Search Results

The SmartPilot YpsoMate NS-A2.25 is indicated for use with the compatible disposable autoinjector to capture and record injection information that provides feedback to the user.

The SmartPilot YpsoMate NS-A2.25 is an optional, battery operated, reusable device designed to be used together with a compatible autoinjector (a single use, needle based, pre-filled injection device for delivery of a drug or biologic into subcutaneous tissue). Figure 1 shows the SmartPilot YpsoMate NS-A2.25 with the paired autoinjector. The SmartPilot YpsoMate NS-A2.25 records device data, injection data and injection process status. The SmartPilot YpsoMate NS-A2.25 also provides guidance feedback to the user during the injection.

Note that the SmartPilot YpsoMate NS-A2.25 does not interfere with autoinjector function.

The provided 510(k) clearance letter details the substantial equivalence of the SmartPilot YpsoMate NS-A2.25 device to its predicate. While it lists various performance tests and standards met, it does not contain specific acceptance criteria values or detailed study results for metrics like sensitivity, specificity, or improvement in human reader performance. This document primarily focuses on demonstrating that the new device does not raise new questions of safety and effectiveness compared to the predicate, due to similar technological characteristics and adherence to relevant safety standards.

Therefore, many of the requested details about acceptance criteria, study design (sample size, data provenance, expert adjudication, MRMC studies), and ground truth establishment (especially for AI-driven performance) cannot be extracted directly from this regulatory document. The information primarily pertains to hardware, software, and usability testing.

However, based on the provided text, here's what can be inferred or stated about the device's acceptance criteria and proven performance:

Device: SmartPilot YpsoMate NS-A2.25

Indication for Use: The SmartPilot YpsoMate NS-A2.25 is indicated for use with the compatible disposable autoinjector to capture and record injection information that provides feedback to the user. Specifically compatible with Novartis/Sandoz Secukinumab (Cosentyx).

1. Table of Acceptance Criteria and Reported Device Performance

The document does not provide a table with quantitative acceptance criteria and reported performance values for metrics typically associated with AI/software performance (e.g., sensitivity, specificity, accuracy of data capture in a clinical context). Instead, it focuses on meeting established engineering, safety, and quality standards.

Here's a summary of the types of performance criteria implied by the successful completion of the listed tests:

Note: The device's primary function is to "capture and record injection information that provides feedback to the user," and it "does not capture dosing information" or "electronically controlled dosing." Therefore, criteria related to dosing volume accuracy or AI interpretation of medical images/signals for diagnosis are not applicable to this device. The focus is on the accurate capture of event data (injection start/end, result) and providing timely feedback, as well as general device safety and functionality.

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

The document describes various types of tests (e.g., Biocompatibility, EMC, Software V&V, Mechanical, Lifetime, Human Factors), but does not specify the sample sizes used for each test dataset.

Data Provenance: The document does not explicitly state the country of origin for the data or whether the studies were retrospective or prospective. Given that Ypsomed AG is based in Switzerland and the testing references international and US standards, the testing likely involved a mix of internal validation, third-party lab testing, and possibly user studies in relevant regions. All tests described are part of preclinical (non-clinical) performance validation, making them inherently prospective for the purpose of demonstrating device function and safety prior to marketing.

3. Number of Experts and Qualifications for Ground Truth

The document does not mention the use of experts in the context of establishing ground truth for the device's functional performance, as it is not an AI-driven diagnostic or interpretative device that relies on human expert consensus for its output. Its performance is evaluated against engineering specifications and physical/software functional requirements. The "Human Factors" testing would involve users, but not necessarily "experts" adjudicating correctness in the sense of accuracy for a diagnostic task.

4. Adjudication Method for the Test Set

Not applicable. The device's performance is determined by meeting pre-defined engineering and regulatory standards and testing protocols, not by expert adjudication of its output, as it does not produce subjective or interpretative results like an AI diagnostic algorithm.

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

Not performed/applicable. An MRMC study is relevant for AI systems that assist human readers in tasks like image interpretation to demonstrate improved diagnostic accuracy. This device is an "Injection Data Capture Device" providing feedback and recording information; it does not involve human readers interpreting data that the device enhances.

