Search Results

The CORVUS system is a radiation treatment planning package designed to allow medical physicists, dosimetrists, and radiation oncologists to create conformal treatment plans using photon (x-ray, gamma ray) external beam radiation therapy. The treatment plans generated by CORVUS are based upon treatment machine-specific data and are intended to provide a guide to delivering external beam radiation therapy which conforms to the target volume defined by the radiation oncologist.

The CORVUS system is valid for use only with external beam photon therapy; calculations for electrons and intracavity sources (Brachytherapy) are NOT supported.

CORVUS is a semi-automatic planning system: For forward planning, the system allows the user to design a treatment plan. For IMRT, rather than simply verifying a user-designed plan, the system itself suggests a plan. A clinician then reviews and approves the plan.

CORVUS is designed to generate plans for treatment delivery systems that can create multiple radiation patterns composed of pencil beams on which the intensity can be individually controlled. The treatment beams are weighted so that when they are projected into the treatment space they superimpose to give the desired dose distribution.

Each radiation field is generated using one of several optimization methods provided with the system, including simulated annealing and gradient descent.

The treatment beams are set not only to deliver the prescribed dose to the identified target volume, but also to keep the dose to other sensitive volumes below user-defined limits. Planning is done volumetrically: the beam weights for treating the entire target volume are generated simultaneously. The dose matrix is volumetric. The dose to each point is calculated to be that received from all beams and from all gantry angles. Dosage is calculated using a finite size pencil beam (FSPB) algorithm based on the beam characterization of clinically measured data. The degree to which a treatment plan is optimized is determined in part by constraints placed on the planning algorithm. The user has direct control over these constraints, which include dose goals to the target structures, dose limits to the sensitive structures, and the specification of arcs or fixed gantry positions in the treatment plan.

CORVUS treatment plans need not have the isocenter located within the target volume. An unlimited number of targets falling within the treatment volume can be planned for at the same time. Dose may be prescribed for up to 32 structures, 29 of them user-selectable, any number of which may be separate targets or radiation-sensitive structures. Each structure can have a separate dose prescription.

Here's a breakdown of the acceptance criteria and the study details for the CORVUS device, based on the provided FDA 510(k) summary:

The CORVUS device is a radiation treatment planning system. The supplied document focuses on demonstrating substantial equivalence to a predicate device (CORVUS 14) rather than a direct clinical performance study against specific acceptance criteria for diagnostic accuracy in the typical sense of AI/ML devices.

Therefore, the acceptance criteria and study detailed below are primarily centered around dosimetric accuracy, software functionality, and compliance with relevant standards, which are the key performance indicators for a treatment planning system seeking 510(k) clearance.

1. Table of Acceptance Criteria and Reported Device Performance

Given that this is a 510(k) submission for a treatment planning system and not a diagnostic AI device, the "acceptance criteria" are generally related to dosimetric accuracy and software performance compared to established benchmarks or the predicate device.

• Calculations for Cobalt-60 based Forward Planning align with expected accuracy (implied by adding this new feature).
• Pencil-beam algorithm and dose calculation options (homogeneous, EPL) maintain accuracy. | - "The accuracy of treatment plans was evaluated through comparison with medical physics measurements 2D array and ion chamber measurements."
• The submission states that the "verification and validation results demonstrate that the CORVUS 15 system met its design requirements and specifications."
• The fundamental scientific technology and intended use/indications are unchanged from the predicate, implying similar dosimetric performance for existing features. |
  | Software Functionality | - All device functionalities work as per its intended use.
• Risk mitigation is ensured for software.
• Conforms to required software standards (IEC 62304, IEC 62366-1, IEC 62083).
• New features (Cobalt-60 Forward Planning) function correctly and safely. | - "Design verification and validation testing was performed to ensure that the device functionality works as per its intended use, all risks are mitigated, is substantially equivalent, and the product conforms to the required standards."
• "The verification activities included system tests, module tests, anomaly verification, code reviews, and run-through integration tests."
• "The validation activities included clinical workflow, treatment planning software usability, dosimetric accuracy, and import-export." |
  | Substantial Equivalence | - No new issues of safety or effectiveness are raised compared to the predicate device (CORVUS 14).
• Intended use and indications for use remain substantially similar or unchanged.
• Fundamental technical characteristics are the same. | - "The fundamental scientific technology of the CORVUS 15 with respect to its predicate device (CORVUS 14) system has not changed. The intended use and indications for use of the device have not changed. Based upon the performance testing results for CORVUS 15 (as detailed in the submission), the system raises no new issues of safety or effectiveness." |

2. Sample Size Used for the Test Set and Data Provenance

The document does not specify a "test set" in the sense of a dataset of patient cases used for clinical efficacy as would be seen for a diagnostic AI. Instead, the testing focuses on system verification and dosimetric validation.

