Search Results

Navient is a computerized surgical navigation system intended as an aid for precisely locating anatomical structures. The Navient system is indicated for any medical condition in which the use of stereotactic surgery may be appropriate, and where reference to a rigid anatomical structure, such as the skull, can be identified relative to a CT, MR-based anatomy model.

Indications:

Example procedures include but are not limited to:

ENT Procedures:

• Transsphenoidal procedures
• Maxillary antrostomies
• Ethmoidectomies
• sphenoidotomies
• Sphenoid explorations
• Turbinate resections
• Frontal sinusotomies
• Intranasal procedures
• Intranasal tumor resections
• All ENT related skull base surgery

Navient is an image guided navigational system intended to assist with preoperative planning and real-time positioning of surgical tools during stereotaxic procedures via optical tracking technology. The system is essentially composed of a computerized main unit (computer), a Navient IR CameraBox, Navient cart, Navient navigation software, and corresponding accessory set.

Navient's guidance function is based on the patient images acquired prior to the procedure, combined with optical measurements of the pose of navigated instruments relative to the patient's anatomy. To enable navigation, the reference instrument/accessory is attached to the patient to enable tracking of the patient's anatomy. The patient images are then spatially registered with the patient's anatomy by matching landmark locations marked on both the image and the patient, followed by matching a path traced by the user on the patient's anatomy with a model of patient's anatomical surface automatically generated from the image data.

The provided FDA 510(k) clearance letter and summary for the Navient Image Guided Navigation System (ENT) do not contain information about the study design or acceptance criteria for AI/algorithm-based performance evaluations. Instead, the document focuses on the system's accuracy, software validation, electrical safety, biocompatibility, and reprocessing validation, all typical for traditional medical devices rather than AI/ML-powered ones.

The document states: "Full system accuracy bench testing: Navient has been validated to the positional accuracy of ≤ 2.0 mm (mean=1.52 mm, STD=0.93 mm, 99% confidence interval of 3.68 mm), with the angular error of ≤ 2.0 deg (mean=1.13 deg, STD=0.43 deg, 99% confidence interval of 2.13 deg). This performance was determined using representative phantoms with system components that are deemed the worst-case in the Navient clinical applications."

This validation refers to the physical navigation system's accuracy in positioning, not the performance of an AI algorithm in tasks like image interpretation or diagnosis. Therefore, I cannot generate the requested table and study details related to AI acceptance criteria and performance based on the specific content provided in this 510(k) document.

The "Navient navigation software" mentioned is described as having a workflow for loading images, planning, setting up, registration, and navigation. This suggests a traditional software interface for guiding the user, rather than an AI/ML algorithm performing diagnostic or predictive functions that would require a ground truth, expert consensus, or MRMC studies.

If we were to hypothetically extract the closest equivalent to "acceptance criteria" for this device, it would be its spatial accuracy, which is a key performance metric for image-guided navigation systems.

Here's a hypothetical structure based on the provided spatial accuracy data, while acknowledging it's not AI-specific:

Hypothetical Acceptance Criteria and System Performance (based on provided spatial accuracy)

Recognizing that the provided document details a traditional image-guided navigation system and not an AI-powered diagnostic/interpretive device, the "acceptance criteria" presented here refer to the system's demonstrated physical accuracy.

1. Table of Acceptance Criteria and Reported Device Performance

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

• Test set sample size: Not explicitly stated as a "test set" in the context of an AI model. The performance data is derived from "Full system accuracy bench testing" using "representative phantoms." The number of measurements or phantom tests isn't specified.
• Data provenance: Not directly applicable as it's a bench test on phantoms, not clinical patient data. The testing was conducted internally by ClaroNav.

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

• Not applicable. The ground truth for positional and angular accuracy in bench testing is defined by precision measurement equipment and physical phantoms, not human experts.

4. Adjudication method for the test set:

• Not applicable. Bench testing does not involve human adjudication.

5. If a multi-reader multi-case (MRMC) comparative effectiveness study was done:

• No. This type of study (MRMC) is typically performed for AI devices that aid human interpretation (e.g., radiologists reading images with AI assistance). The Navient system is a guidance system, not an interpretive AI.

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

• The "Full system accuracy bench testing" represents the standalone performance of the navigation system's hardware and software integration in terms of its ability to track instruments accurately relative to images. It's not an AI algorithm performing a task without human input in the sense of a diagnostic or predictive AI.

7. The type of ground truth used:

• The ground truth for the positional and angular accuracy was established through precise measurements on representative phantoms using calibrated equipment, which is standard for validating the accuracy of surgical navigation systems.

8. The sample size for the training set:

• Not applicable. This is not an AI/ML device that undergoes a training phase on a dataset of examples. Its software processes sensor data and medical images according to deterministic algorithms.

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

• Not applicable, as there is no training set in the context of AI/ML for this device's reported validation.

Ask a specific question about this device