6. Standalone (Algorithm Only) Performance

While the device has software and algorithms to detect injection events and provide feedback, the document does not report "standalone" performance metrics in the way an AI diagnostic algorithm would (e.g., sensitivity, specificity). Its performance is demonstrated through the verification and validation of its hardware and software components (e.g., ability to detect spring position, successful data transfer, correct LED/audible feedback). The "Influence Testing" evaluates its performance in conjunction with the autoinjector, proving it does not negatively interfere.

7. Type of Ground Truth Used

The ground truth for the verification and validation of this device is engineering specifications, physical measurements, and adherence to established regulatory and industry standards. For example:

• Biocompatibility: Measured against established thresholds for cytotoxicity, sensitization, and irritation.
• EMC/Safety: Compliance with current versions of IEC standards.
• Software V&V: Compliance with software lifecycle processes and cybersecurity standards, and correct execution of defined functions (e.g., data recording, feedback activation).
• Mechanical/Lifetime: Physical measurements (e.g., activation force, dimension checks), cycle counts, and functional checks after simulated use/aging.
• Human Factors: User performance and subjective feedback against usability goals.

There is no "expert consensus," "pathology," or "outcomes data" ground truth in the context of its direct function (data capture and feedback).

8. Sample Size for the Training Set

Not applicable. This device is not an AI/machine learning system that requires a "training set" in the conventional sense (i.e., for learning to perform a complex, data-driven task like image recognition or diagnosis). Its functionality is based on programmed logic and sensor readings, not statistical learning from a large dataset.

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

Not applicable, as there is no "training set" for the type of device described. Input signals (e.g., from the inductive sensor about spring position) are processed based on predefined engineering parameters and logical rules to determine injection status, not learned from a dataset.

Ask a specific question about this device

The MallyaG Injection Pen Adapter is indicated for the capture and wireless transmission of dosing information of compatible reusable and disposable pen injectors.

The Novo Nordisk PDS290 Injection pens compatible for growth disorder treatment are:

• Sogroya (somapacitan-beco)
• Norditropin FlexPro

The MallyaG Injection Pen Adapter is a smart sensor composed of a button to be assembled onto a pen injector by covering the injection pen button. It is designed to be mounted on specific Novo Nordisk® PDS290 platform injection pens. A USB cable, necessary to charge the MallyaG device, is also provided in the package. The information is captured through the rotation of the dosing button of the pen into a value indicating the dose increment injected.

The MallyaG Injection Pen Adapter is indicated for the capture and wireless transmission of dosing information of compatible reusable and disposable pen injectors. The manufacturer, Biocorp Production, claims that the device is 99% accurate in matching recorded doses with dialed doses, with a margin of error of +/- 1 increment. This claim is supported by performance data from bench testing guided by ISO 11608-1.

Here's a breakdown of the requested information based on the provided document:

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 the sample size used for the performance testing (bench testing). It also does not specify 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 Qualifications of Those Experts

The document does not provide information on the number of experts used or their qualifications for establishing ground truth for the test set. The performance testing appears to be bench-based, guided by a standard (ISO 11608-1), which typically relies on calibrated equipment and defined procedures rather than expert human assessment for ground truth.

4. Adjudication Method for the Test Set

The document does not describe any adjudication method. Given the nature of bench testing for dose accuracy against a standard, an adjudication method for human interpretation would likely not be relevant.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and the Effect Size of How Much Human Readers Improve with AI vs Without AI Assistance

No, an MRMC comparative effectiveness study was not done. The MallyaG Injection Pen Adapter is a device for capturing and transmitting dose information, not an AI-assisted diagnostic or interpretative tool for human readers. Therefore, there is no discussion of human reader improvement with or without AI assistance.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done

Yes, a standalone performance assessment was done in the form of "Bench testing on performance, using ISO 11608-1 as a guide." This type of testing evaluates the device's ability to accurately capture and transmit dose information independently of human interaction beyond the initial setup and use.

