Search Results

Regulation Number

Type

General and Plastic Surgery (SU)

Panel

Reference & Predicate Devices

K234133

Intended Use

The Dexcom G7 Continuous Glucose Monitoring System (Dexcom G7 CGM System or G7) is a real time, continuous glucose monitoring device indicated for the management of diabetes in persons 2 years and older.

The Dexcom G7 CGM System is intended to replace fingerstick BG testing for diabetes treatment decisions. Interpretation of the Dexcom G7 CGM System results should be based on the glucose trends and several sequential sensor readings over time. The Dexcom G7 CGM System also aids in the detection of episodes of hyperglycemia and hypodlycemia, facilitating both acute and long-term therapy adjustments.

The Dexcom G7 CGM System is also intended to autonomously communicate with digitally connected devices, including automated insulin dosing (AID) systems. The Dexcom G7 CGM System can be used alone or in conjunction with these digitally connected medical devices for the purpose of managing diabetes.

Device Description

The Dexcom G7 Continuous Glucose Monitoring System) is an interoperable continuous glucose monitoring (CGM) system intended to continuously measure the interstitial fluid, calculate the qucose reading and make this available to the user. The G7 System is intended for single patient use at home and requires a prescription.

The G7 System comprises the following primary components: a wearable, consisting of a G7 sensor and transmitter worn on the body and a display device, which can be a G7 Mobile App) on an iOS or Android OS smart device, or a G7 Receiver (Receiver).

To achieve the intended functions and performance of the G7 System, one sensor and at least one display device (App or Receiver) must be used together. The user must pair the display device(s) with each unique sensor to enable communication and start a sensor session. During an active session, the sensor reports new glucose data to the display device every 5-minutes. The displays glucose data and provides alerts and information signals to the user. The reportable glucose range for the G7 System is 40 mg/dL to 400 mg/dL. Glucose values below this range are reported as 'LOW' and glucose values above this range are reported as 'HIGH'. The sensor has an expected wear period of up to 10 days with an extended 12-hour grace period after the sensor session. The grace period allows additional time for the user to change the sensor at a convenient time.

The G7 is an interoperable connected device that can communicate glucose readings and other information wirelessly and securely to and from compatible electronic interfaces via the following secure wireless connections:

• Wireless communication from the transmitter directly to an interoperable device communicating through the same protocol

· The Mobile App communicates to another app on a single mobile platform

· The Mobile App communicates through the cloud to another software device

o Dexcom Partner Web APIs: The Dexcom Partner Web APIs enable secure and reliable communication of CGM data to authorized client software intended to receive the cloud. The Partner Web APIs is not intended to be used by automated insulin delivery systems (AID).

Principle of Operation:

The principles of operation for the Dexcom G7 System renain the same as prior generations of Dexcom CGM System uses a wire-type sensing mechanism that continuously measures interstitial glucose levels and uses a radio transmitter to wirelessly communicate glucose data to the display device for the user to see and use accordingly.

AI/ML Overview

The provided text is a 510(k) summary for the Dexcom G7 Continuous Glucose Monitoring System. It states that the device is substantially equivalent to a previously cleared Dexcom G7 Continuous Glucose Monitoring System (K234133). The key difference between the subject device and the predicate device is limited to "BLE communication and associated labeling updates." Crucially, it explicitly states: "No design change was required to support the extended BLE communication." and "No clinical data was necessary to determine substantial equivalence."

Therefore, the only acceptance criteria and study that proves the device meets the acceptance criteria mentioned in this document pertain to this specific change:

Acceptance Criteria and Study for Dexcom G7 Continuous Glucose Monitoring System (K240902)

Given that the submission is for an updated version of an already cleared device (Dexcom G7 K234133) and the only difference noted is related to BLE communication and associated labeling updates, with no design change, the acceptance criteria and study detailed here are specific to proving the extended BLE communication functionality.

1. Table of Acceptance Criteria and Reported Device Performance:

Acceptance Criteria Category	Specific Acceptance Criteria	Reported Device Performance
Communication Range	Extended BLE communication range specification.	The subject device G7 CGM System met the extended BLE communication range specification.

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

The document states, "No design change was required to support the extended BLE communication." and "No clinical data was necessary to determine substantial equivalence." This implies that the testing for the BLE communication range was likely a non-clinical bench top or engineering test.

Sample Size: Not specified in the provided text, but it would pertain to the number of devices or test repetitions used in the communication range testing.
Data Provenance: Not specified, but generally, non-clinical engineering tests would be controlled laboratory environments. It is implicitly prospective testing conducted specifically for this regulatory submission.

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

This information is not applicable for the specific testing described. The acceptance criteria for BLE communication range would be met through quantitative measurements (e.g., signal strength, data transmission success rate over distance) against a pre-defined engineering specification, not through human expert assessment or ground truth establishment in the traditional sense for medical device diagnostics.

4. Adjudication Method for the Test Set:

This information is not applicable for the specific testing described. Adjudication methods are typically used in clinical studies or image review settings involving human interpretation. The BLE communication range testing would rely on automated data collection and comparison to a specification.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done, and what was the effect size of how much human readers improve with AI vs without AI assistance:

This information is not applicable. This submission is for a continuous glucose monitoring system, not an AI-assisted diagnostic imaging device. The document explicitly states, "No clinical data was necessary to determine substantial equivalence."

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

The primary "performance" discussed for this specific 510(k) is the extended BLE communication range. This would be assessed as a "standalone" technical capability of the device's communication module, without direct human intervention in the communication process itself. However, it's not an "algorithm only" performance assessment in the sense of AI diagnostics.

7. The Type of Ground Truth Used:

For the BLE communication range, the "ground truth" would be the engineering specification for the extended communication range itself, against which the measured performance of the device's BLE module is compared. It's a quantitative, pre-defined target.

8. The Sample Size for the Training Set:

This information is not applicable. There is no mention of a "training set" as this is not an AI/machine learning model where a distinct training phase is described.

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

This information is not applicable for the same reason as above.

Ask a Question

Ask a specific question about this device

K Number

K234070

Device Name

Stelo Glucose Biosensor System

Manufacturer

Date Cleared

2024-03-05

(74 days)

Product Code

SAF

Regulation Number

Type

Panel

Dexcom G7 Continuous Glucose Monitoring System

Reference & Predicate Devices

K231081

Intended Use

The Stelo Glucose Biosensor System is an over-the-counter (OTC) integrated Continuous Glucose Monitor (CCGM) intended to continuously measure, record, analyze, and display glucose values in people 18 years and older not on insulin. The Stelo Glucose Biosensor System helps to detect normal (euglycemic) and low or high (dysglycemic) glucose levels. The Stelo Glucose Biosensor System may also the user better understand how lifestyle and behavior modification, including diet and exercise, impact glucose excursion.

The user is not intended to take medical action based on the device output without consultation with a qualified healtheare professional.

Device Description

The Stelo Glucose Biosensor system (Stelo System) is an over-the-counter (OTC) interoperable continuous glucose monitoring (iCGM) system.

