Search Results

The InPen System is a home-use reusable pen for single-patient use by people with diabetes under the supervision of an adult caregiver, or by a patient aged 7 and older for the self-injection of a desired dose of insulin. The pen injector is compatible with Lilly Humalog® U-100 3.0 mL cartridges, Novo Nordisk Novolog® U-100 3.0 mL cartridges, and Novo Nordisk Fiasp® U-100 3.0 mL cartridges and single-use detachable and disposable pen needles (not included). The pen Smart Insulin pen allows the user to dial the desired dose from 0.5 to 30 units in one-half (1/2) unit increments.

The InPen dose calculator, a component of the InPen app, is indicated for the management of diabetes by people with diabetes under the supervision of an adult caregiver, or by a patient aged 7 and older for calculating an insulin dose or carbohydrate intake based on user entered data.

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 provide patient-specific fixed doses/ meal sizes to be programmed into the software prior to use.

The InPen App is a software application with versions that are compatible with mobile phones running the iOS or Android operating system. The App is a component of the InPen system and is used with the InPen Smart Insulin pen for the management of insulin-requiring diabetes. The InPen App communicates with the InPen Smart Insulin pen to communicate doses that are delivered by the user. The InPen App is also compatible for use with blood glucose (BG) meters, Medtronic Continuous Glucose Monitors (CGMs), and the Dexcom CGMs. The InPen App includes a dose calculator that can calculate and recommend a dose for the user to review and consider as part of following the treatment plan prescribed by the healthcare provider. The dose calculator features in the App require that a healthcare professional provide patient-specific values for various therapy settings for programming into the App prior to use by the patient. These therapy settings include glucose target(s), duration of insulin action time, insulin sensitivity factor(s), and insulin-to-carbohydrate ratio(s) or fixed insulin doses for meal types and sizes. The dose calculator feature is unavailable to the user until these patient-specific values, provided by the healthcare professional, are programmed and an InPen has been paired to the App. A healthcare provider may also provide long-acting insulin settings to be programmed into the InPen App. The App includes a logbook feature that displays the patient's recent activity related to BG values, meal types and sizes, dose calculations, doses by insulin type (rapid- or long-acting), cartridge replacement and priming. The App also provides reminders and alerts that can notify the user to check their glucose, dose insulin (for potential missed meals, correction doses, and longacting insulin doses) and log doses according to schedule, replace a cartridge, or if the insulin pen has been exposed to very low or very high temperatures. The App can generate a supplemental summary report of recent therapy information for review by the patient or healthcare professional (HCP).

The InPen Cloud includes a therapy report component that the user and the health care provider (HCP) can view and print to assess the overall diabetes control and treatment plan. The report displays data based upon user and HCP-defined inputs, such as glucose, insulin and carb trended information, during the defined period, as well as dose calculator usage and alerts and reminders. The Insulin Notification Service (INS) is a subcomponent of the InPen Cloud that can receive Medtronic CGM sensor glucose measurements from the CareLink Cloud. The INS includes two algorithms that assess "real time" sensor glucose measurements to identify whether a user has missed a dose or if their glucose is rising and a correction dose is needed. If either of these conditions exist, a silent notification is sent by the INS to the InPen App. The InPen App confirms the data and can provide and visual alert to the user. The user can act on the alert by assessing their glucose levels followed by calculating a dose utilizing the InPen App dose calculator.

The provided text is a 510(k) summary for the Medtronic MiniMed InPen System. It describes the device, its indications for use, and a comparison to a predicate device to establish substantial equivalence. However, the document does not contain the specific details about acceptance criteria, the study design (including sample size, data provenance, expert ground truth establishment, or adjudication methods), or performance data in the way requested in the prompt.

The "Performance Data" section (Page 10) broadly states that "Software verification and validation testing was performed in accordance with the FDA's Guidance..." and mentions "unit level testing, integration level testing, and systems testing." It also notes "Cybersecurity Testing" and "risk management activities." Finally, it mentions an "additional assessment of the changes to the InPen Cloud occurred through a summative usability evaluation" where "patients...used the InPen Cloud to perform a series of critical tasks involving the use of the InPen App and Cloud, including the additional alerts and reminder."