• Sample Size for Test Set: Not explicitly stated in terms of patient cases. The testing involved "medical physics measurements 2D array and ion chamber measurements." This typically involves a set of phantoms or carefully controlled geometric setups, but the number of such measurements or scenarios is not provided.
• Data Provenance: Not applicable in the context of patient data. The "measurements" would be generated in a lab setting, not from patients.

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

• Number of Experts: Not explicitly stated. The "accuracy of treatment plans was evaluated through comparison with medical physics measurements." This implies evaluation by medical physicists.
• Qualifications of Experts: The involvement of "medical physicists" is mentioned. Their specific years of experience or board certifications are not provided in this summary.

4. Adjudication Method for the Test Set

Not applicable. This is not a study requiring adjudication of expert interpretations of images or patient outcomes. The "ground truth" for dosimetric accuracy would be independent physical measurements.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done

No. This type of study is typically performed for diagnostic or AI-assisted interpretation devices to compare human performance with and without AI assistance. CORVUS is a treatment planning system, and its evaluation focuses on the accuracy of its calculations and software functionality, not on improving human reader performance in interpreting images.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done

Yes, in the context of a treatment planning system. The "dosimetric accuracy" testing evaluates the algorithm's output (calculated dose distributions) against physical measurements, which is an algorithm-only evaluation. While the system is used by humans (medical physicists, dosimetrists, radiation oncologists), the core dose calculation component performs its function algorithmically, and its output is verified independently.

7. The Type of Ground Truth Used

The ground truth used for evaluating dosimetric accuracy was medical physics measurements (2D array and ion chamber measurements). This represents a physical, objective standard for verifying dose distribution accuracy.

8. The Sample Size for the Training Set

Not applicable. CORVUS is a rule-based or physics-based treatment planning system, not a machine learning (ML) or artificial intelligence (AI) device that typically requires a large 'training set' of data in the common sense for algorithm development. Its algorithms are based on established physics models and mathematical optimization techniques.

9. How the Ground Truth for the Training Set Was Established

Not applicable, as there isn't a "training set" in the context of machine learning. The algorithms are built upon fundamental physics and mathematical principles, and characterized/calibrated using machine-specific data (e.g., beam data from a linear accelerator). This characterization data is not a "ground truth for training" in the AI/ML sense but rather input parameters that define the physical behavior of the treatment machine.

Ask a specific question about this device

The CORVUS system is a radiation treatment planning package designed to allow medical physicists, dosimetrists, and radiation oncologists to create conformal treatment plans using photon (x-ray, gamma ray) external beam radiation therapy. The treatment plans generated by CORVUS are based upon treatment machine-specific data and are intended to provide a guide to delivering external beam radiation therapy which conforms to the target volume defined by the radiation oncologist.

The CORVUS system is valid for use only with external beam photon therapy; calculations for electrons and intracavity sources (Brachytherapy) are NOT supported.

CORVUS is a semi-automatic planning system: For forward planning, the system allows the user to design a treatment plan. For IMRT, rather than simply verifying a user-designed plan, the system itself suggests a plan. A clinician then reviews and approves the plan.

CORVUS is designed to generate plans for treatment delivery systems that can create multiple radiation patterns composed of pencil beams on which the intensity can be individually controlled. The treatment beams are weighted so that when they are projected into the treatment space they superimpose to give the desired dose distribution.

Each radiation field is generated using one of several optimization methods provided with the system, including simulated annealing and gradient descent.

The treatment beams are set not only to deliver the prescribed dose to the identified target volume, but also to keep the dose to other sensitive volumes below user-defined limits. Planning is done volumetrically: the beam weights for treating the entire target volume are generated simultaneously. The dose matrix is volumetric. The dose to each point is calculated to be that received from all beams and from all gantry angles. Dosage is calculated using a finite size pencil beam (FSPB) algorithm based on the beam characterization of clinically measured data. The degree to which a treatment plan is optimized is determined in part by constraints placed on the planning algorithm. The user has direct control over these constraints, which include dose goals to the target structures, dose limits to the sensitive structures, and the specification of arcs or fixed gantry positions in the treatment plan.

