(370 days)
This software device is intended to assist in formulating radiation treatment plans for modalities including stereotactic radiosurgery, external photon beam, external electron beam, and brachytherapy radiation treatment of benign or malignant disease processes.
The ADAC Pinnacle3 APEX™ device (APEX) is a radiation treatment planning software. It is an extension and enhancement of the currently marketed ADAC Pinnacle3 software (K926008). The APEX offers the identical imaging, display and stereotactic radiosurgery capabilities as the Pinnacle3, but its intended uses have been expanded to also include brachytherapy, external electron beam, and external photon beam treatment planning capabilities.
Here's an analysis of the provided text, focusing on the acceptance criteria and the study details:
Device: ADAC Pinnacle3 APEX™ device (APEX), a radiation treatment planning software.
Overview: The APEX is an extension of the ADAC Pinnacle3 software (K926008), expanding its intended uses beyond stereotactic radiosurgery to include brachytherapy, external electron beam, and external photon beam treatment planning.
1. Table of Acceptance Criteria and Reported Device Performance
The provided text does not explicitly state numerical "acceptance criteria" or "reported device performance" in a quantitative manner for most of its claims. Instead, it describes equivalence to predicate devices, use of established methodologies, and verification through specific types of testing.
| Feature/Intended Use | Acceptance Criteria (Implied) | Reported Device Performance |
|---|---|---|
| Stereotactic Radiosurgery | Identical capabilities and visualization functions to the predicate device (Pinnacle3 K926008). No adverse effect on safety and efficacy with the addition of other planning functions. | The APEX has identical stereotactic radiosurgery capabilities and visualization functions as the Pinnacle3. The addition of other planning functions has not affected the safety and efficacy for this use. |
| External Photon Beam | Use of a Superposition Convolution Model method for photon dose computation. Performance verified to conform to Task Group #23 of the American Association of Physicists in Medicine bench testing guidelines for photon beam planning software systems. Equivalent other aspects of photon therapy planning (entry of measured physics data, beam modifiers, energies, treatment machines) to predicates (ADAC XL CAD Plan K894706 and Precision Therapy Inc. Render-Plan 3-D K894722). | The APEX uses a Superposition Convolution Model method for photon dose computation, which iteratively models the computed beam profile by superimposing and convolving a kernel scaled in three dimensions according to patient densities.Bench testing conforming to TG#23 guidelines was conducted, with results verifying the performance of the Convolution method.Other aspects (physics data entry, beam modifiers, energies, machines) are equivalent to predicates. This method replaces 2D algorithms with correction factors used by predicates, providing absolute dose per unit fluence instead of relative dose. |
| External Electron Beam | Use of the same physics data entry to create dose lookup tables as predicates (CAD Plan and Render-Plan 3-D). Use of pencil beam algorithms based on Hogstrom et al. Ability to perform both 2-D and 3-D electron dose calculation. Support for 2-20 MeV electron energy. Use of the same dose analysis tools (dose volume histogram and beam weight optimization). | The APEX uses the same physics data entry and pencil beam algorithms (Hogstrom et al.) as predicates. It can perform both 2-D and 3-D calculations. It supports 2-20 MeV electron energy and uses the same dose analysis tools (dose volume histogram and beam weight optimization). |
| Brachytherapy Planning | Substantially equivalent to the ADAC XL CAD Plan System (K894706) in terms of dose calculation methods (Sievert Integral-equivalent table lookup). Provision of intracavitary and interstitial planning for same radiation sources (seeds, tubes, ribbons). Provision of planar display of isodose distribution. | The APEX uses the same Sievert Integral-equivalent table lookup method for dosage calculation and provides intracavitary and interstitial planning for the same source types as the CAD Plan. It also provides planar display of isodose distribution. Parallel testing demonstrated equivalence with the CAD Plan. |
| Brachytherapy - High Dose Rate (HDR) | Claim of HDR calculation and volumetric dose display, with the same dose algorithm as the predicate device (Render-Plan 3-D K894722). | The APEX includes HDR calculation and volumetric dose display, utilizing the same dose algorithm as the Render-Plan 3-D, capable of calculating at higher dose rates. |
| General Performance | Thorough performance, bench, parallel, and beta site testing conducted and verified prior to release. Software development, validation, and verification according to policies and procedures. Verification and Validation documentation available. | Protocols and results of performance testing (bench, parallel, beta site) are provided in Appendix D of the 510(k). Software development, validation, and verification conducted according to policies and procedures, including engineering diagrams, fault tree analyses, and Verification and Validation documentation (Appendix E). |
2. Sample Size Used for the Test Set and Data Provenance
The provided text does not specify sample sizes for test sets or data provenance (e.g., country of origin, retrospective/prospective). It refers to:
- Bench testing (for photon beam and general performance).
- Parallel testing (for brachytherapy).
- Beta site testing (for general performance).
These types of tests would typically involve a range of test cases, but the specific number of cases or patient data sources is not mentioned.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of those Experts
The document does not specify the number or qualifications of experts used to establish ground truth. It focuses on comparison to established predicate devices and adherence to professional guidelines (e.g., AAPM Task Group #23).
