Search Results

The Elekta Medical Linear Accelerator (EMLA) is intended to be used for external beam radiation therapy (EBRT) treatments as determined by a licensed medical practitioner.

It is intended to assist a licensed medical practitioner in the delivery of EBRT to defined target volumes, while sparing surrounding normal tissue and critical organs from excess radiation.

Elekta Synergy and Elekta Harmony are the default entry-level configurations. It is intended to be used for single or multiple fractions using standard dose fractionation, hyperfractionation, and hypofractionation in all areas of the body where such treatment is indicated.

Elekta Infinity and Elekta Harmony Pro are the default mid-level configuration. It is intended to be used for single or multiple fractions using standard dose fractionation, hyperfractionation, hypofractionation and stereotactic delivery (stereotactic body radiation therapy – SBRT; stereotactic ablative radiotherapy – SABR) in all areas of the body where such treatment is indicated.

Versa HD and Elekta Evo are the default high-level configuration. It is intended to be used for single or multiple fractions using standard fractionation, hyperfractionation, hypofractionation and stereotactic delivery (stereotactic body radiation therapy – SBRT; stereotactic ablative radiotherapy – SABR; stereotactic radio surgery - SRS) in all areas of the body where such treatment is indicated and for the treatment of functional disorders, such as trigeminal neuralgia.

The EMLA is indicated for the delivery of curative and palliative intent EBRT to Adult and Pediatric patients with primary benign and malignant tumor and metastasis (or secondaries) anywhere in the body.

The Elekta Medical Linear Accelerator (EMLA) is an external beam, image guided Radiation Therapy device to assist a licensed practitioner in the delivery of ionizing radiation to a defined target volume. The system is located in a radiation-shielded treatment room and consists of several sub-systems, such as, the electron accelerator, beam shaping, imaging, computerized control systems and a treatment table to support the patient with accessories for patient positioning and set-up to deliver therapeutic treatments.

The EMLA is equipped with a MV portal imaging sub-system, i.e. iViewGT, and an optional kV imaging sub-system, i.e. XVI. The table is capable of linear and rotational movements.

The user interface controlling devices are located partly in the treatment room and partly in the control room.

The EMLA is made available in the following models: Elekta Synergy, Elekta Harmony, Elekta Infinity, Elekta Harmony Pro, Versa HD, Elekta Evo. The major differences are described in section VII.

The provided FDA 510(k) clearance letter and summary for the Elekta Medical Linear Accelerator (EMLA) describes performance testing for differences between the subject devices (new EMLA models) and the predicate devices (older EMLA models, K210500). The primary focus of the performance testing detailed in the summary is related to improvements in CBCT image quality and reconstruction.

Here’s a breakdown of the requested information based on the provided text:

1. Table of acceptance criteria and the reported device performance

The document does not explicitly present a table of quantitative acceptance criteria alongside corresponding reported device performance values. Instead, it describes general improvements and conformance to standards. The acceptance is implicitly based on meeting or exceeding the predicate device's performance, especially for the high-definition (HD) reconstruction.

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

• FDK based reconstruction function (Image Quality Evaluation): "acquired image quality phantom data" - Specific sample size not provided, likely laboratory phantom data.
• FDK based reconstruction function (Image Registration Accuracy): "phantom data and CT reference data" - Specific sample size not provided, likely laboratory phantom data.
• HD Reconstruction vs. FDK (Image quality comparison - clinical data sets): A total of 124 different clinical data sets.
• HD Reconstruction vs. FDK (Image registration accuracy - clinical patient CBCT projection data sets): A total of 13 different clinical patient CBCT projection data sets.
• HD Reconstruction vs. FDK (Clinical image quality evaluation - clinical patient data): Specific number of patients/data sets not explicitly stated, but clinical patient data was used for qualitative comparison.

Data Provenance: The document does not explicitly state the country of origin for the clinical data or explicitly state whether it was retrospective or prospective. Given it's clinical data sets for evaluating reconstruction, it's highly likely to be retrospective data collected from clinical operations.