CORVUS treatment plans need not have the isocenter located within the target volume. An unlimited number of targets falling within the treatment volume can be planned for at the same time. Dose may be prescribed for up to 32 structures, 29 of them user-selectable, any number of which may be separate targets or radiation-sensitive structures. Each structure can have a separate dose prescription.

The CORVUS system is a radiation treatment planning package. The provided text, which appears to be an FDA 510(k) summary, focuses on demonstrating substantial equivalence to a predicate device (CORVUS 2011, K151469) rather than detailing specific acceptance criteria and a study proving device performance against those criteria in a typical clinical performance study format.

However, based on the Summary of Testing section (L), we can infer how performance was assessed for the purpose of demonstrating substantial equivalence.

Here's an attempt to structure the information according to your request, extracting what's available and noting what is not explicitly stated in the provided text.

Acceptance Criteria and Device Performance Study for CORVUS (K201350)

The provided document, an FDA 510(k) summary, describes the CORVUS system as a radiation treatment planning package. The core purpose of this submission is to demonstrate substantial equivalence to a predicate device (CORVUS 2011, K151469). Therefore, the "acceptance criteria" and "study" described herein are primarily focused on design verification and validation tests to ensure the new device functions as intended, mitigates risks, and is equivalent to its predicate.

1. Table of Acceptance Criteria (Inferred) and Reported Device Performance

Since explicit, quantifiable acceptance criteria with corresponding performance metrics are not presented in a direct table in this document, they are inferred from the verification and validation (V&V) activities described. The reported performance is generally stated as meeting requirements and demonstrating substantial equivalence.

2. Sample Sizes and Data Provenance for Test Set

The document does not explicitly state the sample size (e.g., number of patient cases, treatment plans) used for the "test set" in the context of clinical plan quality, optimization, or dosimetric accuracy.

• Sample Size: Not explicitly stated. The document refers to "comparison with medical physics measurements including film and ion chamber measurements" and "clinical plan quality / optimization comparison," implying multiple cases or scenarios were tested.
• Data Provenance: Not explicitly stated. It is typical for such V&V studies to use a combination of simulated data, phantom data, and potentially de-identified clinical data. Given the "medical physics measurements" and "clinical plan quality" evaluations, it's likely that phantom data and potentially retrospective clinical cases were used, but the document does not specify country of origin or whether data was retrospective or prospective.

3. Number of Experts and Qualifications for Ground Truth

The document mentions "medical physicists, dosimetrists, and radiation oncologists" as users of the system and participants in activities like "clinical workflow" and "clinical plan quality / optimization comparison." However, it does not specify:

• Number of Experts: Not explicitly stated.
• Qualifications of Experts: It implies that these are qualified professionals (e.g., medical physicists, radiation oncologists) who would typically have extensive experience in radiation therapy planning and delivery. Specific years of experience are not mentioned. Their role would be to evaluate the usability and clinical appropriateness of the plans generated by CORVUS, likely against established clinical best practices or phantom measurements used as ground truth.

4. Adjudication Method for the Test Set

The document does not describe a formal adjudication method (e.g., 2+1, 3+1) for evaluating a test set. The validation activities included "clinical workflow, treatment planning software usability, clinical plan quality / optimization comparison, and dosimetric accuracy," which suggest evaluations were performed by qualified personnel. However, the process for resolving disagreements or establishing a consensus "ground truth" among multiple reviewers is not detailed.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

• Was it done? No, the document does not describe a Multi-Reader Multi-Case (MRMC) comparative effectiveness study evaluating the improvement of human readers with AI vs. without AI assistance. The CORVUS system is a treatment planning tool for physicists, dosimetrists, and oncologists, not an AI-assisted diagnostic or interpretation system for human readers in the traditional sense of an MRMC study. Its function is to generate plans, which are then reviewed and approved by clinicians.
• Effect Size: Not applicable, as no MRMC study was performed.

6. Standalone Performance Study

• Was it done? Yes, a standalone performance assessment akin to an "algorithm only" evaluation was performed. The "dosimetric accuracy" assessment ("comparison with medical physics measurements including film and ion chamber measurements") and "clinical plan quality / optimization comparison" directly evaluate the output of the algorithm in generating treatment plans and calculating doses. This inherently assesses the software's performance independent of real-time human interaction during plan generation, although human experts certainly reviewed and evaluated these outputs.

