Search Results

This device is a self-help tool for individuals aged 18 or older with diagnosed depression. It is intended to be used in addition to usual care and not as a replacement for it.

[Input Description text here]

The provided FDA 510(k) Clearance Letter for the HJY VisualNext 3D Endoscopic Vision System focuses on the device's substantial equivalence to a predicate device, as opposed to a detailed standalone or comparative effectiveness study of an AI-powered diagnostic device. Therefore, many of the requested details, particularly those related to AI algorithm performance (e.g., sample size for test/training sets, data provenance, ground truth establishment, MRMC studies, and effect size of human reader improvement with AI assistance), are not present in this document.

However, based on the information available, here's a breakdown of the acceptance criteria and the study that proves the device meets them:

Device Type: The HJY VisualNext 3D Endoscopic Vision System is an endoscopic vision system, not an AI-powered diagnostic device. Its primary function is to provide 3D visualization during surgical procedures, differentiating it from an AI-based system that might perform automated image analysis or diagnosis.

Acceptance Criteria and Reported Device Performance:

The document outlines acceptance criteria implicitly through the performance of various non-clinical tests. The criteria are met if the device "Pass[es]" the respective tests and demonstrates performance metrics comparable to predefined standards or the predicate device.

Study Details (for the Non-Clinical Performance Testing):

Since the device is a vision system and not an AI algorithm, the traditional "test set" and "training set" concepts as applied to AI models do not directly apply in the same way. The non-clinical testing evaluates the physical and functional performance of the device itself.

1. Sample size used for the Test Set and Data Provenance:

   • Bench Testing: The sample size is not explicitly stated, but it involved "aged and non-aged subject devices" and direct comparison to the predicate device. The data provenance would be laboratory-generated data from device performance measurements.
   • Animal Study Testing: "A porcine animal model" was used. The specific number of animals or trials within the animal study is not provided. The data provenance is described as being from a porcine animal model. This would be prospective data collection, specifically for this study.

2. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications:

   • This metric is not applicable in the context of this device's testing. The "ground truth" for a vision system's performance is typically established by objective physical measurements (e.g., MTF for resolution, calibrated light meters for illumination) and expert subjective evaluation of visual quality in a controlled setting, rather than a consensus on diagnostic findings. The document does not specify the number or qualifications of any human evaluators involved in the image quality assessment during bench or animal testing, only that the data "met the predefined acceptance criteria."

3. Adjudication Method for the Test Set:

   • Not applicable as the testing involves objective performance measurements and comparison against predefined criteria, not diagnostic interpretations requiring adjudication.

4. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done:

   • No. An MRMC study is typically performed for diagnostic devices where human readers interpret medical images, often with and without AI assistance, to measure diagnostic accuracy and efficiency. This device is a surgical visualization tool, not a diagnostic imaging device.

5. If a Standalone (i.e., algorithm only without human-in-the-loop performance) was done:

   • This is not an AI algorithm, so the concept of "standalone performance" of an algorithm is not applicable. The core function of the device is to provide images for human viewing. The non-clinical tests assess the device's ability to produce high-quality images and function as intended.

6. The Type of Ground Truth Used:

   • For Bench Testing: Objective physical measurements (e.g., resolution targets, light sensors, distortion grids) served as the "ground truth" for parameters like FOV, DOF, resolution, etc., along with comparison to the known performance of the predicate device.
   • For Animal Study Testing: The "ground truth" for image quality (resolution, illumination, color representation, contrast, noise) was likely based on objective evaluation against predefined standards and comparative assessment by skilled observers (e.g., surgeons, imaging specialists) who could judge the clarity and utility of the visualization in an anatomical context, compared to the predicate device's 2D view. Anatomical structures within the porcine model served as the "true" objects being visualized.

7. The Sample Size for the Training Set:

   • Not applicable. This device is a hardware system, not an AI algorithm trained on data. There is no "training set" in the context of machine learning.

8. How the Ground Truth for the Training Set was Established:

   • Not applicable, as there is no training set.

Ask a specific question about this device

HJY VisualNext™ Endoscopic Vision System is intended for viewing internal surgical sites during general surgical procedures and for use in visualization of structures within the brain during neurological surgical procedures as well as for viewing internal surgical sites during anterior and posterior spinal procedures, such as nucleotomy, discectomy, and foraminotomy.

The HJY VisualNext™ Endoscopic Vision System is a system used for viewing internal surgical sites during surgical procedures. The system consists of the following components:

• . Endoscope Control Unit (ECU) (Model number: HDSES01)
• . Endoscope (Model number: HDSE201)
  The endoscope is physically connected via a 5m BNC cable to the Endoscope Control Unit (ECU). The Endoscope consists of 2 LED lamps and a CMOS camera, embedded in the proximal end of a rigid metal arthroscope, which captures the image and transmits to and is processed by the Endoscope Control Unit (ECU), subsequently output to and presented on an external monitor. Images are recordable and markable for further analysis. The Endoscope Control Unit (ECU) is not connectable to intranet or Internet.

