(237 days)
The CapsoCam SV-1 video capsule system is intended for visualization of the small bowel mucosa in adults. It may be used as a tool in the detection of abnormalities of the small bowel.
CapsoCam® SV-1 is a single-use, ingestible video capsule that acquires and stores video images in on-board memory while moving through the gastrointestinal tract, propelled by natural peristalsis. The patient retrieves the capsule using the provided retrieval kit, and returns it to the physician who downloads and reviews the images on a personal computer (PC). The capsule is typically excreted within 3 to 30 hours after swallowing.
The CapsoCam® SV-1 video capsule is a panoramic imaging system which includes a panoramic lens system and a multi-array CMOS image sensor. The CapsoCam® has four side-facing cameras arrayed about the circumference of the capsule that together image a full 360° circumference and capture high-resolution color images of the mucosa.
The CapsoCam® SV-1 acquires video images while moving through the patient's gastrointestinal tract. The frame rate is 5 frames per second (FPS) for each of 4 cameras (20 FPS total) the first two hours, and then decreases to 3 FPS per camera (12 FPS total). The initial frame rate is higher because the capsule moves more quickly in the duodenum than in the jejunum or ileum.
After excretion, the capsule is retrieved by the patient using CapsoRetrive, a capsule retrieval system, including a strainer which is placed in the patient's toilet to retrieve the capsule while passing fecal matter and rinse water. The patient then returns the capsule to the clinic, where a technician disinfects the capsule and retrieves the data using the CapsoAccess, Capsule Data Access System (CDAS). The CDAS is comprised of a capsule carrier for holding the capsule opener for cutting off one end of the capsule, and a docking system for forming an electrical contact to the capsule printed circuit board (PCB) through the open end. The data is transferred to the workstation over the link shown schematically in Figure 4 below, from the capsule to the docking system by a serial peripheral interface (SPI) and to the workstation by a universal serial bus (USB).
The reviewing workstation is a personal computer (PC) with the Capso Vieweimage review software installed. The CapsoView@(CVV) software displays the video and employs a variety of image-enhancement, video navigation, and image analysis features to facilitate the physician's diagnosis. The software allows the physician to efficiently annotate individual frames from the video and compile a procedure report.
1. Acceptance Criteria and Device Performance:
The document outlines acceptance criteria based on two co-primary endpoints and several secondary endpoints.
| Metric | Acceptance Criteria (Lower 95% Confidence Limit) | Reported Device Performance (CapsoCam SV-1) |
|---|---|---|
| Co-Primary Endpoint 1: Completed Exams (vs. PillCam SB2) | Not less than -15% difference in proportion | -0.028% (meaning CapsoCam SV-1 was non-inferior) |
| Co-Primary Endpoint 2: Positive Percent Agreement (PPA) | Not explicitly stated as a numerical threshold, but implied to demonstrate similar performance to predicate. | 69.05% (53.97%, 80.93%) |
| Co-Primary Endpoint 2: Negative Percent Agreement (NPA) | Not explicitly stated as a numerical threshold, but implied to demonstrate similar performance to predicate. | 81.94% (71.52%, 89.13%) |
| Secondary Endpoints (Primary Diagnosis Concordance by Type) | Implied to demonstrate similar performance to predicate, confidence intervals provided for each category. | Normal: 81.43% (70.77, 88.81) |
| Vascular: 66.67% (45.37, 82.81) | ||
| Mass/Polyp: 0.00% (0.00, 48.99) | ||
| Blood: 33.33% (6.15, 79.23) | ||
| Ulcerative: 64.29% (38.76, 83.66) | ||
| Other: 50.00% (9.45, 90.55) | ||
| Battery Life | Not explicitly stated as an acceptance criterion in the provided text, but data gathered. | Mean: 16.5 Hrs, Median: 16 Hrs, Standard Deviation: 2.39 Hrs |
2. Sample Size Used for the Test Set and Data Provenance:
- Sample Size: 121 subjects (age 18 to 85)
- Data Provenance:
- Country of origin: USA (Multi Center Pivotal Study)
- Retrospective or Prospective: Prospective (clinical trial was performed "on the CapsoCam® Capsule Endoscope System per Protocol Number SV-1- Comparison of Capso Vision SV-1 to PillCam® SB2")
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts:
- Number of Experts: Three (3) independent readers per set of images.
- Qualifications of Experts: The document states "independent readers, who were blinded to the subject clinical information or suspected diagnosis." Specific medical qualifications (e.g., "radiologist with 10 years of experience") are not provided in the submitted text. They are likely gastroenterologists or other medical professionals specializing in capsule endoscopy review.
4. Adjudication Method for the Test Set:
- Adjudication Method: If any of the three (3) readers in a group disagreed with the Primary Diagnosis or complete exam, the findings were discussed by members of the group until a 2/3 majority or consensus agreement for Primary Diagnosis was reached.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:
- Yes, a comparative effectiveness study was done. The study compared the CapsoCam® SV-1 to the predicate device, PillCam® SB2.
- Effect Size of Human Readers' Improvement with AI vs. Without AI Assistance:
- The provided text does not contain information on human readers improving with AI assistance vs. without AI assistance. The study design involved humans reading images from two different capsule devices (CapsoCam SV-1 and PillCam SB2) in a blinded fashion. It does not describe an AI assistance component for human readers. This was a direct comparison between two devices through human interpretation.
6. Standalone (Algorithm Only) Performance:
- No, a standalone (algorithm only) performance study was not described. The study evaluates the "CapsoCam® Capsule Endoscope System," which includes the capsule and the CapsoView® software. Performance metrics are based on the interpretations of human readers utilizing the device's output, making it a human-in-the-loop assessment rather than a standalone algorithm evaluation. The software primarily focuses on image display, enhancement, navigation, and report generation for human review.
7. Type of Ground Truth Used:
- Type of Ground Truth: Expert Consensus. The ground truth ("Primary Diagnosis" and "completed exam" status) for each case was established by a 2/3 majority or consensus agreement among a group of three independent readers comparing the findings from both CapsoCam and PillCam. The document refers to "the most clinically significant Primary Diagnosis."
8. Sample Size for the Training Set:
- Not explicitly provided. The document describes a "Multi Center Pivotal Study" (test set) but does not detail the sample size or specific methodology for a separate training set. The descriptions of training sets are typically associated with machine learning models, and this document focuses on clinical performance comparison of a diagnostic device through human review.
9. How the Ground Truth for the Training Set Was Established:
- Not applicable / Not provided. As the document does not describe a distinct training set for a machine learning model, the method for establishing its ground truth is not mentioned. The document primarily focuses on the validation study (test set) for regulatory approval.
§ 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.”