Therefore, it is not possible to fill in most of the requested information based on the provided text. The document focuses on regulatory arguments for substantial equivalence rather than detailed clinical or technical study results that would typically include such specifics.

Here's a breakdown of what can be extracted and what cannot:

Information that CANNOT be extracted from the provided text:

1. A table of acceptance criteria and the reported device performance: This detail is not present. The document focuses on showing the device is substantially equivalent to a predicate, not on specific performance metrics against pre-defined acceptance criteria.
2. Sample sizes used for the test set and the data provenance: No specific sample sizes for particular tests (e.g., test sets for algorithms) are mentioned, nor is the data provenance (country, retrospective/prospective). The "summative usability evaluation" is mentioned, but without sample size or details about the data used.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts: Not mentioned.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set: Not mentioned.
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: Not mentioned. This type of study is more common for diagnostic imaging AI, whereas this device is an insulin management system with a dose calculator.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done: While the InPen App has an algorithm for dose calculation and the InPen Cloud has algorithms for notifications, the document doesn't provide performance data for these algorithms in isolation. The "summative usability evaluation" implies human-in-the-loop testing.
7. The type of ground truth used (expert concensus, pathology, outcomes data, etc): Not explicitly stated how the accuracy of the dose calculations or alert effectiveness was verified as "ground truth." The "summative usability evaluation" likely involved checking if users performed tasks correctly, but the underlying ground truth for algorithm accuracy isn't detailed.
8. The sample size for the training set: Not mentioned.
9. How the ground truth for the training set was established: Not mentioned.

Information that CAN be extracted or inferred from the provided text:

• Device Type: Insulin management system (reusable pen, app with dose calculator).
• Regulatory Class: Class II (Product Code: NDC).
• Software Level of Concern: Major.
• Key Software Features: Dose calculator (insulin dose based on carbohydrates or fixed/variable meal sizes), logging, reminders, alerts (missed dose, high glucose).
• New Features (compared to predicate):
  • Two new algorithms in the InPen Cloud (Insulin Notification Service) to identify and send "silent notifications" to the InPen App for missed insulin doses or rising glucose.
  • Two new visual and audible alerts in the InPen App for missed insulin doses or high glucose, for users with a Medtronic CGM.
  • Improved existing Long-Acting Reminder to include both audible and visual alerts.
• Performance Data (General Statement): "Software verification and validation testing was performed in accordance with the FDA's Guidance for Industry and FDA Staff, "Content of Premarket Submissions for Device Software Functions'', and FDA's guidance "General Principles of Software Validation". Verification and validation activities included software testing consisting of unit level testing, integration level testing, and systems testing of the InPen App and InPen Cloud. Cybersecurity Testing was performed... Risk management activities... were undertaken... Additional assessment of the changes to the InPen Cloud occurred through a summative usability evaluation."

In summary, the provided document is a regulatory submission focused on substantial equivalence, not a detailed technical report of study results with specific performance metrics and study methodologies.

Ask a specific question about this device

The InPen is a home-use reusable pen injector for single-patient use by people with diabetes under the supervision of an adult caregiver, or by a patient age 7 and older for the self-injection of a desired dose of insulin. The pen injector is compatible with Lilly Humalog® U-100 3.0 mL cartridges, Novo Nordisk Novolog® U-100 3.0 mL cartridges, and Novo Nordisk Fiasp® U-100 3.0 mL cartridges and single-use detachable and disposable pen needles (not included). The pen injector allows the user to dial the desired dose from 0.5 to 30 units in one-half (1/2) unit increments.

The InPen dose calculator, a component of the InPen app, is indicated for the management of diabetes by people with diabetes under the supervision of an adult caregiver, or by a patient age 7 and older for calculating an insulin dose or carbohydrate intake based on user entered data.

For an insulin dose based on amount of carbohydrates, a healthcare professional must 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 System consists of a manually-controlled pen injector and an app containing a logbook and a dose (bolus) calculator.

