(304 days)
The PENTAX Medical EPK-i5010 Video Processor is intended to be used with the PENTAX camera heads, endoscopes, light sources, monitors and other ancillary equipments for bronchoscopic diagnosis, treatment and video observation.
The PENTAX Medical EPK-i5010 includes PENTAX i-Scan™, a digital, post-processing imaging enhancement technology. i-Scan is intended to be used as an optional adjunct following traditional white light endoscopy and is not intended to replace histopathological sampling. i-Scan is compatible with PENTAX k-series and i-series video bronchoscopes.
The PENTAX EPK-i5010 video processor consists of a video system, integrated light source, monitor, and ancillary equipment. This processor when connected to a compatible EB family bronchoscope is intended for endoscopic diagnostic, treatment and video observation.
The PENTAX i-Scan technology is a digital filter-based image enhancement technique with three modes, i-Scan 1, 2 and 3. i-Scan 1 enhances image topography and edges and i-Scan 2 and 3 enhances the color tone of the image by dissecting and recombining the individual red, green and blue (RGB) components of a white light image.
PENTAX i-Scan™ modes 1, 2, and 3, are intended to give the user an enhanced view of the texture of the mucosal surface and blood vessels. i-Scan 1 provides the user with a view that sharpens surface vessels and enhances surface texture of the mucosa. i-Scan 2 provides the user with increased visibility of blood vessels while also providing the same enhancements to the mucosa achieved in i-Scan 3 provides the user with increased visibility of blood vessels including dimly illuminated far-field regions while also providing the same enhancement to the mucosa achieved in i-Scan 1. The user can select either white light image or i-Scan modes by pressing a pre-programmed button on the scope, by using a pre-programmed foot pedal or by pressing a keyboard button. i-Scan is intended to be used as an optional adjunct following traditional white light endoscopy and is not intended to replace histopathological sampling.
White light is captured from a 300 Watt xenon lamp housed in the EPK-i5010 video processor. All visualization is done with the white light mode first. White light (BGR) illuminates the tissue and transfers the captured light through the video scope or a charged coupled device (CCD). Note that the white light visualization mode is always used first by the physician. The modification of the combination of RGB components for each pixel occurs when the i-Scan function is turned on in the EPK-i5010 video processor. The resulting i-Scan image is then displayed on the observation monitor.
Here's a breakdown of the acceptance criteria and the study details based on the provided document. It's important to note that this document is a 510(k) summary, which focuses on demonstrating substantial equivalence to a predicate device rather than presenting a full clinical trial for a novel AI device. Therefore, some information typically found for AI device studies (like a deep dive into AI model training, specific performance metrics for disease detection, or expert consensus on ground truth for a diagnostic task) is not explicitly detailed.
The device in question is the PENTAX Medical EPK-i5010 Video Processor with EB Family, which includes PENTAX i-Scan™, a digital, post-processing imaging enhancement technology. The primary function described is image enhancement for bronchoscopic diagnosis, treatment, and video observation, as an "optional adjunct following traditional white light endoscopy" and "not intended to replace histopathological sampling."
1. Table of Acceptance Criteria and Reported Device Performance
The document does not provide a table of explicit acceptance criteria with numerical targets for clinical performance (e.g., sensitivity, specificity for a diagnostic task). Instead, the studies aim to demonstrate safety, effectiveness, and substantial equivalence to a predicate device. The performance is assessed through various engineering and non-clinical tests.
| Acceptance Criteria Category | Specific Criteria (Implicitly Met) | Reported Device Performance (Summary) |
|---|---|---|
| Electrical Safety | Compliance with IEC 60601-1:2005+A1:2012, IEC 60601-1-2:2007, IEC 60601-1-6:2010, IEC 60601-2-18:2009 | Confirmed by testing in accordance with mentioned standards. |
| Electromagnetic Compatibility (EMC) | Compliance with IEC 60601-1-2:2007 | Confirmed by testing in accordance with mentioned standards. |
| Software Verification & Validation | Compliance with FDA Guidance for "Content of Premarket Submissions for Software Contained in Medical Devices," IEC 62304:2006 (Class B, Moderate Concern), ISO 14971:2007, IEC 62471:2006 | All testing conducted in compliance with listed standards; V&V demonstrate device performs as intended. |
| Optical Performance (Image Quality) | Demonstrate equivalence of distortion, resolution, and color performance compared to predicate. Demonstrate enhanced view of mucosal surface and blood vessels with i-Scan. | Optical bench testing and animal study images confirmed comparable performance to predicate and demonstrated i-Scan enhancements. |
| Safety and Effectiveness | Device is safe and performs as intended in specified use conditions. | Supported by hardware and software verification, validation, and optical data analysis. |
| Substantial Equivalence | Same intended use, indications for use, fundamental operating principle, and scientific technology as predicate/reference devices. | Established through comparison of technological characteristics and performance data. |
2. Sample Size Used for the Test Set and Data Provenance
- Test Set Description: The "test set" primarily comprises images obtained during a non-clinical animal study and optical bench testing. It's not a clinical diagnostic test set for evaluating specific disease detection performance.