The Stelo Glucose Biosensor system (Stelo System) is an interoperable connected device that measures and displays estimated glucose values for people who are not on insulin. The Stelo System consists of the following components: the Glucose Sensing Subsystem (GSS) and the Mobile Application Subsystem (MAS). The GSS is comprised of the sensor applicator and on-body wearable, which includes a Bluetooth Low Energy (BLE) molded transmitter, adhesive patch and sensor. The sensor is a small and flexible wire, which is inserted by the applicator into subcutaneous tissue where it converts glucose into electrical current. The transmitter's onboard algorithm converts these measurements into estimated glucose values and calculates the glucose rate of change which are sent every 5 minutes to the MAS is an app that can be downloaded to a compatible smart device and that presents glucose readings and qlucose trend to the user every 15 minutes. As such, the most recent displayed glucose value might be up to 15 minutes old. Each sensor session lasts up to 15 days with an extended 12-hour grace period. The grace period allows additional time for the user to change the sensor at a convenient time.

The proposed Stelo System is based on the same mode of operation and mechanism of reaction as the predicate G7 CGM System (K231081), which uses a wire type sensing mechanism that continuously measures interstitial fluid qlucose levels and a BLE enabled radio transmitter to wirelessly communicate CGM data to compatible display devices at regular 5-minute intervals. These data are also able to be reliably and securely transmitted to other digitally connected devices, excluding insulin pens and Automated Insulin Dosing (AID) systems.

The Stelo System uses the same hardware design as the predicate G7 CGM System. The Stelo GSS firmware is designed to support a factory-calibrated only device (without calibration inputs), to extend the sensor wear duration from 10 to 15 days while maintaining the accuracy of the device, and to connect to authorized display devices only (i.e., Stelo MAS). The Stelo MAS includes a redesigned user interface (UI) tailored to the Stelo System's user population to simplify the use of the device. The UI includes an app onboarding specific to the Stelo MAS design and its intended use, the most recent glucose value and trend graph which are updated every 15 minutes, a narrowed glucose range display from 70 mg/dL, and an "Insights" feature providing the time in range percentage with suggestions to help users improve their time in range. The Ul does not provide any glucose or system alerts.

AI/ML Overview

The Dexcom Stelo Glucose Biosensor System is a continuous glucose monitoring (CGM) device intended for over-the-counter (OTC) use by adults aged 18 and older who are not on insulin. The device continuously measures, records, and displays glucose values, helping to detect normal, low, or high glucose levels and allowing users to understand the impact of lifestyle modifications on glucose excursion.

Here's a breakdown of the acceptance criteria and study information:

1. Table of Acceptance Criteria and Reported Device Performance:

The document explicitly states that the device meets the iCGM special controls for clinical performance set forth in 21 CFR 862.1355. While specific numerical acceptance criteria (e.g., MARD values, percentages of readings within certain zones) are not provided in the summary, the general umbrella acceptance criterion is:

Acceptance Criterion	Reported Device Performance
iCGM special controls for clinical performance (21 CFR 862.1355)	Met
Device-related adverse event (AE) incidence	Acceptable. Reported AEs included local irritation (edema) and pain/discomfort.
Nonclinical performance (e.g., electrical, mechanical, environmental, human factors, software, cybersecurity)	Met pre-defined acceptance criteria

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

The document states: "A clinical study was conducted to evaluate the safety and effectiveness of the Stelo Glucose Biosensor System. The effectiveness of the device was evaluated with respect to reference venous plasma sample YSI measurements across the measuring range throughout a 15-day wear duration with a 12-hour grace period in adult (18 years and older) participants with diabetes."

Test Set Sample Size: The exact number of participants is not specified, but it refers to "adult (18 years and older) participants with diabetes."
Data Provenance: Prospective clinical study data. The country of origin is not explicitly stated in the provided text.

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

The ground truth for the clinical study was established using "reference venous plasma sample YSI measurements." This usually refers to laboratory-based glucose analysis performed by trained technicians, not "experts" in the context of interpretation (like radiologists). The number and qualifications of technicians are not mentioned.

4. Adjudication Method for the Test Set:

Not applicable, as the ground truth was established through laboratory reference measurements (YSI) of venous plasma samples, which are generally considered the gold standard and do not typically require adjudication.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done:

No, an MRMC comparative effectiveness study was not done. The study evaluated the standalone performance of the Stelo Glucose Biosensor System against reference measurements. The document does not mention any human readers or human-AI interaction in the context of this study.

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

Yes, a standalone study was done. The clinical study evaluated the Stelo Glucose Biosensor System's effectiveness by comparing its glucose measurements to reference venous plasma YSI measurements, implying an assessment of the device's algorithmic and sensing accuracy without human intervention for interpretation beyond what the user does with the displayed values.

7. The Type of Ground Truth Used:

The ground truth used was laboratory reference measurements ("reference venous plasma sample YSI measurements").

8. The Sample Size for the Training Set:

The document does not provide any information about the training set size for the algorithms within the Stelo Glucose Biosensor System. This information is typically not included in a 510(k) summary unless the submission is for an AI/ML SaMD where the training data characteristics are a primary focus. Given that the Stelo System uses the "same hardware design as the predicate G7 CGM System" and its GSS firmware is designed to support a "factory-calibrated only device (without calibration inputs)," it suggests that the core algorithms might have been trained previously and validated, or leveraged from the predicate device's development.

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

This information is not provided in the document.

Ask a Question

Ask a specific question about this device

K Number

K234133

Device Name

Manufacturer

Dexcom Inc.

Date Cleared

2024-02-26

(59 days)

Product Code

Regulation Number

Type

General and Plastic Surgery (SU)

Panel

Reference & Predicate Devices

K231081

Intended Use

The Dexcom G7 Continuous Glucose Monitoring System (Dexcom G7 CGM System or G7) is a real time, continuous qlucose monitoring device indicated for the management of diabetes in persons 2 years and older.

Device Description

The Dexcom G7 Continuous Glucose Monitoring System (G7 System) is an interoperable continuous glucose monitoring (CGM) system intended to continuously measure the glucose in the interstitial fluid, calculate the glucose reading and make this available to the user.The G7 System is intended for single patient use at home and requires a prescription.

The proposed G7 CGM System is modified to allow users to display glucose data directly from the transmitter on compatible iOS smartwatches.

The G7 is an interoperable connected device that can communicate glucose readings and other information wirelessy and securely to and from compatible electronic interfaces via the following secure wireless connections:

· Wireless communication from the transmitter directly to an interoperable device communicating through the same protocol
· The Mobile App communicates to another app on a single mobile platform
· The Mobile App communicates through the cloud to another software device o Dexcom Partner Web APIs: The Dexcom Partner Web APIs enable secure and reliable communication of CGM data to authorized client software intended to receive the data through the cloud. The Partner Web APIs is not intended to be used by automated insulin delivery systems (AID).

Principle of Operation:

AI/ML Overview

I am sorry, but the provided text does not contain the specific information required to describe the acceptance criteria and the study that proves the device meets those criteria, especially regarding a table of acceptance criteria and reported device performance, sample size for test and training sets, data provenance, expert involvement for ground truth, adjudication methods, MRMC studies, or standalone algorithm performance.