7. The Type of Ground Truth Used

The ground truth for the performance testing (dose accuracy) was implicitly established by the dailing of specific increments on the injection pens and then comparing the device's recorded dose against that known dialed dose. This is a form of measurement against a known input/standard as guided by ISO 11608-1.

8. The Sample Size for the Training Set

The document does not mention a "training set" in the context of device performance. This device is an adapter for capturing data, not a machine learning or AI model that typically requires a training set. The performance testing focuses on verifying the device's accuracy in capturing physical actions.

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

Not applicable, as there is no mention of a training set for this device.

Ask a specific question about this device

The MallyaD Injection Pen Adapter is indicated for the capture and wireless transmission of compatible reusable and disposable pen injectors.

The Novo Nordisk PDS290 Injection pens compatible for diabetes treatment are:

• insulin degludec molecule (Tresiba U-100 FlexTouch and Tresiba U-200 FlexTouch)

• insulin aspart molecule (Fiasp FlexTouch)

• insulin degludec and liraglutide molecules (Xultophy)

The concerned Mallya Injection Pen Adapter, also called Mallyaº, is a smart sensor composed of a button to be assembled onto a pen injector by covering the injection pen button. It is designed to be mounted on specific Novo Nordisk® PDS290 platform injection pens. A USB cable, necessary to charge the Mallya® device is also provided in the package.

The given text is a 510(k) Premarket Notification from the FDA for a medical device called the "MallyaD injection pen adapter (MallyaD)". It details the device's indications for use, technological characteristics, and a comparison to a predicate device. While it mentions various tests and guidelines, it does not contain the specific information required to fully answer your request regarding the acceptance criteria and the study that proves the device meets them in the context of an AI/ML medical device.

The "MallyaD" device, as described, appears to be an adapter that records and transmits dosing information from an injection pen. It's a hardware device with associated software, but there is no explicit mention of it being an AI-powered medical device or using machine learning. The "Software controlled" attribute in the table refers to its internal operation rather than AI/ML algorithms.

Therefore, many of your requested points, which are highly relevant for AI/ML device studies, are not present in this document. For example, there's no mention of:

• A table of acceptance criteria for AI model performance (e.g., sensitivity, specificity, AUC).
• Sample sizes for a test set in the context of an AI model's performance on medical images or other data types.
• Data provenance for an AI test set.
• Number of experts, their qualifications, or adjudication methods for establishing ground truth for an AI model.
• Multi-Reader Multi-Case (MRMC) studies for AI assistance.
• Standalone performance for an AI algorithm.
• Type of ground truth for AI model training or testing (e.g., pathology, outcomes data).
• Sample size or ground truth establishment for an AI training set.

The document primarily focuses on the substantial equivalence to a predicate device based on its intended use, technological characteristics (e.g., BLE communication), and physical attributes, along with general performance tests like dose accuracy, biocompatibility, electrical safety, and cybersecurity. The "Dose accuracy: 99% of the recorded doses match the dialed doses with a margin of error of +/- 1 increment" is a performance claim, but it's for the device's basic function, not an AI/ML model's diagnostic or predictive performance.

In summary, based on the provided text, I cannot describe the acceptance criteria and the study that proves the device meets the acceptance criteria as if it were an AI/ML medical device because the document describes a hardware device with embedded software for data capture and transmission, not an AI/ML algorithm or model for diagnostic or predictive purposes.

Ask a specific question about this device

• Insulclock® is for use by patients using disposable insulin diagnosed with type I or type II diabetes mellitus.
• Insulclock® is for use by people from 18 to 65 years old. Out of this age range, the Insulclock® must be used under supervision.
• Insulclock® is designed for use in any home environment.
• Using a compatible App.

INTENDED USE: INSULCLOCK® v2.0 pro is intended to detect, store and transfer insulin dose-related data, the time of injection and the temperature to which the disposable insulin pen has been exposed to. All this information is sent to a mobile app via integrated SDK.

The INSULCLOCK® v2.0 pro is a reusable data transmitter with Software Development Kit (SDK) that detects and stores insulin dose- related data when attached to a disposable insulin pen and then transfers information to a compatible mobile application via Bluetooth® wireless technology which can display the brand of insulin, dose amount, injection time, and ambient temperature.