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

The document states:

• For the FDK based reconstruction: "reconstruct a CBCT volume image which is suitable for visualizing anatomies to enable certain clinical judgment" - this implies expert judgment in assessment, but does not specify the number or qualifications.
• For the HD Reconstruction: "suitable for visualizing the pelvic anatomies to enable certain clinical judgment".
• "A clinical image quality evaluation was performed between HD Reconstruction function and FDK based reconstruction function, using clinical patient data, where user qualitatively compared image quality between the predicate device and the subject device and reported improved image quality." - This indicates qualitative evaluation by "user", but the number and specific qualifications (e.g., radiologist with X years of experience) are not defined. The document also mentions "Formal validation of the clinical workflows has been performed by competent and professionally qualified personnel" but does not specify them in relation to ground truth establishment for the test set.

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

The document does not describe a formal adjudication method (like 2+1 or 3+1 consensus) for establishing ground truth or evaluating the clinical image quality. It generally refers to qualitative comparison by "user" or "competent and professionally qualified personnel".

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

The document mentions "A clinical image quality evaluation was performed between HD Reconstruction function and FDK based reconstruction function, using clinical patient data, where user qualitatively compared image quality between the predicate device and the subject device and reported improved image quality." It also notes "A clinical survey shows a preference towards the HD Reconstruction of the subject device over the predicate."

This suggests a form of reader study or survey was conducted, where users (human readers) compare images from the legacy FDK method (without the new AI-ML component) to the HD Reconstruction (which "includes an AI-ML based component to estimate the scatter"). However, it's not explicitly labeled as a "multi reader multi case (MRMC) comparative effectiveness study" in the formal sense, and no quantitative effect size of improvement for human readers is provided. The improvement is described qualitatively (e.g., "improved image quality," "better uniformity and HU accuracy," "preference").

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

Yes, standalone algorithm performance evaluation was done for the FDK and HD reconstruction functions. This is evident from:

• "Image quality evaluation was performed for the FDK based reconstruction function, using acquired image quality phantom data, to evaluate uniformity, spatial resolution, low contrast visibility, and geometric accuracy in accordance with IEC 60601-2-68;"
• "Image quality comparison was performed between HD Reconstruction function and FDK based reconstruction function, using data acquired on phantom data, to evaluate uniformity, spatial resolution, low contrast visibility, and geometric accuracy in accordance with IEC 60601-2-68."
• "Image quality comparison was performed between HD Reconstruction function and FDK based reconstruction function, using a total of 124 different clinical data sets, to evaluate uniformity, Hounsfield Unit consistency, signal-to-noise ratio, contrast-to-noise ratio, and contrast consistency."

These evaluations measure the intrinsic performance of the reconstruction algorithms without explicit human interaction beyond setting up the evaluation.

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

The types of ground truth used include:

• Phantom Data: For evaluating image quality metrics (uniformity, spatial resolution, low contrast visibility, geometric accuracy) and image registration accuracy (against CT reference data). Phantoms often have known properties or are designed to allow for quantitative measurement of image accuracy.
• CT Reference Data: Used for comparing image registration accuracy. CT imaging is considered a high-fidelity reference.
• Clinical Patient Data: Used for comparing various image quality metrics (Hounsfield Unit consistency, signal-to-noise ratio, contrast-to-noise ratio, contrast consistency) and for qualitative "user" comparison of image quality. The "ground truth" for clinical image quality comparison would effectively be the subjective assessment of the users or experts performing the comparison against each other, and against the clinical utility standards.
• Simulated Monte Carlo data: Used to train the neural network component of the HD reconstruction that estimates scatter. This implies a simulated ground truth for scatter estimation.

8. The sample size for the training set

The document states: "It includes an AI-ML based component to estimate the scatter to enable its automatic removal from the projection images acquired by the imager ahead of the volume reconstruction. Simulated Monte Carlo data is used to train the network."

The specific sample size for the training set (i.e., the amount of simulated Monte Carlo data) for the AI-ML component is not provided.

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

The ground truth for the training set of the AI-ML component was established through Simulated Monte Carlo data. Monte Carlo simulations are a computational method that can model the physical interactions of radiation with matter, providing a "ground truth" for scatter estimation in this context.

Ask a specific question about this device