7. Type of Ground Truth Used

The ground truth for evaluating the CORVUS system's performance appears to be a combination of:

• Medical Physics Measurements: "Film and ion chamber measurements" served as the ground truth for dosimetric accuracy. These are standard, highly accurate physical measurements used to verify calculated dose distributions in phantoms.
• Clinical Best Practices/Expert Consensus: "Clinical plan quality / optimization comparison" likely relied on assessment against established clinical guidelines, benchmarks, or expert consensus regarding acceptable dose distributions to targets and sensitive structures.

8. Sample Size for the Training Set

The document does not explicitly state a "training set" sample size for the CORVUS system in the context of AI/machine learning. CORVUS is described as a "semi-automatic planning system" using optimization methods like "simulated annealing and gradient descent," and a "finite size pencil beam (FSPB) algorithm based on the beam characterization of clinically measured data."

This description suggests a rules-based system or an optimization engine that may be calibrated with machine-specific data rather than trained on a large dataset of patient cases in the way that contemporary deep learning models are. If "training" refers to the data used to characterize the treatment machines for the FSPB algorithm:

• Sample Size: Not specified. This would typically involve a proprietary dataset collected by the manufacturer to model the performance of specific linear accelerators or Cobalt-60 units.

9. How the Ground Truth for the Training Set Was Established

Given the nature of the device (a treatment planning system using physics-based algorithms and optimization), the "ground truth" for the calibration/development (if we call it training) of the system's core algorithms (like the FSPB algorithm) would generally be established by:

• Physical Measurements: Extensive physical measurements of radiation beam characteristics (e.g., dose profiles, depth doses, output factors) using dosimeters (ion chambers, film, detectors) on phantoms under various geometries and configurations. This "clinically measured data" is used to characterize the treatment machine's behavior, which in turn informs the FSPB algorithm.
• Mathematical/Physics Models: The algorithms themselves are based on established physics principles, which serve as foundational "ground truth" for the calculations.

Ask a specific question about this device

The SOFTDISO system is a standalone software solution designed to be used by medical physicists, radiation oncologists, and dosimetrists to have an overview of the treatment plans delivered from Treatment Systems (LINAC Systems using high energy x-rays) to the patient. This solution is to be used as a quality control check purposes only.

SOFTDISO does not provide any conclusions or policy for interpretation of the results based on the comparison results data. The output from SOFTDISO cannot be and must not be used for changing the planning or treatment strategy or as a means of proving the effectiveness of the quality control process/chain during treatment. It is the physician's responsibility to verify the working of the quality control process/chain and the correctness of the dose delivered.

SOFTDISO software permit an IN-VIVO dosimetric analysis (IVD) in patient for 3DCRT, IMRT and VMAT beams using a Si-EPID portal systems.

The dosimetric analysis consists of two tests elaboration, dose reconstruction in the isocenter point and EPID images y-analysis. In case of out of tolerance levels, the system is capable of underline the right type of quality control useful to remove the causes of discrepancy.

SOFTDISO software is based on the usage of generalized correlation factors between transit signals measured by EPID and doses measured in water equivalent solid phantom, along the beam central axis.

SOFTDISO software uses generalized functions taken from measurements of 70 beams of Varian, Elekta and Siemens linacs. So only a small set of dosimetric measurements must be performed by the user for the software commissioning, and some of those measurements are already performed during the linac's beams calibration. In particular some of the measurements that the user performs include calculating the beam dose in cGy/UM in reference conditions (field 10.10 cm2 at drif =10 cm depth), the beam quality indicator (TPR20, 10) and the attenuation factor WF for wedged beams. The user also performs a measure of the EPID signal in reference condition (field 10.10 cm2 at SED distance) for EPID calibration.

SOFTDISO software can use information coming from record and verify systems through DICOM and DICOM-RT protocols. DICOM protocol is used for transferring images from CT and from EPID and the DICOM-RT protocol is used for transferring information related to the TPS system.