The provided text describes the acceptance criteria and the study that proves the device meets those criteria for the HJY VisualNext Endoscopic Vision System, a neurological endoscope.

Here's the breakdown of the requested information:

1. Table of Acceptance Criteria and Reported Device Performance:

The document lists various non-clinical tests performed, each with a "Purpose" (which implicitly contains the acceptance criteria) and "Results" (reported device performance). Since the original document doesn't provide a consolidated table of acceptance criteria with numerical targets before the results, the "Purpose" section of each test effectively serves as the acceptance criteria statement. For clarity, I will present the key performance parameters.

2. Sample size used for the test set and data provenance:

• Sample Size for Physical/Optical Tests: For most physical and optical performance tests (FOV, Direction of View, Optical Magnification, Distortion, Image Intensity Uniformity, Signal-to-Noise Ratio, Sensitivity, Depth of Field, Image Resolution), the tests were conducted using a "Disposable endoscope and Endoscope Control Unit (ECU) in both non-aged and aged conditions." The specific number of individual units tested for each parameter is not explicitly stated beyond "disposable endoscope."
• Sample Size for Image Quality Test (Biological Tissue Model):
  • ECU: Non-aged and aged ECU units were used. (Quantity not specified, but implied as at least one of each).
  • Endoscopes: "Six pieces of endoscopes each from non-aged and aged conditions were used for brain and spinal surgery, respectively." This means 6 non-aged endoscopes for brain surgery, 6 non-aged for spinal surgery, 6 aged for brain surgery, and 6 aged for spinal surgery.
  • Comparators: "Two cleared devices each for spine and brain were applied for comparing with non-aged and aged subject devices."
• Data Provenance: The tests were non-clinical laboratory studies. The "Image Quality Test utilizing a clinically relevant biological tissue model" was performed using a "live pig model" in "an animal operating room of a facility accredited under AAALACi standards," following "GLP standard." The country of origin for the data is not specified directly, but the company is based in Taiwan. The document states "No clinical test data was used to support the decision of substantial equivalence," indicating these were not human subject trials.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

This information is not provided in the document. For the physical/optical tests, the "ground truth" is established by the measurement apparatus and methods themselves. For the Image Quality Test on a biological tissue model, the results state "The image quality of the non-aged and aged subject device passed the pre-defined acceptance criteria," and was "comparable to that of the FDA-cleared comparator devices." It's highly probable that expert assessment was involved in determining "image quality" or "comparability," but the number and qualifications of these experts are not explicitly mentioned in the provided text.

4. Adjudication method (e.g., 2+1, 3+1, none) for the test set:

This information is not provided in the document. For most non-clinical performance tests, adjudication (as in clinical image review) is not typically applicable. For the image quality assessment in the biological tissue model, while agreement was tested, details of an adjudication method among multiple evaluators (if any) are not specified.

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:

This information is not applicable/not provided. The device is an endoscopic vision system, not an AI-assisted diagnostic tool for image interpretation by human readers. The studies described are non-clinical performance evaluations of the device's optical and imaging capabilities, and comparison of its image quality to a predicate device. No MRMC study regarding human reader performance with or without AI assistance was conducted or is relevant to this submission.

6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done:

This information is not applicable/not provided as the device is not an AI algorithm. It is an endoscopic vision system that provides direct visual output for human use. The performance tests are of the hardware's image capture and display capabilities in a standalone manner (i.e., the system itself creating the image).

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):

For the various physical and optical parameters, the "ground truth" is based on instrumented measurements against defined engineering specifications and comparison to the predicate device's characteristics. For the "Image Quality Test utilizing a clinically relevant biological tissue model," the ground truth for acceptability was based on pre-defined acceptance criteria and comparability to FDA-cleared comparator devices. While expert evaluation likely played a role in assessing "image quality" and "comparability," the final "ground truth" for the test was determined by whether these qualitative and quantitative assessments met the pre-defined criteria. The results state: "The image quality... passed the pre-defined acceptance criteria" and was "comparable to that of the FDA-cleared comparator devices." There is no mention of pathology or outcomes data being used as ground truth.

8. The sample size for the training set:

This information is not applicable/not provided. The HJY VisualNext Endoscopic Vision System is a hardware device (endoscope and control unit) that captures and displays images; it is not an AI/ML device that requires a training set.

9. How the ground truth for the training set was established:

This information is not applicable/not provided as there is no training set for this type of device.

Ask a specific question about this device