(355 days)
The InPen dose calculator, a component of the InPen app, is indicated for the management of diabetes by people with diabetes age 12 and older by calculating an insulin dose or carbohydrate intake based on user entered data. The device is indicated for use with NovoLog® or Humalog® U-100 insulin.
For an insulin dose based on amount of carbohydrates, a healthcare provide patient-specific target blood glucose, insulin-to-carbohydrate ratio, and insulin sensitivity parameters to be programmed into the software prior to use.
For an insulin dose based on fixed/variable meal sizes, a healthcare professional must provide patient-specific fixed doses/ meal sizes to be programmed into the software prior to use.
The InPen app is designed to manage the wireless transfer of insulin dose data from the InPen, log insulin dose data, and provide a dose calculator to aid mealtime insulin dose calculations. The insulin dose calculations provided by the app are meant for patients undergoing multiple daily injection (MDI) therapy. The InPen app is not intended to serve as an accessory to an insulin pump.
The provided text describes the InPen Dose Calculator, outlining its indications for use and
comparing it to a predicate device. However, it does not contain specific acceptance criteria, a
detailed study that proves the device meets those criteria, or the requested specific performance
metrics like sensitivity, specificity, or accuracy.
Here's an attempt to answer the questions based only on the provided text, highlighting
where information is missing:
1. A table of acceptance criteria and the reported device performance
The document does not explicitly state numerical acceptance criteria or present a table of device
performance against such criteria. It generally states that the device "satisfies
all functional performance and safety requirements, meets its intended use, and is safe for the
intended user population."
2. Sample size used for the test set and the data provenance (e.g. country of origin of the data,
retrospective or prospective)
The text mentions a "summative evaluation" where "patients with sufficient diabetes knowledge
completed self-training and then completed a series of critical tasks." However, it does not
specify the sample size of this test set, the country of origin of the data, or whether it was
retrospective or prospective.
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 provided in the document. The "ground truth" for the dose calculations
would theoretically be the correct insulin dose based on provided parameters, but the process of
establishing this for the test set is not detailed.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set
The document does not describe any adjudication method for the 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
The document does not mention a Multi-Reader Multi-Case (MRMC) comparative effectiveness study.
The InPen Dose Calculator is an algorithm to calculate insulin doses, not designed for human
"readers" to interpret medical images or data. Therefore, the concept of human readers improving
with AI assistance in this context does not apply in the manner typically associated with MRMC
studies.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
The "summative evaluation" involved patients using the device, but the core function is an algorithm
calculating doses based on user input. The document states that the "dose calculator uses the
standard approach using healthcare provider specified insulin-to-carbohydrate ratio and insulin
sensitivity factors for making calculations." This implies a standalone algorithmic function based
on pre-programmed parameters. The "Clinical Evidence" section focuses on usability and safety with
human users, rather than solely on the algorithm's performance in isolation.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)
For the dose calculations, the "ground truth" is implied to be the standard approach using
healthcare provider specified insulin-to-carbohydrate ratio and insulin sensitivity factors for
making calculations. The text also mentions: "the calculator includes a consideration for insulin
on-board based on the published study by Mudaliar et al (1999) for the duration of insulin action."
This suggests calculations are validated against established medical formulas and literature.
8. The sample size for the training set
The document does not provide any information about a training set or its sample size. The device
appears to be a rule-based calculator rather than a machine learning model that would typically
require a training set.
9. How the ground truth for the training set was established
As no training set is mentioned or implied for this rule-based dose calculator, this question is not
applicable based on the provided text.
§ 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).