The provided document describes the K170345 510(k) submission for the SOFTDISO device. Here's a breakdown of the acceptance criteria and the study that proves the device meets those criteria, based on the provided text:

Acceptance Criteria and Reported Device Performance

Study Details

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

• External Validation (Dosimetry Accuracy):
  • 3DCRT: 192 tests
  • IMRT: 48 tests
  • VMAT: 60 tests
  • Data Provenance: Conducted at "three different clinical sites" for 3DCRT and IMRT, and "two different clinical sites" for VMAT. The document does not specify the country of origin, but "clinical sites" suggests real-world patient or phantom data. This appears to be prospective data collection as part of validation, though the exact nature (e.g., real patient treatments vs. carefully controlled phantom experiments) is not fully detailed.
• Clinical Validation (Usability and Error Detection):
  • Usability: 1287 treatments
  • Error Detection: 823 patient data (340 3DCRT, 483 VMAT treatments)
  • Data Provenance: The document states "Clinical validation was performed by applying in-vivo dosimetry checks on 1287 treatments" and "performed on 823 patient data." This strongly suggests retrospective or prospective use of real patient data from clinical settings. The country of origin is not specified.

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

The document does not explicitly state the number of experts used to establish the ground truth or their specific qualifications for the test set. The Indications for Use state that "It is the physician's responsibility to verify the working of the quality control process/chain and the correctness of the dose delivered," implying that medical physicists, radiation oncologists, and dosimetrists are involved in the overall quality control process. However, for the validation studies themselves, the method for establishing ground truth and the experts involved are not detailed.

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

The document does not describe any adjudication method used for the test set.

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 multi-reader multi-case (MRMC) comparative effectiveness study was not described. The validation focused on the performance of the device itself against established dosimetric tolerances and its ability to identify errors, rather than directly comparing human reader performance with and without the device.

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

Yes, a standalone performance evaluation was performed as part of the "External validation to validate the accuracy of dosimetry procedure." This involved evaluating the R ratios and gamma analysis indexes of the SOFTDISO software against preset tolerance levels (e.g., ± 5% for R ratios, ≥ 95% for γ%, ≤ 0.3 for γmean). This directly assesses the algorithm's accuracy in comparing planned and measured doses.

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

The ground truth appears to be based on dosimetric comparison against established tolerance levels.

• For the R ratios, the implicit ground truth is the "planned dose" from the Treatment Planning System (TPS) and the "measured dose" from the EPID, with the expectation that these should match within a certain tolerance.
• For the gamma analysis, the ground truth is again the comparison between a "reference EPID image" (likely related to the planned dose distribution) and a "current one," with deviation measured against industry-standard gamma criteria.
• For the clinical validation identifying errors, the ground truth for "errors" (patient setup, laser misalignment, TPS beam-implementation, patient morphological changes) would likely be established through independent clinical assessment or follow-up by the medical physicists and oncologists at the sites. The document doesn't detail how these "errors" were definitively confirmed as ground truth for the study.

8. The sample size for the training set

The document states, "SOFTDISO software uses generalized functions taken from measurements of 70 beams of Varian, Elekta and Siemens linacs." This indicates that these 70 beams likely formed part of the training or commissioning data used to develop these generalized functions within the software, which then allows for site-specific commissioning with a smaller dataset. The exact nature of this "training set" for a machine learning context is not detailed, but the 70 beams are the most relevant number provided for the model's underlying data.

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

For the "generalized functions taken from measurements of 70 beams," the ground truth was established through direct physical measurements. The text mentions:

• "measurements of 70 beams of Varian, Elekta and Siemens linacs."
• "Small set of dosimetric measurements must be performed by the user for the software commissioning, and some of those measurements are already performed during the linac's beams calibration."
• Specifically highlights measurements for "calculating the beam dose in cGy/UM in reference conditions (field 10.10 cm2 at drif =10 cm depth), the beam quality indicator (TPR20, 10) and the attenuation factor WF for wedged beams," and "a measure of the EPID signal in reference condition (field 10.10 cm2 at SED distance) for EPID calibration."

This indicates that the ground truth for the "training" data (the 70 beams and
user-performed commissioning measurements) was established through physical dosimetric measurements performed by qualified personnel (likely medical physicists) in a controlled environment.

Ask a specific question about this device

The CORVUS system is a radiation treatment planning package designed to allow medical physicists, dosimetrists, and radiation oncologists to create conformal treatment plans using photon (x-ray, gamma ray) external beam radiation therapy. The treatment plans generated by CORVUS are based upon treatment machine-specific data and are intended to provide a guide to delivering external beam radiation therapy which conforms to the target volume defined by the radiation oncologist.

