(426 days)
The PillCam COLON 2 Capsule Endoscopy System is indicated to provide visualization of the colon. It is intended to be used for detection of colon polyps in patients after an incomplete optical colonoscopy with adequate preparation, and a complete evaluation of the colon was not technically possible.
The PillCam® COLON 2 capsule endoscopy system includes a single-use ingestible capsule designed to acquire video images during natural propulsion through the digestive system. It is specifically designed to visualize the complex anatomy of the colon. The PillCam COLON 2 capsule is designed to withstand the mechanical forces and chemical environment of the digestive system. The system is comprised of four main subsystems; (1) the ingestible PillCam COLON 2 capsule, (2) the DR 3 PillCam® Recorder, (3) the RAPID® software, and (4) the Given® Workstation.
1. A table of acceptance criteria and the reported device performance
| Acceptance Criteria (Polyp Detection by CCE vs. OC) | Performance for Polyps ≥ 6 mm | Performance for Polyps ≥ 10 mm |
|---|---|---|
| Positive Percent Agreement | 68.8% (95% CI 61.7-75.2%) | 64.9% (95% CI 53.2-75.5%) |
| Negative Percent Agreement | 81.3% (95% CI 77.6-84.6%) | 92.9% (95% CI 90.6-94.8%) |
Note: The document does not explicitly state numerical acceptance criteria values, but rather presents the study's performance characteristics as part of the overall demonstration of effectiveness. The table above reflects the reported performance that supports the device's acceptance.
2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)
- Sample Size for Effectiveness Analysis: 700 subjects successfully completed an investigation with both CCE and OC and were included in the effectiveness analysis.
- Data Provenance: Prospective, multi-center study.
- Country of Origin: 11 enrollment sites in the US and 6 in Israel.
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)
The document states that a "central reader" interpreted the CCE results. For the optical colonoscopy (OC) results, "colonoscopists" evaluated their findings. The number and specific qualifications (e.g., years of experience) of these experts are not explicitly stated in the provided text.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set
The document mentions an initial phase where "colonoscopists were blinded to CCE results when evaluating their OC findings." Following this, if a polyp detected on CCE was not identified by the initial colonoscopy, "the results of the CCE evaluation were unblinded and the colonoscopy was repeated in a second attempt to identify the polyp identified on CCE." This suggests a form of sequential adjudication or unblinding with repeat examination rather than a consensus-based adjudication (like 2+1 or 3+1). The "reference OC polyp chosen for the final determination was the one that was in favor of the device" when there were multiple equally large polyps, which implies a specific rule for handling discrepancies.
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 study was a "multi-center study" comparing CCE (the device) with optical colonoscopy (OC). It primarily evaluated the standalone performance of the CCE system against OC as the ground truth. There is no information provided about a Multi-Reader Multi-Case (MRMC) comparative effectiveness study involving human readers with and without AI assistance. The focus is on the device's ability to detect polyps, not on enhancing human reader performance using the device as an AI assistant.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
Yes, a standalone performance evaluation was conducted. The study "compared CCE with optical colonoscopy (OC) for agreement on absence or presence of colon polyps." The CCE system, including its RAPID software for video viewing and report generation, functions as an algorithm-driven system that outputs video for interpretation. The performance metrics (positive and negative percent agreement) specifically refer to the CCE as a device, implying standalone performance in detecting polyps. The "centralized reader" interpreted the CCE results, but the CCE itself is the classification device.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)
The ground truth used was Optical Colonoscopy (OC) findings, specifically the identification and sizing of colon polyps (≥6 mm or ≥10 mm). OC is considered a clinically acceptable alternative structural imaging method.
8. The sample size for the training set
The provided document does not specify the sample size for a training set. The clinical study described is for evaluating the performance of the already developed device, not for training it.
9. How the ground truth for the training set was established
Since no information on a training set is provided, how its ground truth was established is not detailed in this document.