The InPen is a manual pen injector containing a non-replaceable battery and electronics to communicate via Bluetooth® with the app on an iOS® mobile device. The intended dose is manually set by the user by rotating a dose knob. The insulin is injected by manually depressing the dose knob which causes the piston in the insulin cartridge to expel the intended dose. The InPen is provided in two different models for compatibility with the available U-100 insulin cartridges, i.e. Humalog®, Novolog®, and Fiasp®. The device is provided with Instructions For Use and a Quick Start Guide. The device is used with sterile needles and U-100 insulin cartridges (supplied separately).

The 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.

Here's a breakdown of the acceptance criteria and study information for the InPen System, based on the provided FDA 510(k) summary:

Important Note: The provided document is a 510(k) summary for a substantial equivalence determination. This means the device is being compared to a previously cleared predicate device (K190487). The performance data presented often focuses on demonstrating that the new device meets the same standards as the predicate and that any changes (like expanded indications for use) do not raise new questions of safety or effectiveness. It does not typically include detailed, de novo clinical trial results for the predicate device.

1. Table of Acceptance Criteria and Reported Device Performance

2. Sample Size Used for the Test Set and Data Provenance

The document does not explicitly state a numerical sample size for a "test set" in the context of an algorithm's performance evaluation.

• Provenance: The primary basis for comparison is a previously cleared predicate device (K190487). The "study" referenced for the current submission primarily relates to human factors evaluation and literature review to support the expanded indications (pediatric use). This suggests the data provenance is a combination of:
  • Previously established performance of the predicate device (likely from earlier submissions).
  • New human factors data (prospective) related to the pediatric use case.
  • Literature review (retrospective).

3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications

This information is not explicitly provided in the summary. Since the evaluation primarily relies on demonstrating substantial equivalence to a predicate and conducting human factors testing for expanded use, it's unlikely a separate "ground truth" establishment by a panel of experts, as might be done for an AI diagnostic device, occurred for this specific 510(k) submission. Human factors may involve usability experts, but their role is different from establishing a diagnostic ground truth.

4. Adjudication Method for the Test Set

This information is not explicitly provided. Given the nature of the device (insulin pen and dose calculator) and the type of evaluation described (human factors, substantial equivalence), an adjudication method for a "test set" in the context of expert review for diagnostic ground truth is not typically applicable or detailed in this type of submission.

5. Multi Reader Multi Case (MRMC) Comparative Effectiveness Study

No, an MRMC comparative effectiveness study is not mentioned or explicitly described. This type of study is more common for diagnostic imaging AI systems where human readers interpret cases with and without AI assistance. The InPen system is an insulin delivery and dose calculation device, not a diagnostic imaging tool.

6. Standalone (Algorithm Only) Performance Study

While the device includes a "dose calculator" component (an algorithm), the document does not describe a standalone "algorithm only" performance study in the way one might for a diagnostic AI. The substantial equivalence argument relies on the overall system (pen injector + app) meeting the same performance standards as the predicate, and the clinical parameters used by the dose calculator are provided by a healthcare professional. Human factors testing focuses on the user interface and interaction, rather than an isolated algorithm's numerical output accuracy.

7. Type of Ground Truth Used

For the InPen pen injector:

• Engineering standards: Adherence to ISO 11608-1 (dose accuracy) and ISO 10993-1 (biocompatibility). These standards define the ground truth for mechanical and material performance.

For the InPen App dose calculator (algorithm):

• Clinical parameters/guidelines: The algorithm's calculations rely on patient-specific parameters (target blood glucose, insulin-to-carbohydrate ratio, insulin sensitivity, fixed/variable meal sizes) programmed by a healthcare professional. The "ground truth" for the dose calculation would be its accurate application of these parameters (as per established medical guidelines for insulin dosing) to user-entered data.
• Human Factors Evaluation: For the expanded pediatric indication, human factors testing serves as the ground truth for user interface and usability safety/effectiveness in the intended population.

