(45 days)
Artemis along with the Needle Guide Attachment is used for image-guided interventional and diagnostic procedures of the prostate gland. It provides 2D and 3D visualization of Ultrasound (US) images and register these images with those from other imaging modalities such as Ultrasound, Magnetic Resonance, Computed Tomography, etc. It also provides the ability to display a simulated image of a tracked insertion tool such as a biopsy needle, guidewire or probe on a computer monitor screen that shows images of the target organ and the projected future path of the interventional instrument taking into account patient movement. The software also provides a virtual grid on the live ultrasound for performing systematic sampling of the target organ. Other software features include patient data management, multi-planar reconstruction, segmentation, image measurements, 2D/3D image registration, reporting, and pathology management.
Artemis is intended for treatment planning and guidance for clinical, interventional and/or diagnostic procedures. The device is intended to be used in interventional and diaical setting. Example procedures include, but are not limited to image fusion for diagnostic clinical examinations and procedures, soft tissue ablations and placement of fiducial markers. Artemis is also intents in active surveillance to keep track of previous procedures information and outcomes.
Artemis is designed to display the 2-D live video received from commercially available ultrasound machines and use this 2-D video to reconstruct a 3-D ultrasound image. The system has been designed to work with the clinicians' existing ultrasound machine, TRUS probe, commercially available needle guide, and needle gun combination. Additional software features include patient data management, multi-planar reconstruction, segmentation, image measurement, reporting and 3-D image registration.
Artemis is comprised of a mechanical assembly that holds the ultrasound probe and tracks probe position. The mechanical tracker is connected to a PC-based workstation containing a video digitizing card and running the image processing software. Control of the ultrasound probe and ultrasound system is done manually by the physician, just as it would be in the absence of Artemis. However, by tracking the position and orientation of the ultrasound probe while capturing the video image, the workstation is able to reconstruct and display a 3-D image and 3-D rendered surface model of the prostate.
The reconstructed 3-D image can be further processed to perform various measurements including volume estimation, and can be examined for abnormalities by the physician. Patient information, notes, and images may be stored for future retrieval. Locations for biopsies may be selected by the physician, displayed on the 3-D image and 3-D rendered surface model, and stored. Previously stored 3-D models may be recalled and a stored 3-D model may be aligned or registered to the current 3-D model of the prostate. This is especially useful for patients under active surveillance.
The physician may attach a commercially available biopsy needle guide compatible to the TRUS probe and use the probe and biopsy needle to perform tissue biopsy. Whenever the ultrasound machine is turned on by the physician, the live 2-D ultrasound image is displayed on the screen of Artemis during the biopsy. As the TRUS probe with attached needle guide is maneuvered by the physician, the position and orientation of the probe with respect to the organ is tracked. Artemis is able to add, display and edit loaded plans for biopsy as well as provide the probe position and needle trajectory relative to the 3-D image and 3-D rendered surface model of the prostate.
In addition to standard transrectal needle guidance procedures, Artemis also supports transperineal needle guidance by mounting a Needle Guide Attachment (NGA). A commercially available needle guide compatible with the NGA is used. The NGA provides additional data to track the needle direction angle. When using transperineal mode, the procedure planning, segmentation, registration and navigation are performed in the same way as the standard transrectal procedure. The only difference lies in how the needs to be moved to target the different planned locations. For the transrectal procedure, the needle guide is always attached to the probe. Therefore moving the probe moves the needle guide. In transperineal needle guidance procedures the needle is not attached to the probe. Therefore the NGA needs to be moved to move the needle guide. Artemis highlights the closed target to the current needle guide position.
Artemis offers the physician additional 3-D information for assessing prostate abnormalities, planning and implementing biopsy procedures. The additional image processing features are generated with minimal changes to previous TRUS probe based procedures, and the physician always has access to the live 2-D ultrasound image during prostate assessment or biopsy procedure. The device also provides automated reports with information and pictures from the procedure.
The provided document is a 510(k) summary for the medical device "Artemis". This document focuses on demonstrating substantial equivalence to predicate devices and outlines nonclinical testing performed. However, it does not contain detailed acceptance criteria, specific reported device performance metrics (e.g., sensitivity, specificity, accuracy, dice score), or a study that directly proves the device meets specific performance acceptance criteria related to clinical efficacy or diagnostic accuracy.
The document states: "Nonclinical and performance testing results are provided in the 510(k) and demonstrate that the predetermined acceptance criteria are met. The Artemis has been designed to comply with the applicable standards." However, these detailed results and acceptance criteria are not elaborated within this specific K162474 summary. The provided text primarily focuses on regulatory compliance, safety, and a comparison of technological characteristics with predicate devices to establish substantial equivalence.
