(66 days)
With PillCam SB 2 and SB 3 Capsules
The PillCam Capsule Endoscopy System with a PillCam SB 2 and SB 3 capsules is intended for visualization of the small bowel mucosa.
- o PillCam Capsule Endoscopy System with PillCam SB 2 and SB 3 capsules may be used in the visualization and monitoring of lesions that may indicate Crohn's disease not detected by upper and lower endoscopy.
- PillCam Capsule Endoscopy System with PillCam SB 2 and SB 3 capsules ● may be used in the visualization and monitoring of lesions that may be a source of obscure bleeding (either overt or occult) not detected by upper and lower endoscopy.
- PillCam Capsule Endoscopy System with PillCam SB 2 and SB 3 capsules may be used in the visualization and monitoring of lesions that may be potential causes of iron deficiency anemia (IDA) not detected by upper and lower endoscopy.
The Suspected Blood Indicator (SBI) feature is intended to mark frames of the video suspected of containing blood or red areas.
The PillCam Capsule Endoscopy System with PillCam SB 2 and SB 3 capsules may be used as a tool in the detection of abnormalities of the small bowel and is intended for use in adults and children from two years of age.
With PillCam ESO 3 Capsule
The PillCam Capsule Endoscopy System with PillCam ESO 3 capsules is intended for the visualization of esophageal mucosa in adults and children from 18 years of age..
With PillCam UGI Capsule
The PillCam UGI capsule endoscopy system is intended for visualization of the upper gastrointestinal tract (esophagus, stomach, duodenum). It may be used for visualization of blood in the upper gastrointestinal tract (esophagus, stomach, duodenum) in patients who are hemodynamically stable and at least 18 years of age.
With PillCam COLON Capsule
The PillCam COLON 2 capsule endoscopy system is intended to provide visualization of the colon. It may 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. In addition, it is intended for detection of colon polyps in patients with evidence of gastrointestinal bleeding of lower GI origin. This applies only to patients with major risks for colonoscopy or moderate sedation, but who could tolerate colonoscopy and moderate sedation in the event a clinically significant colon abnormality was identified on capsule endoscopy.
The RAPID Web is a client server based version of the RAPID video review and report creation components of the cleared RAPID 8 software. The RAPID Web is designed to allow users to review and analyze videos created using the cleared RAPID software, and to create reports over a private intranet network or over a secure internet connection. The PillCam capsule studies are created at workflow-relevant locations and stored at a central repository. The RAPID Web server is installed on a central server in an intranet or internet environment with access to the study repository. After the capsule endoscopy procedure was conducted using the cleared RAPID software, Given Workstation/User PC, PillCam capsule, and DR 3 PillCam recorder, end users will use an internet browser with authorization login to access, review, and manage the studies in the repository using workstation, user PC, a laptop or an Apple™ (screen size > 9.7").
The provided text describes the 510(k) submission for RAPID Web, comparing it to its predicate device, RAPID 8.0. The core of the submission revolves around demonstrating that RAPID Web is substantially equivalent to RAPID 8.0, and thus, RAPID Web's "performance" is implicitly tied to how well it replicates the established performance of RAPID 8.0.
Here's an analysis of the acceptance criteria and study as presented in the document:
1. A table of acceptance criteria and the reported device performance
The document does not explicitly state numerical acceptance criteria in a table format for RAPID Web's diagnostic performance. Instead, it states the acceptance criterion informally as "RAPID 8.0 met the pre-specified sensitivity" in the context of a comparative performance study. The reported performance is that the analysis of "matched and un-matched pathologies" showed this criterion was met.
| Acceptance Criteria (Stated or Implied) | Reported Device Performance |
|---|---|
| RAPID Web must perform equivalently to RAPID 8.0 in identifying pathologies. (Implied: RAPID 8.0's established diagnostic sensitivity for pathology identification, including "blood or red areas" marked by SBI, is maintained.) | A comparative performance study showed that "RAPID 8.0 met the pre-specified sensitivity" when RAPID Web and RAPID 8.0 were used to identify pathologies in the same video and video segments. |
| Software validation performed per FDA guidance. | "the software was validated and documented per FDA's software guidance document." |
| No new issues of safety or effectiveness introduced by technological differences. | "The minor differences in indications do not alter the intended diagnostic use of the device and do not affect its safety and effectiveness when used as labeled." "In addition, the minor technological differences between the RAPID Web and its predicate devices raise no new issues of safety or effectiveness." |
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 Test Set: The document states that a "comparative performance study wherein trained professionals identified pathologies in the same video and video segments that we displayed using RAPID Web and RAPID 8.0 was performed." However, it does not specify the number of cases (videos or video segments) used in this test set.