The document is a 510(k) premarket notification letter from the FDA to Dexcom Inc. for their G7 Continuous Glucose Monitoring System, confirming its substantial equivalence to a predicate device. It briefly mentions non-clinical testing performed to establish performance characteristics related to the "Direct to Watch (DTW) feature" (EMC, communication range, human factors, software V&V, cybersecurity) and states that these tests demonstrate the system meets "pre-defined acceptance criteria." However, it does not detail what those acceptance criteria are or present the performance results against them. It also explicitly states, "No clinical data was necessary to determine substantial equivalence."

Therefore, I cannot fulfill your request for the detailed information as the source text does not provide it.

Ask a Question

Ask a specific question about this device

K Number

K231081

Device Name

Dexcom G7 Continuous Glucose Monitoring (CGM) System

Manufacturer

Date Cleared

2023-05-15

(28 days)

Product Code

Regulation Number

Type

Panel

Dexcom G6 Continuous Glucose Monitoring (CGM) System

Reference & Predicate Devices

K213919

Intended Use

The Dexcom G7 Continuous Glucose Monitoring (CGM) System is a real time, continuous glucose monitoring device indicated for the management of diabetes in persons 2 years and older.

The Dexcom G7 CGM System is intended to replace fingerstick BG testing for diabetes treatment decisions. Interpretation of the Dexcom G7 CGM System results should be based on the glucose trends and several sequential sensor readings over time. The Dexcom G7 CGM System also aids in the detection of episodes of hyperglycemia and hypoglycemia, facilitating both acute and long-term therapy adjustments.

Device Description

The Dexcom G7 Continuous Glucose Monitoring System) is an interoperable connected device that measures and displays estimated dlucose values for people with diabetes. The G7 System consists of the following components: the Glucose Sensing Subsystem (GSS), the Mobile Application Subsystem (MAS), the Receiver Subsystem (RVS), The GSS is comprised of the sensor applicator and on-body wearable, which includes a Bluetooth Low Energy (BLE) transmitter, adhesive patch and sensor. The sensor is a small and flexible wire, which is inserted by the applicator into subcutaneous tissue where it converts glucose into electrical current. The sensor has an expected wear period of up to 10 days with an extended 12-hour grace period after the sensor session. The grace period allows additional time for the user to change the sensor at a convenient time.

The transmitter is pre-connected to the sensor and is cradled into the applicator needle inside the applicator housing. The applicator external housing consists of a cap and shroud which utilize a threaded cap and seal to create the sterile barrier system. A deployment lock mechanism prevents insertion of the on-body wearable until the applicator is pressed against the insertion site. The insertion is then completed with a single button press vertical spring deployed mechanism, which introduces the sensor via the subcutaneous tissue while also placing the embedded wearable onto the body. The wearable adheres to the skin via an adhesive patch.

After deployment, the transmitter initiates automatic wakeup and session start. The sensor's small and flexible wire converts qlucose to electrical current and the transmitter samples the electrical current produced by the sensor. The transmitter's onboard algorithm converts these measurements into estimated qlucose values and calculates the glucose rate of change; the data are sent every 5 minutes to the MAS and/or the RVS. The MAS and RVS are display devices that present the current ducose reading and glucose trend to the user. Both display devices alert the user when glucose levels are outside of a target zone and when specific system states occur. The G7 System is designed to communicate to one or both display devices simultaneously.

The G7 System is also designed to communicate estimated glucose values, trend and system information to other compatible electronic interfaces via the following secure wireless connections:

. Wireless communication from the transmitter directly to an interoperable device communicating through the same protocol
. The app communicates to another app on a single mobile platform
The app communicates through the cloud to another software device ●
- Dexcom Partner Web APIs: The Dexcom Partner Web APIs enable secure and o reliable communication of CGM data to authorized client software intended to receive the data through the cloud. The Partner Web APIs is not intended to be used by automated insulin delivery systems (AID).

The proposed G7 CGM System uses the same mode of operation and mechanism of reaction as the predicate G7 CGM System (K213919). The proposed G7 CGM System uses an alternate GSS wearable adhesive.

AI/ML Overview

The provided text describes a 510(k) premarket notification for the Dexcom G7 Continuous Glucose Monitoring (CGM) System. The submission is for a modification to an already cleared device, specifically changing the adhesive patch.

Since this is a submission for a modification, the document frequently refers to the "predicate device" (K213919), which is essentially the previous version of the Dexcom G7. The current submission argues that the modified device is "substantially equivalent" to this predicate. As such, the performance data presented (or referenced indirectly) in this document likely refers to the performance of the predicate device, which the modified device is claimed to match.

Here's an analysis of the acceptance criteria and study information, extracting what's available and noting what is not explicitly stated in this particular document:

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

The document states: "The proposed Dexcom G7 CGM System was verified and validated according to Dexcom's internal design control process and in accordance with special controls for integrated continuous glucose monitoring systems. This testing demonstrated that the proposed system performed according to its specifications; and the proposed system has met its technological and performance criteria which have not changed from the predicate device."

This statement asserts that the device meets its performance criteria and specifications, and that these criteria have not changed from the predicate device. However, the exact quantitative acceptance criteria (e.g., specific MARD values, accuracy at different glucose ranges) and the reported device performance against those criteria are not explicitly provided in this document. This information would typically be found in the full 510(k) submission not included here, likely in the sections detailing the clinical study for the original device clearance (K213919).

2. Sample size used for the test set and the data provenance (e.g., country of origin of the data, retrospective or prospective)

This document explicitly states it is a modification to a previously cleared device. It only mentions that the "proposed Dexcom G7 CGM System was verified and validated according to Dexcom's internal design control process and in accordance with special controls for integrated continuous glucose monitoring systems."

Therefore, the specific sample size, data provenance (country of origin), and whether the study was retrospective or prospective are not provided in this document. This information would be crucial for the original submission (K213919) that established the performance of the Dexcom G7.

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)

Given that this is a Continuous Glucose Monitoring (CGM) system, the "ground truth" for glucose measurements is typically established using a reference laboratory blood glucose analyzer, not human experts. Therefore, the concept of "number of experts" is not applicable in the context of establishing ground truth for glucose values in this type of device.

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

Since the ground truth is established by a laboratory analyzer, there is no adjudication method involving human experts for the glucose values themselves.

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 Dexcom G7 is a standalone medical device that measures glucose; it is not an AI-assisted diagnostic imaging device that involves human readers interpreting images. Therefore, a multi-reader multi-case (MRMC) comparative effectiveness study with human readers assisting AI is not applicable to this device.

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

This document describes the Dexcom G7 CGM System as "a real time, continuous glucose monitoring device." Its core function is to measure and display estimated glucose values. The device includes a "transmitter's onboard algorithm" that "converts these measurements into estimated glucose values and calculates the glucose rate of change." It also states, "The Dexcom G7 CGM System is intended to replace fingerstick BG testing for diabetes treatment decisions." This strongly implies that the device is intended for standalone performance in providing glucose measurements that can be used directly for treatment decisions, without requiring human interpretation of raw sensor data or assistance from human operators to derive the glucose values themselves.