- Sample Size:
- Animal Study: A "library of images" was obtained. The exact number of individual images or animal subjects is not specified.
- Bench Testing: Images were generated from the porcine pulmonary location using two PENTAX bronchoscopes (one HD, one standard def) and one Olympus bronchoscope. The number of images is not specified.
- Data Provenance:
- Country of Origin: Not specified, but given Pentax's global presence, it could be from various locations.
- Retrospective/Prospective: The animal study and bench testing were likely prospectively generated for the purpose of this submission. The historical data for the reference predicate (PENTAX EPK-i5010 Video Processor K122470) would be retrospective for comparison, but the new data generated for the current submission is prospective.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications
- Number of Experts: The document does not specify the number of human experts used for formal "ground truth" establishment in the context of disease diagnosis.
- Qualifications of Experts: Not specified. Given the nature of an image enhancement device, the evaluation of image quality (e.g., "enhanced view of the texture of the mucosal surface and blood vessels") would likely involve experienced endoscopists, but their roles and specific qualifications are not detailed. The primary goal was to show technical performance and equivalence, not a direct clinical diagnostic improvement verified by experts.
4. Adjudication Method for the Test Set
The document does not describe any formal adjudication method (e.g., 2+1, 3+1 consensus) for establishing ground truth or evaluating the test set images. The evaluation appears to be technical in nature (e.g., optical performance metrics) and qualitative assessment of image enhancement.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
Data for an MRMC comparative effectiveness study, which would quantify how much human readers improve with AI vs. without AI assistance, is not provided or mentioned. The i-Scan technology is a "digital, post-processing imaging enhancement technology," which means it's an imaging aid. The study presented here focuses on technical performance and equivalence, not on a direct comparison of diagnostic accuracy with and without the i-Scan function by multiple human readers across many cases.
6. Standalone (Algorithm Only Without Human-in-the-Loop Performance) Study
The device is an "imaging enhancement technology" intended to be an "optional adjunct following traditional white light endoscopy" and specifically "not intended to replace histopathological sampling." Therefore, a "standalone" performance evaluation in the context of diagnostic accuracy was not conducted nor would it be appropriate since the device is designed as a human-in-the-loop aid, not a standalone diagnostic algorithm. The reported optical performance and image enhancement capabilities are effectively the "standalone" performance of the image processing aspect.
7. The Type of Ground Truth Used
Given that the device is an image enhancement tool and not a diagnostic AI that provides a specific disease classification, the "ground truth" in this context refers more to:
- Reference Optical Measurements: For bench testing, established physical and optical performance parameters (distortion, resolution, color accuracy) serve as ground truth for technical equivalence.
- Real-world Image Characteristics: The animal study images provide a real-world, yet controlled, "ground truth" for evaluating how the i-Scan technology enhances relevant anatomical features (mucosal surface, blood vessels) compared to white light and other technologies (Olympus NBI). This is a qualitative assessment of the image output itself.
- Compliance with Standards: For electrical, EMC, and software, the ground truth is adherence to the specified international standards.
There is no mention of pathology confirmation or outcomes data being used as ground truth for evaluating the diagnostic performance of the i-Scan feature for specific lesions or conditions.
8. The Sample Size for the Training Set
The document does not specify a training set size. The i-Scan technology is described as a "digital, post-processing imaging enhancement technology" with "three modes, i-Scan 1, 2, and 3." This description suggests it might be based on fixed digital filters or algorithms rather than a machine learning model that requires a "training set" in the conventional sense (i.e., for learning from data to make predictions). If any machine learning elements are present, they are not detailed as such, and no training data is specified.
9. How the Ground Truth for the Training Set Was Established
As no "training set" (in the machine learning sense) is specified, there is no information on how ground truth for a training set was established. The i-Scan modes are described as enhancing "image topography and edges" and "color tone" by dissecting and recombining RGB components, implying a rule-based or filter-based approach rather than a data-trained predictive model.
§ 874.4680 Bronchoscope (flexible or rigid) and accessories.
(a)
Identification. A bronchoscope (flexible or rigid) and accessories is a tubular endoscopic device with any of a group of accessory devices which attach to the bronchoscope and is intended to examine or treat the larynx and tracheobronchial tree. It is typically used with a fiberoptic light source and carrier to provide illumination. The device is made of materials such as stainless steel or flexible plastic. This generic type of device includes the rigid ventilating bronchoscope, rigid nonventilating bronchoscope, nonrigid bronchoscope, laryngeal-bronchial telescope, flexible foreign body claw, bronchoscope tubing, flexible biopsy forceps, rigid biopsy curette, flexible biopsy brush, rigid biopsy forceps, flexible biopsy curette, and rigid bronchoscope aspirating tube, but excludes the fiberoptic light source and carrier.(b)
Classification. Class II.