Based on the available text, here's what can be extracted and what is missing:
1. Table of Acceptance Criteria and Reported Device Performance
| Criterion Type | Acceptance Criteria | Reported Device Performance |
|---|---|---|
| Safety & Effectiveness | Device labeling contains instructions for use and necessary cautions, warnings, and notes. Risk Management procedure identifies and controls potential hazards. | Passed all in-house testing criteria validating design, function, and specifications. Measurement validation using phantoms, clinical CT, and MRI images showed performance as well as or better than predicate devices, and demonstrated safety and effectiveness. |
| Regulatory Compliance | Complies with applicable standards (listed below for Emissions, Immunity, Risk, Usability). | Designed to comply with: |
- IEC/EN 60601-1-2:2007/AC:2010, EN 55011:2009+A1:2010, CISPR 11:2009+A1:2010, IEC 61000-3-2:2005+A1:2009 +A2:2009, EN 61000-3-2:2006+A1:2009 +A2: 2009, IEC 61000-3-3:2008, EN 61000-3-3:2008 (Emissions)
- IEC/EN 60601-1-2:2007/AC:2010, IEC 61000-4-2:2008, EN 61000-4-2:2009, IEC 61000-4-3:2006+A1:2008 +A2:2010, EN 61000-4-3:2006+A1:2008 +A2:2010, IEC 61000-4-4: 2004+A1:2010, EN 61000-4-4:2004+A1:2010, IEC 61000-4-5:2005, EN61000-4-5:2006, IEC61000-4-6:2004/A2:2006, EN61000-46:2009, IEC 61000-4-8:2009,EN61000-4-8:2010,IEC61000-4-11:2004,, EN61000-4-11:2004 (Immunity)
- EN/ISO 14971:2012, IEC 62366:2007, IEC 60601-1-6:2010 (Risk and Usability) |
Note: The document explicitly states "Nonclinical and performance testing results are provided in the 510(k) and demonstrate that the predetermined acceptance criteria are met," but the specific numerical acceptance criteria for performance (e.g., accuracy of measurement, registration error tolerance, segmentation precision) and the corresponding quantitative results are not included in this summary. The "Reported Device Performance" column above summarizes the claim of meeting criteria rather than the data itself.
Regarding the study proving the device meets acceptance criteria:
The document mentions "Measurement validation using, phantoms, clinical CT, and MRI images were used to show that Artemis preforms as well as or better than the other predicate devices and furthermore shows that Artemis was safe and effective." This indicates that nonclinical testing was performed.
However, the specific "study" details are limited in this summary. This document is a summary and refers to more detailed "Nonclinical and performance testing results...provided in the 510(k)." The information below is based only on what is explicitly stated in the provided text.
2. Sample size used for the test set and the data provenance
- Test Set Sample Size: Not specified in the provided summary.
- Data Provenance: The document mentions "phantoms, clinical CT, and MRI images." It does not specify the country of origin of the clinical data. It is implied these were retrospective images used for validation, as it doesn't mention a prospective clinical trial for this 510(k).
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 summary.
4. Adjudication method for the test set
- This information is not provided in the summary.
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
- A MRMC comparative effectiveness study involving human readers with and without AI assistance is not mentioned in this summary. The device's primary function is image-guided interventional and diagnostic procedures (e.g., visualization, registration, navigation, segmentation, measurement for prostate biopsies/treatment), which are often tools for physicians rather than AI for interpretation of images. The comparison focuses on the device's technical performance against predicate devices, not on physician performance improvement.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
- The document describes Artemis as providing "2D and 3D visualization," "ability to fuse and register images," "display a simulated image of a tracked insertion tool," "virtual grid," "patient data management, multi-planar reconstruction, segmentation, image measurements, 2D/3D image registration, reporting, and pathology management." These functionalities are tools to assist a human physician. The text "Control of the ultrasound probe and ultrasound system is done manually by the physician, just as it would be in the absence of Artemis" further indicates a human-in-the-loop system. While several features (segmentation, registration, measurement) are algorithmic, the context suggests these are components of a larger system used by a physician, rather than a standalone diagnostic AI algorithm. Therefore, a standalone performance evaluation in the typical sense of a diagnostic AI is not specifically detailed or claimed in this summary.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)
- The summary states "Measurement validation using, phantoms, clinical CT, and MRI images." For phantoms, the ground truth would be the known physical dimensions or properties of the phantom. For clinical CT and MRI images, the method for establishing ground truth (e.g., pathology, surgical findings, expert measurement, or other imaging modalities) is not specified.
8. The sample size for the training set
- The document describes "nonclinical testing" and "measurement validation." It does not provide details on a "training set" for a machine learning algorithm, which suggests the device might not heavily rely on a machine learning model that requires a distinct training/test split in the way modern AI algorithms do for clinical inference. The discussion around "software source code for basic system functionality" points to more traditional image processing and navigation algorithms. Therefore, a "training set" in the context of deep learning is not applicable or mentioned in this summary.
9. How the ground truth for the training set was established
- As a training set is not explicitly mentioned or clearly applicable based on the summary, how its ground truth was established is not provided.
§ 892.2050 Medical image management and processing system.
(a)
Identification. A medical image management and processing system is a device that provides one or more capabilities relating to the review and digital processing of medical images for the purposes of interpretation by a trained practitioner of disease detection, diagnosis, or patient management. The software components may provide advanced or complex image processing functions for image manipulation, enhancement, or quantification that are intended for use in the interpretation and analysis of medical images. Advanced image manipulation functions may include image segmentation, multimodality image registration, or 3D visualization. Complex quantitative functions may include semi-automated measurements or time-series measurements.(b)
Classification. Class II (special controls; voluntary standards—Digital Imaging and Communications in Medicine (DICOM) Std., Joint Photographic Experts Group (JPEG) Std., Society of Motion Picture and Television Engineers (SMPTE) Test Pattern).