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

The ground truth for a CGM device is established by comparison with a highly accurate reference method for measuring blood glucose, typically a laboratory-grade blood glucose analyzer (e.g., YSI analyzer) measuring venous blood samples. This is mentioned implicitly by the nature of CGM device testing.

8. The sample size for the training set

This document pertains to a 510(k) submission for a modified device, specifically a change in adhesive. While the device (and its underlying algorithm) would have been developed using training data, the size of that training set is not provided in this specific document. It would have been part of the original K213919 submission.

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

Similar to the ground truth for the test set, the ground truth for any training data used for the algorithm in a CGM device would have been established by comparison with a highly accurate reference method for measuring blood glucose, usually a laboratory blood glucose analyzer. This information is not explicitly provided in this document but is standard practice for CGM development.

In summary:

This 510(k) submission for the Dexcom G7 focuses on demonstrating substantial equivalence of a modified device (specifically, a new adhesive) to a previously cleared predicate device. It asserts that the modifications do not change the intended use or fundamental scientific technology, and that the device continues to meet its performance specifications. However, the detailed performance data, study design, and sample sizes for the original clinical validation (which would contain the answers to many of these questions) are not included in this summary document. This document emphasizes that "the proposed G7 CGM System uses the same mode of operation and mechanism of reaction as the predicate G7 CGM System (K213919)" and its "technological and performance criteria ... have not changed from the predicate device."

Ask a Question

Ask a specific question about this device

K Number

K223931

Device Name

Manufacturer

Date Cleared

2023-01-11

(12 days)

Product Code

Regulation Number

Type

Panel

Dexcom G7 Continuous Glucose Monitoring System

Reference & Predicate Devices

K213919,K221259

Intended Use

The Dexcom G6 Continuous Glucose Monitoring System) is a real time, continuous glucose monitoring device indicated for the management of diabetes in persons age 2 years and older.

The Dexcom G6 System is intended to replace fingerstick blood glucose testing for diabetes treatment decisions. Interpretation of the Dexcom G6 System results should be based on the glucose trends and sequential readings over time. The Dexcom G6 System also aids in the detection of evisodes of hyperglycemia, facilitating both acute and long-term therapy adjustments.

The Dexcom G6 System is also intended to autonomously communicate with digitally connected devices, including automated insulin dosing (AID) systems. The Dexcom G6 System can be used alone or in conjunction with these digitally connected medical devices for the purpose of managing diabetes.

Device Description

The proposed Dexcom G6 CGM System is based on the same physical principles and fundamental design as the predicate but includes an alternative receiver. This receiver has identical hardware to the G7 receiver (K213919). The Dexcom G6 CGM System is designed to function as intended with either the proposed or current commercial receiver. The proposed receiver has the same function as the commercial receiver.

The Dexcom G6 Continuous Glucose Monitoring System is an interoperable connected device that measures and displays glucose values for patients with diabetes. The G6 CGM System consists of three main components: a sensor/applicator, a Bluetooth Low Energy (BLE) transmitter, and a BLE enabled display device (receiver and/or mobile application). The sensor is a small and flexible wire, which is inserted by the applicator into subcutaneous tissue where it converts ducose into electrical current. The sensor has an expected wear time of up to 10 days. The transmitter is connected to the sensor and is worn on the body. The transmitter samples the electrical current produced by the sensor and converts these measurements into estimated glucose values (EGV) using an onboard algorithm. The transmitter sends glucose data to the receiver and/or mobile app, which displays the current qlucose reading (updated every 5 minutes) and glucose trends. The receiver and/or mobile app also alerts the user when glucose levels are outside of a target zone and when other important system conditions occur. The G6 CGM System is designed to be used alone or in conjunction with digitally connected devices. The G6 CGM System can communicate Estimated Glucose Values (EGV) and other information wirelessly and securely to and from these digitally connected devices.

AI/ML Overview

Based on the provided text, the acceptance criteria and study details for the Dexcom G6 Continuous Glucose Monitoring (CGM) System are not explicitly detailed in a table format with specific performance metrics. This document primarily focuses on demonstrating substantial equivalence to a predicate device, K221259, rather than presenting a comprehensive clinical performance study.

However, I can infer some information based on the context:

Acceptance Criteria and Reported Device Performance

The document states that the proposed Dexcom G6 CGM System performs "according to specifications and meets the technological and performance criteria which have not changed from the predicate device." This implies that the acceptance criteria for the new device are the same as those established for the predicate device (Dexcom G6 CGM System, K221259). Since the core components (sensor, applicator, transmitter, and app requirements) are the same, the performance is expected to be identical to the predicate.

While quantitative acceptance criteria are not explicitly listed in this document, for CGM devices, such criteria typically relate to accuracy, precision, and reliability of glucose measurements compared to a reference method (e.g., YSI analyzer for blood glucose). The predicate device would have met specific MARD (Mean Absolute Relative Difference) targets and operated within defined accuracy zones (e.g., Consensus Error Grid Analysis).

Inferred Table of Acceptance Criteria and Reported Device Performance:

Acceptance Criteria Category	Specific Metric (Inferred)	Acceptance Threshold (Inferred from predicate)	Reported Device Performance (Inferred from predicate)
Accuracy	Mean Absolute Relative Difference (MARD)	(Not specified, but generally 95-99% in A+B)	Met predicate device's established CEGA performance
Reliability	Sensor wear time	Up to 10 days	Up to 10 days
Functionality	Communication with display devices (receiver/mobile app)	Reliable and timely data transmission	Reliable and timely data transmission
Functionality	Alerting capabilities (hyper/hypoglycemia)	Accurate and timely alerts	Accurate and timely alerts
Safety	Human Factors (usability)	Validated through human factors testing	Validated through human factors testing

Study Details:

Sample size used for the test set and the data provenance:
The document states: "Therefore, performance testing and software verification and validation testing completed for the predicate device (K221259) remain applicable." This indicates that the performance data used to demonstrate safety and effectiveness for this specific submission (K223931) is largely derived from the studies conducted for the predicate device (K221259). As such, the sample size and data provenance for the predicate device's original studies would be relevant, but are not provided in this document. For K223931, the testing specifically focused on the new receiver.
- No specific sample size for a test set is provided within this document for K223931.
- Data provenance for K223931's specific testing of the new receiver is not described (e.g., country of origin, retrospective/prospective). It mentions "firmware and hardware testing."
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
This information is not provided in the document. The ground truth for CGM studies typically involves laboratory reference methods (e.g., YSI blood glucose measurements), not expert consensus in the same way as imaging studies.
Adjudication method (e.g., 2+1, 3+1, none) for the test set:
This information is not provided in the document. Adjudication methods are not typically applicable to CGM device performance studies, as ground truth is established through a quantitative laboratory reference method.
If a multi-reader multi-case (MRMC) comparative effectiveness study was done:
No, an MRMC comparative effectiveness study was not done. This type of study is relevant for diagnostic imaging where human readers interpret results, often with and without AI assistance. For a CGM device, the performance is measured against a quantitative reference method, not human interpretation.
If a standalone (i.e. algorithm only without human-in-the loop performance) was done:
Yes, implicitly. The core of a CGM system's performance, particularly its accuracy (MARD, CEGA), is essentially a standalone (algorithm-only) assessment, as the device's algorithm generates the glucose values. The document states the "Glucose Value Estimation Algorithm" is a "Joint Probability Algorithm" which is the "Same" as the predicate. The performance testing for the predicate would have covered this standalone aspect.
The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
While not explicitly stated for K223931, CGM studies generally use laboratory reference blood glucose measurements (e.g., YSI glucose analyzer) as the ground truth. This is the standard for evaluating the accuracy of glucose monitoring devices.
The sample size for the training set:
The document does not provide information regarding the sample size for the training set. Since the "Glucose Value Estimation Algorithm" is the "Same" as the predicate device, it implies that the algorithm was trained using data collected for the development of the original G6 system.
How the ground truth for the training set was established:
The document does not provide information on how the ground truth for the training set was established. However, similar to the test set, it would typically be established using laboratory reference blood glucose measurements during the development and training phases of the original Dexcom G6 system's algorithm.

