Search Results

The CONTOUR®NEXT EZ blood glucose monitoring system is an over the counter (OTC) device utilized for self-testing by persons with diabetes at home for the quantitative measurement of glucose in whole blood, is for single-patient use only, and should not be shared. The CONTOUR NEXT EZ blood glucose monitoring system is indicated for use with fresh fingertip capillary whole blood samples. The clinical utility of this device is to aid in monitoring the effectiveness of your diabetes control program.

The CONTOUR® NEXT EZ blood glucose monitoring system is not intended for use for the diagnosis of or screening for diabetes mellitus and is not intended for use on neonates.

The CONTOUR® NEXT test strips are intended for self-testing by persons with diabetes for the quantitative measurement of glucose in whole blood samples from 20 to 600 mg/dL.

The CONTOUR®NEXT control solutions are aqueous glucose solutions intended for use in self-testing by people with diabetes as a quality control check.

Device Description

The Contour NEXT EZ Blood Glucose Monitoring System consists of:

Contour NEXT EZ Blood Glucose Meter
Contour NEXT Blood Glucose Test Strips
Contour NEXT Control Solutions

AI/ML Overview

Acceptance Criteria and Device Performance Study for Contour NEXT EZ Blood Glucose Monitoring System

1. Table of Acceptance Criteria and Reported Device Performance

Performance Characteristic	Acceptance Criteria	Reported Device Performance
Accuracy	Non-clinical (ISO 15197 Section 7.3/7.4.1): A minimum of 95% of individual glucose results to fall within ±15 mg/dL of YSI analyzer results at glucose < 75 mg/dL, and within ±20% at glucose ≥ 75 mg/dL.	Non-clinical: 100% of data fell within ±20% or ±15 mg/dL. More precisely, 100% within ±10 mg/dL (<75mg/dL) and 99.2% within ±10% (≥ 75 mg/dL).

	Clinical (ISO 15197:2003 Section 8 internal criteria): 95% of individual glucose results to fall within ±15 mg/dL of manufacturer's method results at glucose < 75 mg/dL, and within ±20% at glucose ≥ 75 mg/dL.	Clinical: 100% of results meet the above accuracy requirements. More precisely, 100% within ±10 mg/dL (<75 mg/dL) and 96.3% within ±10% (≥ 75 mg/dL).
Precision (Repeatability)	Internal Acceptance Criterion: Repeatability test must perform within the established accuracy requirements of Cpk ≥ 0.65.	All Cpk values were greater than 0.65, ranging from 3.7 to 5.9 across different glucose intervals.
Linearity/Reportable Range	System Linearity: (1) At least 95% of assay results within ±10% for samples ≥100 mg/dL or within ±10 mg/dL for samples <100 mg/dL relative to YSI plasma glucose values. (2) Linear regression slope from 0.93 to 1.07, intercept from -8.9 to +9.4, correlation coefficient > 0.990.	100% of 717 data points were within ±10 mg/dL (<100mg/dL) or ±10% (≥100mg/dL) of YSI reference. Regression equation: y = 0.967(x) + 1.246 (slope and intercept within range), r = 0.998 (>0.990).
Detection Limit	For blood with extreme glucose levels, the meter must display "LO" or "HI" error messages.	All 72 readings at 5 mg/dL reported "LO", and all 288 readings at 900 mg/dL and higher reported "HI". Pass.
Analytical Specificity (Hematocrit)	Assay bias within 10 mg/dL (glucose <100mg/dL) or 10% (glucose ≥100mg/dL) compared to YSI references or 40% Hct whole blood samples. No trend of increasing imprecision or bias with increasing hematocrit.	Results indicate assay bias is within the specified limits. No trend of increasing imprecision or bias with increasing sample hematocrit.

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

Non-clinical Accuracy Study:

Sample Size: 100 blood samples tested in replicates, resulting in 300 data points for the first replicate (6 meters x 3 lots x ~16 samples/lot, adjusted to 100 samples total).
Data Provenance: Not explicitly stated, but implies laboratory-controlled and prepared blood samples.

Clinical Accuracy Study:

Sample Size: 115 subjects.
Data Provenance: Clinical site, fresh fingertip capillary blood samples (prospective study).