The CORVUS system is valid for use only with external beam photon therapy; calculations for electrons and intracavity sources (Brachytherapy) are NOT supported.

CORVUS is a semi-automatic planning system: rather than simply verifying a user-designed plan, the system itself suggests a plan. A clinician then reviews and approves the plan.

CORVUS is designed to generate plans for treatment delivery systems that can create multiple radiation patterns composed of pencil beams on which the intensity can be individually controlled. The treatment beams are weighted so that when they are projected into the treatment space they superimpose to give the desired dose distribution.

Each radiation field is generated using one of several optimization methods provided with the system, including simulated annealing and gradient descent.

The treatment beams are set not only to deliver the prescribed dose to the identified target volume, but also to keep the dose to other sensitive volumes below user-defined limits. Planning is done volumetrically: the beam weights for treating the entire target volume are generated simultaneously. The dose matrix is volumetric. The dose to each point is calculated to be that received from all beams and from all gantry angles. Dosage is calculated using a finite size pencil beam (FSPB) algorithm based on the beam characterization of clinically measured data. The degree to which a treatment plan is optimized is determined in part by constraints placed on the planning algorithm. The user has direct control over these constraints, which include dose goals to the target structures, dose limits to the sensitive structures, and the specification of arcs or fixed gantry positions in the treatment plan.

CORVUS treatment plans need not have the isocenter located within the target volume. An unlimited number of targets falling within the treatment volume can be planned for at the same time. Dose may be prescribed for up to 32 structures, 29 of them user-selectable, any number of which may be separate targets or radiation-sensitive structures. Each structure can have a separate dose prescription.

Here's an analysis of the acceptance criteria and study information for the CORVUS Radiation Therapy Planning System, based on the provided document:

This document describes a premarket notification (510(k)) for the CORVUS Radiation Therapy Planning System (K151469), which is an accessory to medical devices of Major Level of Concern. The submission focuses on the upgrade from CORVUS 09 to CORVUS 2011, primarily adding support for Cobalt-60 based external beam radiation treatment planning (Gamma Tomotherapy) and updated operating system/hardware.

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria provided are mostly qualitative (e.g., "clinically acceptable," "similar") or defined by pass/fail thresholds for dosimetric accuracy.

2. Sample Size Used for the Test Set and Data Provenance

• Sample Size: The document does not explicitly state the numerical sample size for patient cases or plans used in the test sets for most clinical validation tests. It mentions "various pencil beam sizes" (Test 4), "multiple test results" (Test 4), and "all Cobalt validation plans" (Test 8), which are not specific numbers. For Test 1 (Linac Dose Equivalence) and Test 2 (Clinical Comparison), it refers to "the same patients" and "treatment plans created for Cobalt treatment," but no specific count of patients or plans is given.
• Data Provenance: The document does not specify the country of origin of the data. It also does not explicitly state whether the data was retrospective or prospective. Given the context of clinical comparison and re-calculation, it is likely retrospective, using existing patient data or phantom measurements, but this is not definitively stated.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of those Experts

• The document mentions that the CORVUS system is "designed to allow medical physicists, dosimetrists, and radiation oncologists to create conformal treatment plans." And that "It is the physician's responsibility to verify that the dose distributions... are appropriate for a particular patient."
• However, for the specific validation studies on the test set, the document does not explicitly state the number of experts or their qualifications used to establish the "ground truth" or to review the clinical acceptability of plans. The "clinical acceptability" (Test 2) implies expert review, but details are not provided. The dosimetric accuracy tests (Tests 3, 4, 8) rely on direct physical measurements (film and ionization chambers) rather than expert consensus as the ground truth.

4. Adjudication Method for the Test Set

• The document does not describe any explicit adjudication method for expert review (e.g., 2+1, 3+1). For the dosimetric tests, the adjudication is based on direct measurement comparison against defined physical thresholds (e.g., 4% / 4mm).

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

• No, a MRMC comparative effectiveness study was not done as described in the document. The studies were primarily focused on comparing the CORVUS 2011 algorithm's output against:
  • Previous version (CORVUS 09) outputs (Tests 1, 2, 7)
  • Physical measurements (Tests 3, 4, 8)
  • Expected computational outcomes (Test 5)
• There is no mention of human readers, their performance without AI (the planning system itself), or an effect size of improvement with AI assistance. The system is a planning tool, not an AI for interpretation or diagnosis that would augment human review in that manner.