Ask a Question

Ask a specific question about this device

K Number

K213919

Device Name

Manufacturer

Date Cleared

2022-12-07

(357 days)

Product Code

Regulation Number

Type

Panel

Dexcom G6 Continuous Glucose Monitoring (CGM) System, Dexcom G6 Glucose Program Continuous Glucose Monitoring (CGM) System, Dexcom G6 Professional Continuous Glucose Monitoring (CGM) System

Reference & Predicate Devices

K201328

Intended Use

The Dexcom G7 Continuous Glucose Monitoring (CGM) System is a real time, continuous glucose monitoring device indicated for the management of diabetes in persons 2 years and older.

The Dexcom G7 CGM System is intended to replace fingerstick BG testing for diabetes treatment decisions. Interpretation of the Dexcom G7 CGM System results should be based on the glucose trends and sequential sensor readings over time. The Dexcom G7 CGM System also aids in the detection of episodes of hyperglycemia and hypoglycemia, facilitating both acute and long-term therapy adjustments.

Device Description

The Dexcom G7 Continuous Glucose Monitoring System (G7 System) is an interoperable connected device that measures and displays estimated glucose values for people with diabetes. The G7 System consists of the following components: the Glucose Sensing Subsystem (GSS), the Mobile Application Subsystem (MAS), the Receiver Subsystem (RVS). The GSS is comprised of the sensor applicator and on-body wearable, which includes a Bluetooth Low Energy (BLE) molded transmitter, adhesive patch and sensor. The sensor is a small and flexible wire, which is inserted by the applicator into subcutaneous tissue where it converts glucose into electrical current. The sensor has an expected wear period of up to 10 days with an extended 12-hour grace period after the sensor session. The grace period allows additional time for the user to change the sensor at a convenient time.

The molded transmitter is pre-connected to the sensor and is cradled into the applicator needle inside the applicator housing. The applicator external housing consists of a cap and shroud which utilize a threaded cap and seal to create the sterile barrier system. A deployment lock mechanism prevents insertion of the on-body wearable until the applicator is pressed against the insertion site. The insertion is a single button press vertical spring deployed mechanism which introduces the sensor via the needle into the subcutaneous tissue, while also placing the embedded wearable onto the body which adheres to the skin via an adhesive patch.

After deployment, the molded transmitter initiates automatic wakeup and session start. The sensor's small and flexible wire converts glucose to electrical current and the transmitter samples the electrical current produced by the sensor. The transmitter's onboard algorithm converts these measurements into estimated glucose values and calculates the glucose rate of change which are sent every 5 minutes to the MAS and/or the RVS. The MAS and RVS are display devices which present the current qlucose reading and glucose trend to the user. Both display devices alert the user when glucose levels are outside of a target zone and when specific system states occur. The G7 System is designed to communicate to one or both display devices simultaneously.

The G7 System is also designed to communicate estimated glucose values, trend and system information to other compatible electronic interfaces via the following secure wireless connections:

Wireless communication from the transmitter directly to an interoperable device communicating through the same protocol
. The app communicates to another app on a single mobile platform
. The app communicates through the cloud to another software device.
Dexcom Partner Web APIs: The Dexcom Partner Web APIs enable o secure and reliable communication of CGM data to authorized client software intended to receive the data through the cloud. The Partner Web APIs is not intended to be used by automated insulin delivery systems (AID).

AI/ML Overview

The provided document (K213919) is a 510(k) premarket notification for the Dexcom G7 Continuous Glucose Monitoring (CGM) System. While it discusses the device's characteristics and compares them to a predicate device (Dexcom G6), it does not contain the specific acceptance criteria or detailed results of a clinical study that proves the device meets those criteria in the typical format of a clinical trial report.

The document states:
"The Dexcom G7 CGM System was verified and validated according to Dexcom's internal design control processes and in accordance with special controls for integrated continuous glucose monitors. This testing demonstrated that the System performed accordingly to its specifications and that the technological and performance criteria are comparable to the predicate device."

This indicates that internal testing was performed, but the detailed results, sample sizes, ground truth establishment, or expert involvement are not explicitly provided in this summary.

Therefore, I cannot fully provide all the requested information. However, based on the information provided and general knowledge of CGM device regulatory requirements, I can infer and construct some parts of the answer.

Here's an attempt to answer your request based on the available information and reasonable assumptions for a 510(k) submission for a CGM device:

Device: Dexcom G7 Continuous Glucose Monitoring System (K213919)

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly present a table of acceptance criteria with corresponding performance data. However, for continuous glucose monitoring systems, a key performance metric is accuracy compared to a reference blood glucose measurement. This is typically assessed using metrics like Mean Absolute Relative Difference (MARD) and various error grid analyses (e.g., Consensus Error Grid, Parkes Error Grid).

Inferred Acceptance Criteria (Typical for iCGM) and Assumed Reported Performance (Based on 510(k) Clearance for a similar device):

Performance Metric	Acceptance Criteria (Typical for iCGM)	Reported Device Performance (Inferred from 510(k) clearance, specific data not provided in this document)
Accuracy (MARD)	MARD (Mean Absolute Relative Difference) against YSI reference 95% of points in Zones A+B for all glycemic ranges (adults and children); very few, if any, points in Zones C, D, E.	The device would have met the stringent requirements for clinical accuracy as assessed by Consensus Error Grid or similar error grid analyses, demonstrating that a vast majority of CGM readings are clinically accurate (Zone A) or lead to benign or no treatment errors (Zone B). Specific percentages for each zone are not presented in this summary document but are a standard part of CGM submissions.
Rate of Change Accuracy	Performance metrics demonstrating reliable detection of rapid glycemic changes (e.g., >80% accuracy for alerts/alarms for rapid rise/fall).	The system's ability to aid in the detection of hyperglycemia and hypoglycemia and facilitate acute/long-term therapy adjustments implies meeting performance criteria for rate of change and alert accuracy. Specific metrics are not presented.
Sensor Warm-Up Time	Target: Within 30 minutes	Reported: Within 30 minutes (specifically 27 minutes)
Sensor Life	Target: Up to 10 days with a grace period	Reported: Up to 10 days with a 12-hour grace period
Minimum Measuring Range	40-400 mg/dL	Reported: 40-400 mg/dL
Replace Fingerstick BG Testing	Clinical data supporting the ability to replace fingerstick BG testing for diabetes treatment decisions based on accuracy and reliability.	The indication for use explicitly states the system "is intended to replace fingerstick BG testing for diabetes treatment decisions." This implies that the underlying clinical data met the necessary performance thresholds for this claim.