8. Sample Size for the Training Set

This information is not provided. The document is concerned with the current device's performance relative to a predicate, not the development or training of a de novo AI algorithm. The "dose calculator" is likely based on established physiological formulas and clinical guidelines rather than machine learning trained on a large dataset in the conventional AI sense.

9. How the Ground Truth for the Training Set Was Established

Since a "training set" in the context of de novo machine learning is not implied or described, the method for establishing ground truth for such a set is not applicable or provided. The dose calculation logic is based on well-established medical algorithms and clinical parameters set by healthcare professionals.

Ask a specific question about this device

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.

Ask a specific question about this device

The InPen is a home-use reusable pen injector for single-patient use by people with diabetes age 12 and older for the self-injection of a desired dose of insulin. The pen injector is compatible with Lilly Humalog® U-100 3.0 mL cartridges, Novo Nordisk Novolog® U-100 3.0 mL cartridges, and Novo Nordisk Fiasp® U-100 3.0 mL cartridges and single-use detachable and disposable pen needles (not included). The pen injector allows the user to dial the desired dose from 0.5 to 30 units in one-half (1/2) unit increments.

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. Prior to use, a healthcare professional must provide the patient-specific target blood glucose, insulin-to-carbohydrate ratio, and insulin sensitivity parameters to be programmed into the software.

The InPen System consists of a manually-controlled pen injector and an app containing a logbook and a dose (bolus) calculator.

The InPen is a manual pen injector containing a non-replaceable battery and electronics to communicate via Bluetooth® with the app on an iOS® mobile device. The intended dose is manually set by the user by rotating a dose knob. The insulin is injected by manually depressing the dose knob which causes the piston in the insulin cartridge to expel the intended dose. The InPen is provided in two different models for compatibility with the available U-100 insulin cartridges, i.e. Humalog®, Novolog®, and Fiasp®. The device is provided with Instructions For Use and a Quick Start Guide. The device is used with sterile needles and U-100 insulin cartridges (supplied separately).

The 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 a 510(k) premarket notification for the InPen System, focusing on its substantial equivalence to a previously cleared predicate device (K160629). It primarily addresses the device's compatibility with an additional insulin cartridge (Fiasp®) and the InPen app's dose calculator.

Based on the provided information, I can extract the following details regarding acceptance criteria and performance study:

1. A table of acceptance criteria and the reported device performance:

The document doesn't explicitly present a table of "acceptance criteria" for the entire InPen System in the typical sense of quantitative targets. Instead, it demonstrates continued conformance to existing standards and principles for both the pen injector and the app.

Here's a summary of the attributes compared and the reported performance, indicating that the new device meets the same standards as the predicate:

2. Sample size used for the test set and the data provenance:

• InPen Pen Injector (Fiasp® compatibility):
  • Sample Size: Not explicitly stated as a number of devices or injections. The compliance is demonstrated by "demonstrating conformance to ISO 11608-1...and Part 2" and "hazard analysis according to ISO 14971". This typically involves a defined number of test units and test conditions as specified by the ISO standards.
  • Data Provenance: Not specified, but generally, ISO standard testing is conducted in controlled laboratory environments. No indication of retrospective or prospective human clinical data.
• Labeling Changes Validation:
  • Sample Size: Not explicitly stated but mentions "human factors testing." Human factors testing involves a representative user sample, but the specific number is not provided.
  • Data Provenance: Not specified. This would typically be prospective data collected during human factors studies.
• InPen Dose Calculator (Fiasp® compatibility):
  • Sample Size: Not applicable. The compatibility was shown through "clinical literature review." No new test set was used for this aspect of the calculator.
  • Data Provenance: Clinical literature (specifically referencing "Mudaliar et al (1999)").

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

• For the InPen Pen Injector and Labeling Changes: Not explicitly stated. For ISO standard compliance, the "ground truth" is typically the standard itself, validated by accredited testing labs or internal quality assurance. For human factors, experts review the results, but their specific number and qualifications are not detailed.
• For the InPen Dose Calculator: No new "ground truth" was established based on expert consensus for the Fiasp® compatibility beyond the existing clinical literature review. The "ground truth" for the original algorithm would have been established during the predicate device's clearance, likely through clinical trials and medical expert validation, but this information is not in the provided text.

