Search Results

Ask a specific question about this device

The Colormate III™ is to be used as an aid to the physician in monitoring the status of newborn babies for the development of hyperbilirubinemia. Following the physician's examination within the first hours after birth, newborn babies are initially measured and periodically monitored by the Colormate III™ for incremental changes in the yellow content of the skin color as compared to the baseline Colormate III™ measurements. Babies with Colormate III™ test results indicative of hyperbilirubinemia are to be re-evaluated by the attending physician for appropriate patient management.

Colormate III™ (Colormate III or CM III) is a computer assisted noninvasive transcutaneous bilirubinometer which, through colorimetric technology, illuminates the skin of newborn babies and measures the yellow content of the skin color. The incremental changes of these color readings measured over time are compared to the newborn's baseline readings. These data are then automatically processed to provide a numerical index of estimated bilirubin count in mg/dl that has been shown to correlate with total serum bilirubin within a clinically useful range. The Colormate III, like other colorimeters, generates a set of tristimulus color coordinates for each measurement. It then performs a series of algorithms to produce final measurement results.

The provided document describes the Colormate III™, a transcutaneous bilirubinometer, and a clinical study conducted to demonstrate its performance. Here's a breakdown of the requested information:

1. Table of Acceptance Criteria and Reported Device Performance

The document doesn't explicitly state "acceptance criteria" as a set of predefined thresholds that the device must meet. Instead, it presents performance metrics against visual assessment as a comparator, focusing on correlation, bias, and sensitivity. We can infer the performance goals from the reported results.

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

• Total Study Population: 1317 newborns
• Test Set for Performance Comparison (Colormate III vs. Visual Assessment vs. Serum Bilirubin): 171 premature and full-term newborns who had concurrent Colormate III measurements, clinical visual estimates, and blood serum tests. These 171 newborns yielded 360 concurrent measurements.
• Data Provenance: The document does not explicitly state the country of origin. However, the FDA 510(k) submission context implies the study was conducted to support marketing in the U.S. The study was a prospective clinical trial.

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

• Number of Experts: The document mentions "the physician" and "clinicians" in the context of visual assessment, implying multiple physicians were involved in the visual estimates. However, it does not specify the exact number of physicians or their qualifications (e.g., years of experience, specialty).
• Qualifications: Not specified. It can be inferred they were attending physicians or clinicians experienced in neonatal care.

4. Adjudication Method for the Test Set

• The document does not describe an explicit adjudication method for reconciling discrepancies between multiple visual assessments by clinicians. The study compares the Colormate III to "the clinicians' visual assessment," suggesting either a single visual assessment per case or an implicit consensus, but no formal adjudication process is detailed.

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

• Type of Study: This was a comparative effectiveness study comparing the Colormate III (an AI-assisted device in its processing of colorimetric data) against human visual assessment without AI assistance and against the gold standard of serum bilirubin.
• Effect Size of Human Readers Improve with AI vs. Without AI Assistance:
  • The study does not evaluate "human readers improve with AI vs. without AI assistance" in the typical MRMC sense where human readers interpret images with and without AI output.
  • Instead, it directly compares the device's algorithmic output ("Colormate III") against human visual assessment.
  • The "effect size" is presented by comparing the correlation coefficient with serum bilirubin:
    • Colormate III: r = 0.90
    • Clinicians' visual assessment: r = 0.67
  • And by sensitivity for detecting hyperbilirubinemia (>12.0 mg/dl):
    • Colormate III: 89%
    • Clinicians' visual assessment: 61%
  • The Colormate III shows significantly higher accuracy and sensitivity than visual assessment.

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

• Yes, a standalone performance assessment was done. The "Colormate III" performance metrics (correlation r=0.90, bias, sensitivity=89%) represent the algorithm's performance when measuring skin color and processing it to provide a numerical index of estimated bilirubin. The device's output is then used by a physician, but the reported performance metrics focus on the device's ability to correlate with blood serum levels.

7. Type of Ground Truth Used

• The primary ground truth used for evaluating the Colormate III and visual assessment was total serum bilirubin concentration (BRC), measured via blood tests.
• The document also describes how "a cutoff of 12.0 mg/dl of bilirubin" was used as an "acceptable level of clinical concern" for calculating sensitivity.

8. Sample Size for the Training Set

• The document states: "Prior to undertaking a pivotal clinical trial to evaluate the Colormate III's performance, two phases of studies were conducted to validate the clinical approach and test criteria and determine calibration algorithms for subsequent validation."
• This indicates that a training/calibration phase was performed before the pivotal clinical trial, but the sample size for these training/calibration studies is not explicitly provided.

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

• The document implies that the ground truth for the training/calibration phase was also established using serum bilirubin blood tests in conjunction with Colormate III measurements. It states these studies were used to "determine calibration algorithms," which would necessitate comparing device readings to a reference standard (serum bilirubin) to build or refine its internal models. However, detailed methodology for training ground truth establishment is not provided.

Ask a specific question about this device