Models:

• INSULCLOCK® v2.0 PRO for Kwikpen® (Lilly®)
• INSULCLOCK® v2.0 PRO for Flex-touch® (Novo Nordisk®)

The provided FDA 510(k) summary for the INSULCLOCK® v2.0 PRO focuses on demonstrating substantial equivalence to a predicate device through non-clinical testing and does not contain information about acceptance criteria and a study proving those criteria are met in the context of device performance metrics typically reported for AI/ML devices (e.g., sensitivity, specificity, accuracy against a ground truth dataset).

This document mainly describes:

• The device's intended use and technological characteristics.
• Comparison of the device with a predicate device (Tempo Smart Button K212217).
• Summary of non-clinical testing performed to meet various safety, performance, and software standards (e.g., Biocompatibility, Software and Cybersecurity, Electrical Safety and EMC, Performance standards like UNE-EN 11608-1 for dose accuracy).
• A brief mention of a Human Factors validation study.
• A statement that clinical testing was not required for substantial equivalence.

Therefore, most of the information requested in your prompt regarding acceptance criteria for device performance/AI metrics, sample sizes for test sets, expert adjudication, MRMC studies, standalone performance, and ground truth establishment for AI/ML models cannot be extracted from this document.

However, I can provide what information is available concerning the requested points:

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

The document does not provide specific acceptance criteria or reported device performance in terms of metrics like sensitivity, specificity, or accuracy that would be relevant to an AI/ML device's diagnostic or predictive capabilities. Instead, it refers to compliance with established industry standards.

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

Not applicable to the information provided. The document outlines non-clinical engineering and usability testing, not performance evaluation against a labeled dataset.

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

Not applicable, as no ground truth establishment for a diagnostic or predictive AI/ML model is described.

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

Not applicable.

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 applicable.

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

Not applicable. The device, INSULCLOCK® v2.0 PRO, is a hardware device with an SDK that detects and transmits data, rather than a standalone AI algorithm making diagnostic or predictive decisions. Its performance is related to accurate data capture and transmission, which is evaluated through engineering standards compliance.

7. The type of ground truth used

For the specific performance aspect mentioned (Dose Accuracy via UNE-EN 11608-1), the ground truth would be the actual dose delivered by the compatible insulin pens, measured by standard metrological methods defined within that standard. For other standards, compliance is verified against the requirements of the standard itself (e.g., electrical safety tests, biocompatibility tests). There is no "expert consensus" or "pathology" ground truth described here.

8. The sample size for the training set

Not applicable. The document does not describe the development or training of an AI/ML model on a dataset.

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

Not applicable.

Ask a specific question about this device

The SoloSmart Injection Pen Adapter is indicated for the capture and wireless transmission of dosing information from compatible reusable and disposable pen injectors.

The device is made of a cap to be assembled onto a pen injector by covering the injection pen button. The overall device size is 21.4mm diameter & 23.3mm height. A USB cable, necessary to charge the SoloSmart® device is also provided in the package.

The provided FDA 510(k) summary for the SoloSmart Injection Pen Adapter does not contain information related to a study proving the device meets acceptance criteria for an AI/ML-based medical device. This document describes a traditional medical device (an adapter for injection pens) and its substantial equivalence to a predicate device, focusing on its physical and functional characteristics, rather than AI performance.

Therefore, I cannot provide the requested information about acceptance criteria for an AI/ML device, sample sizes for test sets, expert ground truth establishment, MRMC studies, standalone performance, or training set details, as this information is not present in the provided text.

The document discusses performance data related to the physical device and its ability to accurately record and transmit dosing information, but not the performance of an AI algorithm.