§ 876.1330 Colon capsule endoscopy system.
(a)
Identification. A prescription, single-use ingestible capsule designed to acquire video images during natural propulsion through the digestive system. It is specifically designed to visualize the colon for the detection of polyps. It is intended for use only in patients who had an incomplete optical colonoscopy with adequate preparation, and a complete evaluation of the colon was not technically possible.(b)
Classification. Class II (special controls). The special controls for this device are:(1) The capsule must be demonstrated to be biocompatible.
(2) Non-clinical testing data must demonstrate the mechanical and functional integrity of the device under physically stressed conditions. The following performance characteristics must be tested and detailed protocols must be provided for each test:
(i) Bite test to ensure that the capsule can withstand extreme cases of biting.
(ii) pH resistance test to evaluate integrity of the capsule when exposed to a range of pH values.
(iii) Battery life test to demonstrate that the capsule's operating time is not constrained by the battery capacity.
(iv) Shelf-life testing to demonstrate that the device performs as intended at the proposed shelf-life date.
(v) Optical testing to evaluate fundamental image quality characteristics such as resolution, field of view, depth of field, distortion, signal-to-noise ratio, uniformity, and image artifacts. A test must be performed to evaluate the potential of scratches, caused by travelling through the gastrointestinal tract, on the transparent window of the capsule and their impact on the optical and color performance.
(vi) An optical safety analysis must be performed based on maximum (worst-case) light exposure to internal gastrointestinal mucosa, and covering ultraviolet, visible, and near-infrared ranges, as appropriate. A mitigation analysis must be provided.
(vii) A color performance test must be provided to compare the color differences between the input scene and output image.
(viii) The video viewer must clearly present the temporal or spatial relationship between any two frames as a real-time lapse or a travel distance. The video viewer must alert the user when the specific video interval is captured at a frame rate lower than the nominal one due to communication errors.
(ix) A performance test evaluating the latency caused by any adaptive algorithm such as adjustable frame rate must be provided.
(x) If the capsule includes a localization module, a localization performance test must be performed to verify the accuracy and precision of locating the capsule position within the colon.
(xi) A data transmission test must be performed to verify the robustness of the data transmission between the capsule and the recorder. Controlled signal attenuation should be included for simulating a non-ideal environment.
(xii) Software validation, verification, and hazards analysis must be provided.
(xiii) Electrical equipment safety, including thermal and mechanical safety and electromagnetic compatibility (EMC) testing must be performed. If the environments of intended use include locations outside of hospitals and clinics, appropriate higher immunity test levels must be used. Labeling must include appropriate EMC information.
(xiv) Information demonstrating immunity from wireless hazards.
(3) The clinical performance characteristics of the device for the detection of colon polyps must be established. Demonstration of the performance characteristics must include assessment of positive percent agreement and negative percent agreement compared to a clinically acceptable alternative structural imaging method.
(4) Clinician labeling must include:
(i) Specific instructions and the clinical and technical expertise needed for the safe use of the device.
(ii) A detailed summary of the clinical testing pertinent to use of the device, including the percentage of patients in which a polyp was correctly identified by capsule endoscopy, but also the percent of patients in which the capsule either missed or falsely identified a polyp with respect to the clinically acceptable alternative structural imaging method.
(iii) The colon cleansing procedure.
(iv) A detailed summary of the device technical parameters.
(v) A detailed summary of the device- and procedure-related complications pertinent to use of the device.
(vi) An expiration date/shelf life.
(5) Patient labeling must include:
(i) An explanation of the device and the mechanism of operation.
(ii) Patient preparation procedure.
(iii) A brief summary of the clinical study. The summary should not only include the percentage of patients in which a polyp was correctly identified by capsule endoscopy, but also the percent of patients in which the capsule either missed or falsely identified a polyp with respect to the clinically acceptable alternative structural imaging method.
(iv) A summary of the device- and procedure-related complications pertinent to use of the device.