- Data Provenance: The document does not specify the country of origin of the data or whether the data was collected retrospectively or prospectively.
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 implies that "trained professionals" were involved in identifying pathologies. However, it does not specify the number of experts used to establish the ground truth or their specific qualifications (e.g., years of experience, medical specialty). The study design appears to be a comparison of user performance with two different software versions, rather than a separate ground truth establishment phase by independent experts for the test set itself. The "ground truth" for this comparative study seems to be derived from the consensus or established interpretations of "trained professionals" using the predicate device.
4. Adjudication method (e.g., 2+1, 3+1, none) for the test set
- The document does not describe an explicit adjudication method (like 2+1 or 3+1) for the test set. The study compares "matched and un-matched pathologies," suggesting a direct comparison of findings between the two software versions, likely relying on individual assessments by the trained professionals.
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 described is a "comparative performance study" where "trained professionals identified pathologies in the same video and video segments that we displayed using RAPID Web and RAPID 8.0." This sounds like a multi-reader, multi-case study (MRMC) in the sense that multiple readers would use the two different software versions on the same cases.
- However, this is not a study on human readers improving with AI vs. without AI assistance. RAPID Web and RAPID 8.0 are video review and analysis software. The "Suspected Blood Indicator (SBI) feature" is mentioned as "intended to mark frames of the video suspected of containing blood or red areas," which is an algorithmic assistance feature. The study assessed if this feature, within the new web-based context, maintained its performance from the predicate.
- The document does not report any effect size demonstrating improvement of human readers with either RAPID Web or RAPID 8.0 over a non-assisted reading scenario. The study aims to demonstrate substantial equivalence between the two software versions.
6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done
- The document does not describe a standalone algorithm-only performance study. The performance evaluation is explicitly framed as a "comparative performance study wherein trained professionals identified pathologies," indicating human-in-the-loop performance. The SBI feature is an algorithm, but its performance is assessed within the context of human review.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)
- The ground truth for the "pathologies" identified in the comparative study appears to be based on the interpretations of "trained professionals" themselves, as they "identified pathologies." The comparison is between the findings obtained using RAPID Web and those obtained using RAPID 8.0, with RAPID 8.0's established performance serving as the reference. There is no mention of an independent, higher-standard ground truth like biopsy/pathology reports or long-term clinical outcomes data being used to validate the accuracy of these identified pathologies in this particular comparative study. The study's goal is equivalence, not re-proving clinical accuracy against an external gold standard for the new platform.
8. The sample size for the training set
- The document does not provide any information about a training set or its sample size. This submission is for a new version of existing software (RAPID Web being a client-server version of RAPID 8.0), not necessarily a new AI model that would require a distinct training phase described in the submission. The "Suspected Blood Indicator (SBI) feature" is part of the established functionality from the predicate device, not a new AI algorithm being introduced or trained for RAPID Web.
9. How the ground truth for the training set was established
- As no training set is described in the document, there is no information on how ground truth for a training set was established.
§ 876.1300 Ingestible telemetric gastrointestinal capsule imaging system.
(a)
Identification. An ingestible telemetric gastrointestinal capsule imaging system is used for visualization of the small bowel mucosa as an adjunctive tool in the detection of abnormalities of the small bowel. The device captures images of the small bowel with a wireless camera contained in a capsule. This device includes an ingestible capsule (containing a light source, camera, transmitter, and battery), an antenna array, a receiving/recording unit, a data storage device, computer software to process the images, and accessories.(b)
Classification. Class II (special controls). The special control is FDA's guidance, “Class II Special Controls Guidance Document: Ingestible Telemetric Gastrointestinal Capsule Imaging Systems; Final Guidance for Industry and FDA.”