a subcutaneous Basal transition insulin dose when IV insulin therapy is no longer required.
The provided document is a 510(k) summary for the EndoTool IV 1.10 device, a glycemic management software support program. It states that the device is substantially equivalent to a predicate device (EndoTool Glucose Management System, K132547).
Based on the provided text, the device is a software-only device that calculates blood glucose management recommendations. The 510(k) summary explicitly states that clinical studies and animal studies were not required, and performance testing was limited to software verification and validation testing in accordance with IEC 62304 and FDA software validation guidance. Therefore, much of the requested information regarding acceptance criteria for a clinical study, human reader performance, types of ground truth for training/test sets, and sample sizes for training/test sets, is not available in this document as such studies were not conducted for the clearance of this specific iteration of the device.
Here's what can be extracted from the provided text:
1. A table of acceptance criteria and the reported device performance
The document does not provide a specific table of acceptance criteria tied to quantitative device performance metrics (e.g., accuracy, precision) in the way one might expect for a diagnostic or therapeutic device that undergoes clinical trials. Instead, the acceptance criteria are implicit in the software verification and validation testing objectives: ensuring the software meets its requirements and functions correctly as intended.
| Acceptance Criteria (Implicit) | Reported Device Performance |
|---|---|
| Software meets requirements as defined by IEC 62304 and FDA software validation guidance. | "Software verification and validation testing was conducted per IEC 62304 and included the following:Requirements-based, module and integration testsEndo-Tool regression testing including:Requirements-based testing for risk-related requirementsAutomated algorithm test cases for the complete applicationStatic analysis of the software code" |
| The device performs as well as the predicate device (K132547) and raises no new safety or effectiveness issues. | "The indications for use and performance of the EndoTool IV is substantially equivalent to that of predicate device (K132547), raising no safety or effectiveness issues, and performing as well as 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: Not applicable/not specified. The testing described is software verification and validation, not a clinical study with a patient test set.
- Data Provenance: Not applicable. No clinical data was used for performance testing.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts
- Not applicable. As no clinical test set was used, no human experts were involved in establishing ground truth for performance evaluation of the algorithm against patient data. Software testing relies on defined requirements and expected computational outputs as its "ground truth."
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set
- Not applicable. No clinical test set.
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. The document explicitly states: "Clinical testing was not required to demonstrate the safety and effectiveness of the EndoTool IV." Therefore, no MRMC study or assessment of human reader improvement with AI assistance was conducted or reported here.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
- Yes, in the sense that the software's internal logic and calculations were verified through "Automated algorithm test cases" and "Static analysis of the software code." However, this is against predetermined computational correctness, not against clinical outcomes or human performance. The device is intended to provide "recommended guidance" and is "not a substitute for, but rather an adjunct to clinical reasoning," implying it's always intended to be used with a human in the loop. The "standalone" performance here refers to its computational correctness rather than its diagnostic or therapeutic performance in isolation from clinical context.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)
- For the software verification and validation, the "ground truth" is defined by the software requirements and expected computational outputs based on those requirements. It is a mathematical/logic-based ground truth, not a clinical one derived from patient outcomes or expert consensus on clinical data.
8. The sample size for the training set
- Not applicable/not specified. This device is described as a software program that calculates doses based on input data, not as a machine learning model that undergoes a "training" phase on a dataset of patient outcomes. Its logic is presumably rule-based or uses established physiological models.
9. How the ground truth for the training set was established
- Not applicable. No training set described.
§ 868.1890 Predictive pulmonary-function value calculator.
(a)
Identification. A predictive pulmonary-function value calculator is a device used to calculate normal pulmonary-function values based on empirical equations.(b)
Classification. Class II (performance standards).