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

The document does not specify the exact sample size for the clinical test set. However, 510(k) submissions for iCGM devices typically involve robust clinical studies with hundreds of participants across various age groups (adults and children/adolescents from 2 years old, as per the indication for use) to evaluate performance under real-world conditions and in controlled clinical settings (e.g., hyperglycemic and hypoglycemic clamps).

The data provenance is not specified. However, for a major medical device company like Dexcom, these studies are typically multi-center, prospective clinical trials conducted in the United States and potentially other geographies, adhering to GCP (Good Clinical Practice) guidelines. The document does not specify whether the data was retrospective or prospective, but clinical validation for a device of this class would almost certainly involve prospective data collection.

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

For CGM device validation, ground truth is not established by "experts" in the sense of image readers, but rather through highly accurate laboratory reference methods.

Ground Truth Method: Reference blood glucose measurements obtained from a YSI Glucose Analyzer, which is a gold standard laboratory method.
Personnel Qualifications: Highly trained clinical staff and laboratory technicians who follow strict protocols for blood sample collection and YSI analysis. Not typically "experts" like radiologists, but rather clinical and lab professionals.

4. Adjudication Method for the Test Set:

Not applicable in the context of a CGM device. Adjudication methods (e.g., 2+1, 3+1) are typically used in studies where human readers are interpreting images or other subjective data, and a consensus or authoritative reading is needed for ground truth or comparison. For CGM, the "ground truth" is a direct, objective laboratory measurement (YSI).

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

Not applicable. MRMC studies are used for evaluating the impact of AI on human readers for diagnostic imaging, not for continuous glucose monitors, which provide quantitative measurements. The Dexcom G7 System does not involve human "reading" of data in the same way as, for example, a radiology AI.

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

Yes, in essence. The primary performance assessment of a CGM device like the Dexcom G7 is its stand-alone accuracy (the algorithm's calculations of glucose values) when compared to a laboratory reference. While the device is intended for human use and interpretation of trends, its fundamental accuracy metrics (e.g., MARD, Error Grid analysis) represent the performance of the device's sensor and algorithm without human intervention in the primary glucose measurement and calculation. The "human-in-the-loop" aspect comes into play with treatment decisions, but the device's numerical output itself is an "algorithm only" performance metric.

7. The Type of Ground Truth Used:

The ground truth used for validating the Dexcom G7's accuracy would be concurrent blood glucose measurements obtained from a YSI Glucose Analyzer (or an equivalent traceable clinical chemistry laboratory method). This is considered the gold standard for glucose measurement in clinical studies.

8. The Sample Size for the Training Set:

The document does not explicitly state the sample size for the training set. For data-driven medical devices like CGMs, the training data (for calibration algorithms, signal processing, etc.) is typically derived from extensive internal research and development studies, often involving hundreds to thousands of sensor wear periods and millions of data points collected over years. This data would encompass a wide range of glucose values, patient populations, and physiological conditions to ensure the algorithm is robust.

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

The ground truth for the training set would be established in the same manner as the test set: through concurrent, highly accurate laboratory reference blood glucose measurements (YSI) taken alongside CGM readings. This ensures that the algorithm learns from precise and reliable reference data. These training data sets often include induced hypoglycemia and hyperglycemia to ensure algorithm performance across the full dynamic range of glucose.

Ask a Question

Ask a specific question about this device

K Number

K221259

Device Name

Manufacturer

Date Cleared

2022-07-29

(88 days)

Product Code

QBJ,QDK,QII

Regulation Number

Type

Panel

Dexcom G6 Glucose Program Continuous Glucose Monitoring (CGM) System

Reference & Predicate Devices

K201328,K203089,K191833

Intended Use

The Dexcom G6 Continuous Glucose Monitoring System (Dexcom G6 System) is a real time, continuous glucose monitoring device indicated for the management of diabetes in persons age 2 years and older.

The Dexcom G6 System is intended to replace fingerstick blood glucose testing for diabetes treatment decisions. Interpretation of the Dexcom G6 System results should be based on the glucose trends and several sequential readings over time. The Dexcom G6 System also aids in the detection of episodes of hyperglycemia and hypoqlycemia, facilitating both acute and long-term therapy adjustments.

The Dexcom G6 Glucose Program Continuous Glucose Monitoring System (Dexcom Glucose Program System) is a real time, continuous glucose monitoring device indicated for the management of diabetes in persons 2 years and older.

The Dexcom Glucose Program System is intended to replace fingerstick blood ducose testing for diabetes treatment decisions for persons with diabetes who are not at significant risk of severe hypoglycemia. Interpretation of the Dexcom Glucose Program System results should be based on the glucose trends and several sequential sensor readings over time. The Dexcom Glucose Program System also aids in the detection of episodes of hyperglycemia and hypoglycemia, facilitating long-term therapy adjustments.

The Dexcom Glucose Program System is also intended to autonomously communicate with digitally connected devices. The Dexcom Glucose Program System can be used alone or in conjunction with these digitally connected devices or services for the purpose of managing diabetes.

The Dexcom G6 Professional Continuous Glucose Monitoring System (Dexcom G6 Pro System) is a real time continuous glucose monitoring device indicated for the management of diabetes in persons age 2 years and older in a home environment while under the supervision of a healthcare professional. The Dexcom G6 Pro System is intended to replace fingerstick blood qlucose testing for diabetes treatment decisions. Interpretation of the real-time Dexcom G6 Pro System results should be based on the glucose trends and several sequential readings over time.

The Dexcom G6 Pro System may also be used as a retrospective glucose recording device indicated for assessing glycemic variability in persons age 2 years and older in a home environment while under the supervision of a healthcare professional. Retrospective interpretation of data recorded by the Dexcom G6 Pro System should be conducted solely by a healthcare professional.

The Dexcom G6 Pro System aids in detecting glucose excursions facilitating care plan adjustments. The Dexcom G6 Pro System is also intended to interface with digitally connected devices. The Dexcom G6 Pro System can be used alone or in coniunction with these digitally connected medical devices for managing diabetes or assessing glycemic variability.

Device Description

The proposed Dexcom G6 CGM System, Dexcom G6 Glucose Program CGM System, and Dexcom G6 Professional CGM System are based on the same physical principles and fundamental design as the predicate for each respective system but includes a modified adhesive patch. The adhesive patch adheres the transmitter holder to the user's body. The Dexcom G6 CGM System, Dexcom G6 Glucose Program CGM System, and the Dexcom G6 Professional CGM System are designed to function as intended with either the proposed or current adhesive patch. The proposed adhesive patch has the same form, fit, and function as the commercial adhesive patch and, from the users' perspective, functions identically.