6. Standalone (Algorithm Only Without Human-in-the-Loop Performance) Study

• Yes, the studies are largely standalone in terms of the algorithm's performance. The tests primarily evaluate the calculations and outputs of the CORVUS 2011 system itself (dose equivalence, dosimetric accuracy, consistency of calculations, transfer of data).
• While the system is a "semi-automatic planning system" that "suggests a plan" which a "clinician then reviews and approves," the validation tests described focus on the accuracy and equivalence of the generated plans and calculations themselves, rather than the overall human-in-the-loop workflow's effectiveness. Test 2 (Clinical comparison) does allude to "clinical efficacy" and "clinical acceptability" but without specifying a human-in-the-loop scenario. The dosimetric tests are purely algorithmic output vs. physical measurement.

7. Type of Ground Truth Used

• Mixed:
  • For dosimetric accuracy (Tests 3, 4, 8): Physical measurements using EDR2 film and ionization chambers (0.125 cc chamber, CC01 ionization chamber). This is a objective, empirical ground truth.
  • For clinical comparisons and equivalence (Tests 1, 2, 7): The ground truth is established by comparison to the predicate device's output (CORVUS 09) and expert consensus on "clinical acceptability" or "clinical similarity," though the specifics of this expert consensus are not detailed.
  • For dose comparison consistency (Test 5) and system transfer (Test 6): Expected computational outcomes and predefined data formats (DICOM RT) serve as the ground truth.

8. Sample Size for the Training Set

• The document does not provide any information about the sample size for a training set. CORVUS is described as a "semi-automatic planning system" using "optimization methods including simulated annealing and gradient descent," and dosage is calculated "using a finite size pencil beam (FSPB) algorithm based on the beam characterization of clinically measured data." This suggests a model-based approach with measured beam data, rather than a machine learning model that would typically have a "training set" in the modern sense. The "clinically measured data" for beam characterization itself would be a foundational dataset, but its size is not specified as a "training set."

9. How the Ground Truth for the Training Set Was Established

• As a "training set" in the modern AI/ML sense is not explicitly mentioned, the concept of establishing ground truth for it is not addressed. The "beam characterization of clinically measured data" for the FSPB algorithm would have its ground truth established through direct physical measurements of radiation beam properties.

Ask a specific question about this device

Intended Use: CORVUS is intended for use as a planning tool for conformal radiation therapy. Using operator-supplied input and patient scans, it creates a plan for treatment delivery systems and generates a set of beam weights that, when applied to a compatible system, facilitates delivery of an intensity-modulated 3D conformal radiation therapy treatment. CORVUS is intended only to suggest a delivery plan. It is the physician's responsibility to verify that the dose distributions which would result from plan implementation are appropriate for a particular patient. The CORVUS system is intended to be used as an integrated system with a modulating device for planning and delivery of conformal radiation therapy. The modulating device can be the NOMOS MIMIC, nomosSTAT MLC, or a supported MLC. CORVUS produces radiation fields which are modulated to conform to the projected tumor volume plus margins. The system tries to achieve target goals while sparing sensitive structures.

Indications for Use: The CORVUS system is a radiation treatment planning package designed to allow medical physicists, dosimetrists, and radiation oncologists to create conformal treatment plans using photon (x-ray) external beam radiation therapy. The treatment plans generated by CORVUS are based upon treatment machine-specific data and are intended to provide a guide to delivering external beam radiation therapy which conforms to the target volume defined by the radiation oncologist. The CORVUS system is valid for use only with external beam photon therapy; calculations for electrons and intracavity sources (Brachytherapy) are NOT supported.

CORVUS is a semi-automatic planning system: rather than simply verifying a userdesigned plan, the system itself suggests a plan. A clinician then reviews and approves the plan. CORVUS is designed to generate plans for treatment delivery systems that can create multiple radiation patterns composed of pencil beams on which the intensity can be individually controlled. The treatment beams are weighted so that when they are projected into the treatment space they superimpose to give the desired dose distribution. Each radiation field is generated using one of several optimization methods provided with the system, including simulated annealing and gradient descent. The treatment beams are set not only to deliver the prescribed dose to the identified target volume, but also to keep the dose to other sensitive volumes below user-defined limits. Planning is done volumetrically: the beam weights for treating the entire target volume are generated simultaneously. The dose matrix is volumetric. The dose to each point is calculated to be that received from all beams and from all gantry angles. Dosage is calculated using a finite size pencil beam (FSPB) algorithm based on the beam characterization of clinically measured data. The degree to which a treatment plan is optimized is determined in part by constraints placed on the planning algorithm. The user has direct control over these constraints, which include dose goals to the target structures, dose limits to the sensitive structures, and the specification of arcs or fixed gantry positions in the treatment plan. CORVUS treatment plans need not have the isocenter located within the target volume. An unlimited number of targets falling within the treatment volume can be planned for at the same time. Dose may be prescribed for up to 32 structures, 29 of them userselectable, any number of which may be separate targets or radiation-sensitive structures. Each structure can have a separate dose prescription.