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set:

• The document does not describe any adjudication method (like 2+1 or 3+1 refereeing) for establishing ground truth or evaluating performance. The assessment relies on compliance with ISO standards, hazard analysis, human factors testing, and clinical literature review.

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 MRMC comparative effectiveness study was done or is mentioned in the provided text. The InPen System is not an AI-assisted diagnostic tool where "human readers" (e.g., radiologists) would be evaluating cases with or without AI assistance. The dose calculator is a direct calculation tool, not an AI interpretation system.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:

• Yes, for the InPen Pen Injector's compatibility with Fiasp®: Conformance to ISO 11608-1 (Part 1 and Part 2) is a standalone performance test of the device's mechanical and functional aspects, independent of a human user in a clinical setting at the time of testing.
• Yes, for the InPen Dose Calculator's compatibility with Fiasp®: The assessment was based on "clinical literature review," which is an algorithm-only (or calculation-principle-only) evaluation against existing clinical understanding, without direct human-in-the-loop testing for this specific change. The app's calculator algorithm itself performs independently once parameters are set.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):

• InPen Pen Injector: Conformance to ISO 11608-1 standards (which define precision, accuracy, and other physical characteristics of injection systems) and ISO 14971 (risk management). These are engineering and safety standards, representing an accepted "ground truth" for device performance characteristics.
• Labeling Changes: Likely human factors engineering principles and results of user testing, representing the "ground truth" for usability and clarity.
• InPen Dose Calculator: Clinical literature review (specifically mentioning Mudaliar et al (1999) for duration of insulin action). This implies the "ground truth" for the dose calculation algorithm's underlying physiological models and parameters is derived from established medical knowledge and clinical research.

8. The sample size for the training set:

• The document describes a 510(k) submission for a device change (Fiasp® compatibility) and the app remaining unchanged. There is no mention of a "training set" in the context of machine learning or AI algorithm development for the current submission. The InPen dose calculator, as described, performs calculations based on user-entered data and pre-programmed parameters (target BG, I:C ratio, ISF) provided by a healthcare professional, rather than being an adaptive AI model trained on a dataset. The original algorithm for the predicate device would have been developed and validated, but details about its training set are not in this document.

9. How the ground truth for the training set was established:

• As noted above, no "training set" for a machine learning algorithm is discussed in this submission. For the underlying principles of the dose calculator, the "ground truth" for its rules and parameters would have been established through established medical guidelines, clinical trials, and medical expertise during the development of the original algorithm for the predicate device. The current submission relies on the existing algorithm and its previous validation, plus a literature review for Fiasp® compatibility, not a new training process.

Ask a specific question about this device

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. Prior to use, a healthcare professional must provide the patient-specific target blood glucose, insulin-to-carbohydrate ratio, and insulin sensitivity parameters to be programmed into the software.

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.

Here's a breakdown of the acceptance criteria and study information for the InPen Dose Calculator, based on the provided FDA 510(k) summary:

This document primarily focuses on demonstrating substantial equivalence to a predicate device (K160629) rather than presenting a novel clinical study with quantitative performance metrics against specific acceptance criteria. Therefore, the "acceptance criteria" discussed below are inferred from the demonstrated equivalence and risk management, not explicit numerical thresholds.

Acceptance Criteria and Reported Device Performance

Given that this 510(k) is for demonstrating substantial equivalence to a previously cleared device (K160629) and not presenting new clinical performance data with explicit numerical acceptance criteria, the "acceptance criteria" are primarily established by the equivalence of the product's attributes and the successful completion of verification and validation activities. The reported "device performance" is therefore that it functions identically to the predicate device in its calculations and features, and that risks are mitigated.

Study Information

The submission details primarily focus on demonstrating substantial equivalence to a predicate device (InPen System K160629) and robust software verification and validation (V&V), rather than a traditional clinical study with a test set of patient cases.