Dexcom G6 CGM System: The Dexcom G6 Continuous Glucose Monitoring System is an interoperable connected device that measures and displays glucose values for patients with diabetes. The G6 CGM System consists of three main components: a sensor/applicator, a Bluetooth Low Energy (BLE) transmitter, and a BLE enabled display device (receiver and/or mobile application). The sensor is a small and flexible wire, which is inserted by the applicator into subcutaneous tissue where it converts glucose into electrical current. The sensor has an expected wear time of up to 10 days. The transmitter is connected to the sensor and is worn on the body. The transmitter samples the electrical current produced by the sensor and converts these measurements into estimated glucose values (EGV) using an onboard algorithm. The transmitter sends glucose data to the receiver and/or mobile app, which displays the current glucose reading (updated every 5 minutes) and glucose trends. The receiver and/or mobile app also alerts the user when glucose levels are outside of a target zone and when other important system conditions occur. The G6 CGM System is designed to be used alone or in conjunction with digitally connected devices. The G6 CGM System can communicate Estimated Glucose Values (EGV) and other information wirelessly and securely to and from these digitally connected devices.

Dexcom G6 Glucose Program CGM System: The Dexcom G6 Glucose Program Continuous Glucose Monitoring System measures and displays glucose values and trends for patients with diabetes who are not at significant risk of severe hypoglycemia. The system is factory calibrated and provides continuous qlucose readings at five-minute intervals for up to ten days of use. The system consists of a sensor/applicator, a Bluetooth Low Energy (BLE) transmitter, and a mobile CGM software module (app module) that is embedded within a third-party program provider's mobile app (host app).

The sensor is a small and flexible wire, which is inserted by the applicator into subcutaneous tissue where it converts glucose into electrical current. The transmitter is connected to the sensor and is worn on the body. The transmitter samples the electrical current produced by the sensor and converts these measurements into estimated glucose values (EGV) using an onboard algorithm. The transmitter sends glucose data to the app module, which displays the current glucose reading (updated every 5 minutes) and glucose trends. The app module does not include any glucose related alarm or alerts but will alert the user when important system conditions occur.

Dexcom G6 Professional CGM System: The Dexcom G6 Professional CGM System is a continuous glucose monitor that offers an introduction to CGM for users who would benefit from the supervision of their qualified Healthcare Professional (HCP) during early or initial use of CGM.

The G6 Professional CGM System consists of three main components: a sensor/applicator delivery system, a transmitter, and a mobile application (app). The sensor is a small and flexible wire inserted into subcutaneous tissue where it converts glucose into electrical current. The G6 Professional CGM System transmitter is connected to the sensor and is worn on the body. It samples the electrical current produced by the sensor and converts these measurements into glucose readings using an onboard algorithm. The G6 Professional CGM System transmitter's firmware includes an auto-start feature which enables the transmitter to start a session immediately upon attachment of the transmitter to the on-body wearable. The G6 Professional CGM System transmitter can be used as a retrospective CGM data logger, and it can also send real-time estimated glucose values to the G6 Mobile Application. The HCP elects which type of CGM session the patient receives (retrospective vs. real-time). The G6 Professional CGM System transmitter firmware supports a single-use 10-day sensor session per transmitter. The G6 Mobile Application displays the current glucose reading (updated every 5 minutes) and qlucose trends (up to 24 hours) from the transmitter.

AI/ML Overview

This document, a 510(k) Premarket Notification from the FDA, details the equivalence of several Dexcom G6 Continuous Glucose Monitoring (CGM) Systems to previously cleared predicate devices. The core change in the subject devices compared to their predicates is a modified adhesive patch.

Therefore, the acceptance criteria and study proving the device meets these criteria would primarily relate to the performance of the device with this new adhesive patch, demonstrating it does not adversely affect the system's safety and effectiveness.

Based on the provided text, here's a breakdown of the requested information:

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

The document states that the proposed devices performed "according to their respective specifications" and "met their technological and performance criteria which have not changed from the predicate devices." Because the submission is based on substantial equivalence due to a minor change (adhesive patch), specific performance metrics and their acceptance criteria are not explicitly detailed in this summary; rather, the underlying assumption is that the performance remains consistent with the already-cleared predicates.

Therefore, the table would look like this:

Acceptance Criteria (Not explicitly detailed, assumed to be consistent with predicate performance)	Reported Device Performance
Demonstrated safety and effectiveness of the CGM system for its intended use.	Performed according to respective specifications.
Maintained technological and performance criteria of the predicate devices.	Met technological and performance criteria.
Adhesive patch does not adversely impact safety or effectiveness.	Substantially Equivalent with no adverse impact on safety or effectiveness.

For a CGM system, typical performance metrics would include:

Accuracy (e.g., MARD - Mean Absolute Relative Difference): How close the CGM readings are to reference blood glucose values.
Time in Range (TIR): Percentage of time glucose readings are within a target range.
Detection of Hypo/Hyperglycemia: Ability to accurately identify low and high glucose events.
Sensor Wear Time: The specified duration the sensor can be worn (stated as up to 10 days).
Alarm/Alert Performance (where applicable): Timeliness and accuracy of alerts.
Adhesion Performance: Ensuring the new patch maintains the sensor on the body for the intended wear time without causing adverse skin reactions.

However, these specific quantitative criteria are not provided in this 510(k) summary, as the submission focuses on the impact of the adhesive patch change.

2. Sample size used for the test set and the data provenance (e.g., country of origin of the data, retrospective or prospective)

The document mentions that the devices "were verified and validated according to Dexcom's internal design control process." It does not provide explicit details about the sample sizes used in the testing or the specific provenance of the data (country of origin, retrospective/prospective). This level of detail is typically found in the full 510(k) submission, not the summary.

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. For CGM systems, the "ground truth" for glucose values is typically established using a highly accurate laboratory reference method (e.g., YSI analyzer) for blood samples, rather than human expert consensus, as it's a quantitative measurement.

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

This information is not provided as it's not relevant for device performance evaluation of a CGM system based on quantitative measurements against a reference. Adjudication methods like 2+1 or 3+1 are common in diagnostic image interpretation studies where human expert consensus on image findings serves as ground truth.

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

An MRMC study is relevant for AI-assisted image interpretation. This document describes a Continuous Glucose Monitoring (CGM) system, which is a sensor-based medical device that measures glucose levels directly. It is not an AI-assisted diagnostic imaging device, and therefore, an MRMC comparative effectiveness study would not be applicable or performed for this device.

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

For a CGM system, the device (sensor + transmitter + algorithm) generates glucose values autonomously. The "human-in-the-loop" is the user who interprets these values for diabetes management. The system is inherently designed to provide standalone glucose readings. The "Estimated Glucose Value (EGV) using an onboard algorithm" indicates standalone algorithmic performance. The document states: "Interpretation of the Dexcom G6 System results should be based on the glucose trends and several sequential readings over time" and the system "aids in the detection of episodes of hyperglycemia and hypoglycemia, facilitating both acute and long-term therapy adjustments."