Precision (Repeatability) Study:

Sample Size: Venous blood tested at five glucose concentration ranges. 10 meters, 10 replicates per meter for one test strip lot (n=100 for each blood sample level, total unknown but implies 500 data points across 5 intervals).
Data Provenance: Not explicitly stated, but implies laboratory-controlled venous blood.

Linearity/Reportable Range Study:

Sample Size: Eight aliquots of fresh venous blood, each at a different glucose concentration. 30 replicate data points per aliquot for each of three test strip lots (8 glucose levels * 30 replicates * 3 lots = 720 data points, with 717 used for analysis after removing data >600mg/dL).
Data Provenance: Laboratory-controlled fresh venous blood.

Detection Limit Study:

Sample Size: 8 Contour meters (3 readings per meter) with 3 test strip lots; 72 readings for low glucose (5 mg/dL) and 288 readings for high glucose (900-1800 mg/dL).
Data Provenance: Not explicitly stated, but implies laboratory-controlled blood samples.

Analytical Specificity (Hematocrit) Study:

Sample Size: Fresh venous blood samples (15% to 65% hematocrit), adjusted to two plasma glucose concentrations (40 mg/dL and 550 mg/dL). 12 replicates per sample collected on six Contour NEXT meters for each of three test strip lots.
Data Provenance: Laboratory-controlled fresh venous blood samples.

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

Ground Truth Method: The ground truth for all performance studies (Accuracy, Precision, Linearity, Detection Limit, Analytical Specificity) was established by comparison to the YSI 2300 STAT PLUS glucose analyzer.
Expert involvement: The document does not specify direct involvement of human "experts" to establish the ground truth for individual test set samples. Instead, the YSI appliance serves as the reference standard. The traceability section states the YSI analyzer is traceable to the hexokinase method, which was developed collaboratively by the FDA, CDC, NIST, and AACC. This implies expert validation of the reference method, rather than individual sample adjudication by experts.

4. Adjudication Method for the Test Set

None specified for individual samples. The YSI 2300 STAT PLUS glucose analyzer provided a single reference value for each blood sample, acting as the definitive ground truth. There was no mention of multiple expert reads or an adjudication process for discrepancy resolution for the test samples themselves.

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

No Multi-Reader Multi-Case (MRMC) comparative effectiveness study was described. The studies focused on the performance of the device itself against a reference standard or internal criteria. The clinical study involved users self-testing, but it was not designed as an MRMC study to compare human readers' performance with and without AI assistance.

6. Standalone (Algorithm Only) Performance Study

Yes, standalone performance was done. The "Accuracy" section under "Performance Data" describes "System Accuracy Evaluation" where "Six Contour NEXT EZ meters, three lots of test strips and 100 blood samples were tested in replicates." The reported results are directly from the device's measurement compared to the YSI reference, indicating standalone algorithm performance. Similarly, all other non-clinical studies (Precision, Linearity, Detection Limit, Analytical Specificity) evaluate the inherent performance of the device's system and algorithm.

7. Type of Ground Truth Used

YSI 2300 STAT PLUS glucose analyzer results. This is stated to be traceable to the hexokinase method, which is a recognized laboratory reference method for glucose measurement. The hexokinase method itself is "incorporated in a Bayer procedure that utilizes NIST Standard Reference Material 917, dry D-glucose." This indicates a highly accurate and standardized laboratory reference.

8. Sample Size for the Training Set

Not explicitly stated. The document is a 510(k) summary for a blood glucose monitoring system, which typically relies on established electrochemical principles rather than machine learning algorithms that require large training sets in the same way AI-powered image analysis tools do. The "measurement algorithm" was modified, but the detail on its development and any associated "training set" (in the sense of machine learning) is not provided in this summary. Instead, the focus is on verification and validation studies using test sets.

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

Not applicable/Not explicitly stated in the context of machine learning training data. As mentioned above, the device likely relies on a deterministic algorithm for glucose measurement based on electrochemical detection. Therefore, the concept of a "training set" with established ground truth as used in machine learning is not directly addressed. The device is verified and validated against established laboratory reference methods and internal criteria.

Ask a Question

Ask a specific question about this device

Page 1 of 1