Here's a breakdown of what is in the provided text, and why it doesn't align with the request for AI/ML device performance:

• Device Type: SoloSmart Injection Pen Adapter, a physical device that attaches to insulin pens to capture and transmit dosing data. It is not described as an AI/ML device.
• Performance Data (Section 13): This section outlines "Bench testing on performance" for the physical adapter, including:
  • Dose accuracy of pen not affected by SoloSmart®.
  • Recording and transmission Accuracy.
  • Dose Prime Differentiation.
  • Complete Dose Notification.
  • Incomplete Injection Identification.
  • Physical Interaction with pen.
• Other Testing: Biocompatibility, Lifetime, Electrical safety, EMC and Radiocommunication, Cybersecurity Testing and Software Verification and Validation. These are standard tests for medical devices, not specific to AI/ML performance.
• Human Factor Validation (Section 14): Confirms usability of the device form factor and instructions for use.
• Substantial Equivalence (Section 8 & 12): The core of the 510(k) submission is to demonstrate that the SoloSmart® is substantially equivalent to a legally marketed predicate device (Mallya Injection Pen Adapter). This comparison focuses on indications for use, intended use, fundamental scientific technology, and technological characteristics (e.g., Bluetooth Low Energy communication, mechanism for recording dose).

In summary, the provided text does not describe an AI/ML-based medical device or any studies related to its AI performance. Thus, it is impossible to extract the requested information regarding AI acceptance criteria and proof of meeting those criteria.

Ask a specific question about this device

The Mallya Injection Pen Adapter is indicated for the capture and wireless transmission of dosing information from compatible reusable and disposable pen injectors.

The device is made of two parts to be assembled onto a pen injector. The Mallya base part (about 6 x 3 x 2 cm) which clips around the body of the injection pen and the Mallya button (about 2.5 x Ø2 cm) that slips over the injection pen button. A USB cable, necessary to charge the MALLYA device, and a reset key are also provided in the package.

This is a description of the Mallya Injection Pen Adapter, K222689. The document is for a 510(k) premarket notification, indicating it's a medical device.

1. Table of Acceptance Criteria and Reported Device Performance

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

The document does not specify the exact sample size used for the test set (e.g., number of injections, number of devices tested). It mentions "Bench testing on performance, using ISO 11608-1 as a guide" and "Design Verification." It does not provide information on the country of origin of the data or whether the data was retrospective or prospective.

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

This document does not provide details on the number of experts or their qualifications for establishing ground truth for the test set. The tests appear to be primarily engineering and performance-based (bench testing, dose accuracy assessments) rather than involving subjective expert interpretation of medical images or conditions.

4. Adjudication method for the test set

The document does not discuss any adjudication method (e.g., 2+1, 3+1). The nature of the tests (bench testing, electrical safety, lifetime, etc.) suggests objective measurements rather than subjective assessments requiring adjudication.

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

A multi-reader multi-case (MRMC) comparative effectiveness study was not done. The Mallya Injection Pen Adapter is a device that captures and transmits dosing information; it is not an AI-powered diagnostic or assistive tool for human readers interpreting medical cases. Therefore, the concept of improving human readers with AI assistance does not apply to this device.

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

The device itself is a standalone mechanism for capturing and transmitting data from an injection pen. Its performance (e.g., dose recording accuracy, transmission accuracy) would inherently be "standalone" in the sense that the device performs these functions automatically without direct human intervention in the data capture and transmission process itself. However, it requires a human to operate the injection pen. The specific "standalone" performance details (like sensitivity, specificity, accuracy metrics typically associated with algorithms) are not presented in this summary beyond general statements of accuracy.

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

The type of ground truth for most of the performance tests would be objective physical measurements or established engineering standards. For example:

• Dose accuracy: The ground truth would be the actual dose dialed and delivered by the injection pen, measured precisely using calibrated equipment.
• Recording and transmission accuracy: The ground truth would be the known, actual dose input/occurrence, compared to what the Mallya device records and transmits.
• Electrical safety/EMC: Ground truth is defined by compliance with the referenced IEC standards.
• Lifetime: Ground truth is reaching or exceeding the specified operational and shelf life under test conditions.

8. The sample size for the training set

The document does not mention any "training set." This type of device does not typically involve machine learning or AI models that require training data in the conventional sense. Its functionality is based on sensing physical rotation and transmitting data, rather than learning patterns from a dataset.

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

As there is no mention of a training set or machine learning components, this question is not applicable.