Here's an analysis of the provided text regarding the acceptance criteria and the study proving the device meets them, structured as requested:

Device Name: CORVUS Radiation Therapy Treatment Planning System (Model: CORVUS 09)

1. Table of Acceptance Criteria and Reported Device Performance

The provided document describes the CORVUS 09 as an update to an existing device (CORVUS 5.0M) and focuses on the equivalence of the updated device to its predicate, with specific improvements. Therefore, the "acceptance criteria" are implied to be equivalence to the predicate device in overall treatment planning quality and improved accuracy in specific areas.

2. Sample Size Used for the Test Set and Data Provenance

• Test Set Sample Size: The document does not specify a distinct "test set sample size" in terms of patient data or specific clinical cases.
  • For dosimetric validation, it states: "A total of 53 comparisons with measurement or comparisons with Monte-Carlo were completed." This refers to specific measurements against known standards.
  • For system functionality, it states: "A total of 62 system tests passed the criteria."
• Data Provenance: The document does not specify the country of origin of data or whether it was retrospective or prospective. The validation appears to be primarily focused on physical measurements and comparisons with Monte-Carlo simulations, rather than clinical patient data.

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

• Number of Experts: Not explicitly stated as a specific number.
• Qualifications of Experts: For the LDI algorithm's treatment plan quality validation, it states: "qualified personnel who are familiar with using an inverse treatment planning system in a clinical setting." For other validations, it refers to "medical physicists skilled in the art of conformal radiation therapy." No specific years of experience are mentioned.

4. Adjudication Method for the Test Set

The document does not describe an adjudication method (e.g., 2+1, 3+1) for establishing ground truth, as the testing primarily involves comparing device outputs to physical measurements (film, MOSFETs, ion-chambers) and Monte-Carlo simulations, or through direct comparison of plan quality by qualified personnel.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done

No. The document describes technical performance validation and comparison to a predicate device, not a human reader study in combination with AI assistance. Therefore, no effect size of human readers improving with/without AI assistance is provided. The device (CORVUS) is the planning tool, and ActiveRx (a feature of CORVUS 09) involves a user in the optimization process, but this is not described as an MRMC comparative effectiveness study where human readers interpret output to make diagnostic decisions.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done

Yes, implicitly. The "performance testing data" section focuses on the algorithm's accuracy (dosimetric validation, leakage calculation, system tests) independent of a human's final clinical interpretation, although a physician approves the plan. The system "suggests a plan" and "generates a set of beam weights" which are then reviewed. The validation studies for dose calculation and plan quality, comparing against physical measurements and Monte-Carlo, are standalone assessments of the algorithm's output.

7. The Type of Ground Truth Used

The ground truth for the performance testing was established using:

• Physical Measurements: Film measurements, MOSFETs (Metal-Oxide-Semiconductor Field-Effect Transistors), and ion-chambers.
• Computational/Simulated Data: Monte-Carlo calculations.
• Clinical/Expert Criteria: "Clinical criteria which are associated with conformal therapy" for evaluating treatment plan quality, as assessed by "qualified personnel" or "medical physicists."

8. The Sample Size for the Training Set

The document does not provide information on the sample size used for the training set. It describes optimization algorithms (simulated annealing, gradient descent) but not how a training set was used to develop or train the system. This suggests that the system's dose calculation and optimization methods are based on physics-based models and algorithms rather than a machine learning model that requires a distinct "training set" in the conventional sense.

9. How the Ground Truth for the Training Set Was Established

As no specific training set is mentioned in the context of machine learning, there is no information on how its ground truth was established. The system's underlying physics models and algorithms would have been developed based on established principles and characterized with machine-specific data mentioned in the "Indications for Use" section.

Ask a specific question about this device