1. Sample size used for the test set and the data provenance:

   • No explicit "test set" of patient cases with clinical outcomes is mentioned in this summary for demonstrating diagnostic or predictive accuracy. The performance data section refers to "Software Verification and Validation Testing" and "Risk Analysis."
   • It is likely that comprehensive software testing (unit testing, integration testing, system testing) was performed on a variety of input scenarios, but the specific "sample size" of test cases for these software tests is not quantified in this summary.
   • Data provenance is not applicable in the context of a clinical test set from patient data, as no such study is described. The V&V activities would involve internally generated test data.

2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

   • Not applicable as there is no mention of a clinical test set requiring expert ground truth establishment for diagnostic or predictive accuracy. The "ground truth" for the software's calculations would be the mathematically correct output based on the predefined algorithm, parameters, and input data.

3. Adjudication method for the test set:

   • Not applicable, as no clinical test set with expert adjudication is described.

4. 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 MRMC comparative effectiveness study is mentioned. The device is a "dose calculator" with specific instructions for healthcare professionals to program patient-specific parameters. It assists the patient in calculating doses based on these established parameters and user input, rather than augmenting human interpretation of complex medical images or data that would typically feature in an MRMC study.

5. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:

   • The "Software Verification and Validation Testing" would essentially be a standalone evaluation of the algorithm's performance against its specifications, assuming various inputs. The summary states: "Companion Medical has demonstrated the InPen dose calculator is appropriate for its intended use through the use of hazard analysis according ISO 14971. The dose calculator uses the standard approach using healthcare provider specified insulin-to-carbohydrate ratio and insulin sensitivity factors for making calculations. In addition, 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 indicates the algorithm's core functionality was evaluated.

6. The type of ground truth used:

   • For the software's calculation accuracy: The ground truth would be the mathematically correct insulin dose or carbohydrate intake as determined by the predefined algorithms and formulas using the input parameters (target blood glucose, insulin-to-carbohydrate ratio, insulin sensitivity, insulin on-board, and user-entered data). This would be established by independent calculation or a trusted reference implementation of the algorithm.
   • For risk management and safety: Ground truth is implicitly established by adherence to standards like ISO 14971 and relevant FDA guidance documents.

7. The sample size for the training set:

   • Not applicable. This device is a rule-based dose calculator, not a machine learning model that typically requires a "training set" of data to learn from. Its "knowledge" is encoded within the algorithms and patient-specific parameters provided by a healthcare professional.

8. How the ground truth for the training set was established:

   • Not applicable, as there is no training set for this type of device. The algorithm's logic is based on established medical formulas and and a published study for insulin on-board (Mudaliar, et al., 1999).

Ask a specific question about this device

The InPen is a home-use reusable pen injector for single-patient use by people with diabetes age 12 and older for the self-injection of a desired dose of insulin. The pen injector is compatible with Lilly Humalog® U-100 3.0 mL cartridges of insulin and single-use detachable and disposable pen needles (not included). The pen injector allows the user to dial the desired dose from 0.5 to 30 units in one-half (1/2) unit increments.

The InPen is a home-use reusable pen injector for single-patient use by people with diabetes age 12 and older for the self-injection of a desired dose of insulin. The pen injector is compatible with Novo Nordisk Novolog® U-100 3.0 mL cartridges of insulin and single-use detachable and disposable pen needles (not included). The pen injector allows the user to dial the desired dose from 0.5 to 30 units in one-half (1/2) unit increments.

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. Prior to use, a healthcare professional must provide the patient-specific target blood glucose, insulin-to-carbohydrate ratio, and insulin sensitivity parameters to be programmed into the software.

The InPen System consists of a manually-controlled pen injector and an app containing a logbook and a dose (bolus) calculator.

The InPen is a manual pen injector containing a non-replaceable battery and electronics to communicate via Bluetooth® with the app on an iOS® mobile device. The intended dose is manually set by the user by rotating a dose knob. The insulin is injected by manually depressing the dose knob which causes the piston in the insulin cartridge to expel the intended dose. The InPen is provided in two different models for compatibility with the available U-100 insulin cartridges, i.e. Humalog® and Novolog®. The device is provided with Instructions For Use and a Quick Start Guide. The device is used with sterile needles and U-100 insulin cartridges (supplied separately).