Ask a specific question about this device

The Tempo Smart Button is intended to detect, store, and transfer insulin dose-related data from a Tempo Pen to a compatible application such as the myTempo App (App).

The Tempo Smart Button is indicated for single-patients 18 years or older who are:

• . diagnosed with type 1 or type 2 diabetes mellitus,
• using prefilled insulin Tempo Pens [Basaglar® Tempo Pen (insulin glargine) injection 100 units/mL, Lyumjev®
• Tempo Pen (insulin lispro-aabc) injection 100 units/mL, Humalog® Tempo Pen (insulin lispro injection) 100 units/mL],
• using a compatible application (App).

The Tempo Smart Button is a reusable data transmitter that detects and stores insulin doserelated data when attached to a disposable Tempo Pen and then transfers information to a compatible mobile application via Bluetooth® wireless technology. The insulin dose-related data is used by the mobile application which can display the brand of insulin, dose amount, date, and time.

Additionally, the Tempo Smart Button is available as part of a Welcome Kit. The Tempo Welcome Kit and Refill Kit provide the convenience of having a single kit that contains devices needed for insulin administration and blood glucose monitoring.

The provided text does not contain specific acceptance criteria or a detailed study description for the Tempo Smart Button's performance related to its primary function of detecting, storing, and transferring insulin dose-related data. While general statements about meeting acceptance criteria and various tests are made, the specifics requested in your prompt are largely absent.

Here's an analysis of what information is available:

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

The document broadly states:

• "Lilly completed module testing for general design requirements, mechanical characterization, and accuracy of electronic dose capture. In all instances, the module met the acceptance criteria."
• "Bench Testing: Testing confirms that the module does not impact Tempo Pen compliance to specifications for dialing torque, injection force, and dose accuracy, nor does it impact Tempo Pen compliance to ISO 11608-1."

However, specific numerical acceptance criteria (e.g., accuracy percentage, tolerance levels) and the corresponding reported performance values for "accuracy of electronic dose capture" or other key functions are not provided in the text.

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

• Test Set Sample Size: Not specified.
• Data Provenance: Not specified (e.g., country of origin, retrospective/prospective). The document only mentions "Bench Testing" and "module testing."

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

• Number of Experts: Not specified.
• Qualifications of Experts: Not specified.

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

• Adjudication Method: Not specified.

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:

• This device is an "injection data capture device" and not an AI-assisted diagnostic or imaging tool. Therefore, an MRMC comparative effectiveness study involving human readers improving with AI assistance would not be applicable or expected for this type of device.

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

• The device's core function is autonomous data capture and transfer. The statement "accuracy of electronic dose capture. In all instances, the module met the acceptance criteria" implies standalone performance testing was conducted for this functionality. However, detailed results are missing.

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

• For the "accuracy of electronic dose capture," the ground truth would likely be the actual dose dialed on the pen, measured independently, or validated through mechanical means, rather than expert consensus or pathology. This is inferred from the bench testing context, but not explicitly stated.

8. The sample size for the training set:

• The document mentions "Software Verification and Validation" but does not refer to machine learning model training or a "training set" in the context of data used for algorithm development. This suggests the primary function is not based on a learnable AI model from a training set, but rather on direct electronic sensing and data processing.

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

• As a "training set" in the context of machine learning is not mentioned, this information is not applicable and hence not provided.

Summary of Available Information from the Provided Text:

The document confirms that the Tempo Smart Button underwent "Bench Testing" and "module testing for general design requirements, mechanical characterization, and accuracy of electronic dose capture," and "In all instances, the module met the acceptance criteria." It also states the device "does not impact Tempo Pen compliance to ISO 11608-1."

However, the specific quantitative acceptance criteria and the detailed results of these tests (e.g., specific accuracy rates for dose capture), along with the sample sizes, data provenance, and ground truth establishment methods for those specific performance aspects, are not disclosed in the provided FDA letter and 510(k) summary. The information focuses more on regulatory compliance (biocompatibility, electrical safety, software validation, cybersecurity) and substantial equivalence to a predicate device rather than in-depth performance study results for the core data capture function.

Ask a specific question about this device