This confirms that the algorithm provides direct, actionable results, and its performance is evaluated as a standalone system generating glucose values.

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

The ground truth for a CGM system is typically established by concurrent blood glucose measurements obtained using a highly accurate laboratory reference method (e.g., YSI glucose analyzer) from venous blood draws. While not explicitly stated in this summary, this is the standard for CGM validation. The device's "Estimated Glucose Values (EGV)" are compared against these laboratory reference values.

8. The sample size for the training set

This document describes a 510(k) submission for a device change (adhesive patch) to existing, cleared devices. The core algorithms (e.g., "Joint Probability Algorithm") are already established and cleared from prior submissions (the predicates). Therefore, specific training set information for the algorithms themselves is not detailed here, as the algorithms are not being fundamentally changed or retrained. The testing mentioned in this document (verification and validation) would be focused on demonstrating that the new adhesive patch does not negatively impact the already-established performance of the system.

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

As in point 8, the algorithms are inherited from previously cleared devices. For the original development of these algorithms, the ground truth would have been established through extensive clinical studies involving human subjects, where CGM readings were compared against simultaneous, high-accuracy laboratory reference blood glucose measurements. The process would involve collecting paired data points (CGM reading and reference blood glucose) across a wide range of glucose values, physiological states, and patient populations.

Ask a Question

Ask a specific question about this device

K Number

K203089

Device Name

Manufacturer

Date Cleared

2021-08-17

(308 days)

Product Code

QDK

Regulation Number

Type

Panel

Dexcom G6 Continuous Glucose Monitoring (CGM) System

Reference & Predicate Devices

K200876

Intended Use

The Dexcom Glucose Program System is intended to replace fingerstick blood glucose testing for diabetes treatment decisions for persons with diabetes who are not at significant risk of severe hypoglycemia. Interpretation of the Dexcom Glucose Program System results should be based on the glucose trends and several sequential readings over time. The Dexcom Glucose Program System also aids in the detection of episodes of hyperglycemia, facilitating long-term therapy adjustments.

Device Description

The Dexcom G6 Glucose Program Continuous Glucose Monitoring (CGM) System measures and displays glucose values and trends for patients with diabetes who are not at significant risk of severe hypoglycemia. The system is factory calibrated and provides continuous glucose readings at five-minute intervals for up to ten days of use. The system consists of a sensor/applicator, a Bluetooth Low Energy (BLE) transmitter, and a BLE enabled mobile CGM display.

The sensor is a small and flexible wire, which is inserted by the applicator into subcutaneous tissue where it converts glucose into electrical current. The transmitter is connected to the sensor and is worn on the body. The transmitter samples the electrical current produced by the sensor and converts these measurements into estimated glucose values (EGV) using an onboard algorithm. The transmitter sends glucose data to the mobile CGM displays the current glucose reading (updated every 5 minutes) and glucose trends. The mobile CGM display does not include any glucose related alarm or alerts but will alert the user when important system conditions occur.

The subject of this submission is a change to the primary display mobile device software. Compared to the predicate device which uses a standalone mobile CGM application, the proposed Glucose Program System uses a new mobile CGM software module (app module) that is embedded within a third party program provider's mobile app (host app).

This change is to help encourage CGM retention for those not at significant risk of severe hypoglycemia by providing Dexcom Glucose Program CGM functions and program provider's functions on a single unified mobile application. The app module is designed as a finished sovereign software that interacts directly with smart device hardware/operating system and maintains the same core CGM functionality as the predicate device, independent of the host app. The proposed Glucose Program System uses the same sensor/applicator and transmitter as the predicate device, with only changes to the mobile CGM software.

The change to the primary display mobile device software described in this submission does not impact the standalone G6 mobile application used in the G6 CGM System (last cleared K200876) and only affects the Glucose Program System which is designed for payor-sponsored, valuebased health programs.

AI/ML Overview

The document describes the Dexcom G6 Glucose Program Continuous Glucose Monitoring (CGM) System and its comparison to a predicate device (K200876). The core of the information provided shows that the new device, referred to as the "subject device," is substantially equivalent to the predicate, with the primary change being a modification to the mobile display software.

Here's an analysis of the provided text in relation to the requested information:

1. Table of acceptance criteria and the reported device performance

The document does not explicitly present a table of acceptance criteria with corresponding reported device performance values. Instead, it asserts "substantial equivalence" to a predicate device, implying that the subject device meets the same performance standards as the predicate. The description highlights that the subject device "has the same core CGM functionality and meets the same system requirements as the predicate G6 Glucose Program System."

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

The document does not specify a sample size for a test set that was used to prove the device meets acceptance criteria. It mentions that "All testing referenced in the predicate device (K200876) in accordance with special controls for integrated continuous glucose monitors remain applicable." This suggests that new clinical performance testing with human subjects was not conducted for this specific submission because the underlying hardware and core algorithms are the same as the predicate.

Regarding data provenance (country of origin, retrospective/prospective), this information is also not provided as new clinical trials were not detailed for this submission.

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

This information is not provided in the document. As stated above, new clinical performance data from human subjects is not detailed for this submission, as the claim is based on substantial equivalence to a predicate device for which such testing would have been performed.

4. Adjudication method for the test set

This information is not provided in the document.

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 type of study is not applicable to this device. The Dexcom G6 Glucose Program CGM System is a continuous glucose monitoring device, not an AI-assisted diagnostic imaging or interpretation tool that would involve "human readers" in the context of an MRMC study.

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

The device itself is a continuous glucose monitoring system with an onboard algorithm that converts sensor measurements into estimated glucose values. The document states, "The transmitter samples the electrical current produced by the sensor and converts these measurements into estimated glucose values (EGV) using an onboard algorithm." This confirms a standalone algorithm that processes raw sensor data. The claim of substantial equivalence implies that the performance of this standalone algorithm has been previously validated as part of the predicate device's clearance.

7. The type of ground truth used

For a glucose monitoring device, the typical "ground truth" for evaluating accuracy would be laboratory reference standard blood glucose measurements (e.g., from a YSI analyzer). While the document doesn't explicitly state the ground truth used for this specific submission, it relies on the predicate device's clearance, which would have established accuracy against such a standard. The section "Glucose Value Estimation Algorithm" mentions an "Optimized Joint Probability Algorithm with improved data availability," which would be evaluated against such ground truth.

8. The sample size for the training set

The document does not provide information regarding the sample size for the training set. This is likely because the core algorithm remains the same as the predicate device, and new training data for algorithm development is not part of this specific submission.

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

The document does not provide this information. It only states that the "Optimized Joint Probability Algorithm" is used, implying that any training and its associated ground truth establishment would have occurred during the development of this algorithm (likely for the predicate device).

Ask a Question

Ask a specific question about this device

K Number

K201328

Device Name

Manufacturer

Date Cleared

2021-07-15

(422 days)

Product Code

Regulation Number

Type

Panel