The 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.

Here's an analysis of the acceptance criteria and supporting study for the InPen System, based on the provided document:

1. Table of Acceptance Criteria and Reported Device Performance

The document doesn't explicitly present a direct table of "acceptance criteria" alongside "reported device performance" for each specific requirement in the way a formal test report might. However, it states that the device was tested against specific standards and "met the acceptance criteria." I will synthesize this information from the "Performance Data" section.

2. Sample Size Used for the Test Set and Data Provenance

• Test Set Sample Size: The document mentions a "comprehensive Human Factors Summative Study" involving "patients with sufficient diabetes knowledge." However, it does not specify the exact sample size for this study.
• Data Provenance: The document does not explicitly state the country of origin for the data used in the Human Factors Summative Study. It is implied to be a prospective study as it involved patients completing tasks with the InPen System.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications

The document does not provide information on the number or qualifications of experts used to establish ground truth for the clinical (Human Factors Summative) test set. The assessment appears to be based on observed user performance and identification of critical errors, rather than expert-adjudicated diagnostic "ground truth."

4. Adjudication Method for the Test Set

The document does not specify an adjudication method (e.g., 2+1, 3+1, none) for the Human Factors Summative Study. The focus was on identifying critical errors during user interaction, which would likely be observed and documented by study facilitators/observers, rather than requiring expert adjudication of different interpretations.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and Effect Size

No, a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not done. This type of study is typically used for diagnostic devices involving humans interpreting results (e.g., radiologists reading images) and comparing performance with and without AI assistance. The InPen System is a pen injector with a dose calculator, and its evaluation focused on functionality, safety, and human factors, not on improving human reader performance in a diagnostic context.

6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done

Yes, a form of standalone performance was implicitly done for the dose calculator component. The document states:

• "The dose calculator uses the standard approach using healthcare provider specified insulin-to-carbohydrate ratio and insulin sensitivity factors for making calculations."
• "In addition, 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 indicates that the algorithm's calculation logic was assessed against established medical standards and a specific published clinical model, which represents a standalone evaluation of its computational accuracy and adherence to a defined medical algorithm.

7. The Type of Ground Truth Used

• For biocompatibility, EMC, electrical safety, and bench testing, the ground truth was adherence to established national and international standards (ISO, IEC).
• For the software verification and validation, the ground truth was adherence to FDA guidance documents for software in medical devices.
• For the Human Factors Summative Study, the ground truth was the absence of "critical errors that could lead to a hazard" during patient use, as determined by observation during the study.
• For the dose calculator, the ground truth for its logic was based on established medical practice/formulas (HCP-specified parameters) and a published clinical study/model (Mudaliar, et al. 1999 for insulin on-board).

8. The Sample Size for the Training Set

The document does not explicitly mention a training set sample size. The InPen System as described is a hardware device (pen injector) with a software component (app/dose calculator). The "training" for such devices typically refers to the data used during the development of the algorithms (firmware for the pen, logic for the app). This information is generally part of the internal development process and not usually reported as a "sample size" in FDA submission summaries unless a specific machine learning model with a distinct training phase is being described. Given the nature of the device (a calculator using established formulas rather than a predictive AI model trained on patient data), a traditional "training set" for an AI model may not be applicable.

9. How the Ground Truth for the Training Set Was Established

As noted above, a distinct "training set" in the context of an AI/ML model for diagnostic or predictive purposes is not explicitly described. For the algorithmic logic of the dose calculator, the ground truth was established by:

• Medical consensus/standards: The use of "healthcare provider specified insulin-to-carbohydrate ratio, and insulin sensitivity parameters." These are standard, well-accepted medical formulas.
• Published clinical research: The "consideration for insulin on-board based on the published study by Mudaliar, et.al. (1999)." This study's findings served as the basis for incorporating this aspect into the calculator's logic.

Ask a specific question about this device