4. Adjudication Method for the Test Set
The document does not mention any adjudication method (e.g., 2+1, 3+1, none) 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
No MFRC comparative effectiveness study was done or mentioned. This device is a radiation treatment planning software that computes dose and assists in plan formulation, not a diagnostic AI system designed to assist human readers in image interpretation or diagnosis. Therefore, a study assessing human reader improvement with/without AI assistance in the traditional sense is not applicable to the description provided.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
Yes, the studies described are essentially standalone performance evaluations of the algorithm's accuracy and equivalence to predicate devices and established methodologies.
- For External Photon Beam, the "bench testing quidelines [sic] for photon beam planning software systems" from Task Group #23 of the AAPM evaluate the algorithm's performance in computing photon dose.
- For Brachytherapy, "parallel testing was conducted to demonstrate the equivalence of brachytherapy planning in these two systems," which implies a direct comparison of the software's output against a known or accepted standard (the predicate).
These tests assess the software's output and calculations directly, without human intervention as part of the performance metric.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)
The document implies the following types of "ground truth" or reference standards:
- Established Methodologies/Physics: For external photon beam, adherence to and verification against Task Group #23 of the American Association of Physicists in Medicine bench testing guidelines serves as a form of ground truth/standard. The Superposition Convolution Model itself is a well-established physics model.
- Predicate Device Equivalence: For stereotactic radiosurgery, external electron beam, and brachytherapy, the ground truth for establishing safety and effectiveness is largely based on equivalence to previously cleared predicate devices (ADAC Pinnacle3, ADAC XL CAD Plan, Precision Therapy Inc. Render-Plan 3-D). This means the predicates' established performance serves as the benchmark.
- Measured Beam Data: For photon dose calculation, the model "uses measured beam data to model iteratively the computed beam profile." This "measured beam data" can be considered a form of ground truth for characterizing the physical properties of the radiation.
8. The sample size for the training set
The document does not specify a sample size for a training set. As a radiation treatment planning software, it's more likely to be developed based on physics principles, mathematical models, and previously validated clinical data/libraries rather than a "training set" in the machine learning sense. The "measured beam data" mentioned for photon dose calculation might be extensive, but it's not described as a "training set" in the context of an algorithm learning from examples.
9. How the ground truth for the training set was established
Since a "training set" in the machine learning context is not explicitly mentioned, the method for establishing its ground truth is also not described. The underlying principles are based on established physics and previously validated clinical practices/device performances.
{0}------------------------------------------------
KGK1681
APR - 8 1996
SMDA 510(k) SUMMARY .
A. OVERVIEW
Ka51581
MAC LABORATORIES
540 ALDER DRIVE
Milpitas, CA 95035
This 510(k) application requests premarket clearance for the ADAC Pinnacle3 APEX™ device (APEX), a radiation treatment planning software manufactured by ADAC Laboratories of Milpitas, California. This software device is intended to assist in formulating radiation treatment plans for modalities including stereotactic radiosurgery, external photon beam, external electron beam, and brachytherapy radiation treatment of benign or malignant disease processes.
The APEX software system is an extension and enhancement of the currently marketed ADAC Pinnacle3 software (K926008), which also is also a radiation treatment software with a similar array of imaging. display, and dose computation capabilities. However, its approved intended use is only for stereotactic radiosurgery. The APEX offers the identical imaging, display and stereotactic radiosurgery capabilities, but its intended uses have been expanded to also include brachytherapy, external electron beam, and external photon beam treatment planning capabilities.
The predicate devices for the additional intended uses are the ADAC XL CAD Plan (K894706) and Precision Therapy Inc. Render-Plan 3-D (K894722). Each of the intended uses are discussed below.
STEREOTACTIC RADIOSURGERY B.
The Pinnacle3 is the predicate device for the APEX for the stereotactic radiosurgery intended use. The two devices have identical stereotactic radiosurgery capabilities and identical visualization functions. The addition of the other planning functions has not affected the safety and efficacy of the APEX for this use.
{1}------------------------------------------------
C. EXTERNAL PHOTON BEAM
The APEX software also contains the capability to formulate treatment plans for external photon beam therapy, as do both the CAD Plan and the Render-Plan 3-D. However, the APEX uses a Superposition Convolution Model method to compute photon dose, replacing the table lookup methods used by these predicates. This Convolution method uses measured beam data to model iteratively the computed beam profile. To compute dose, the associated kernel is superimposed on the dose grid and convolved with the primary fluence distribution. The kernel is then scaled in three dimensions according to patient densities (taken directly from patient images) to account for patient heterogeneities, beam modifiers, etc., to arrive at absolute dose per unit fluence.
The predicate devices use two dimensional algorithms to calculate photon dose, whereby measured beam data is used with a series of correction factors to account for patient heterogeneities, beam modifiers, etc., for a manually specified patient contour. This algorithm calculates relative dose per unit fluence.
The other aspects of photon therapy planning in the APEX system -- entry of measured physics data, beam modifiers, the energies and treatment machines supported -- are equivalent to the predicates.
Task Group #23 of the American Association of Physicists in Medicine has produced bench testing quidelines for photon beam planning software systems. Testing of the photon beam performance of the APEX that conforms to these guidelines has been conducted. The results verify the performance of the Convolution method for computing photon dose, and is provided in the 510(k).
{2}------------------------------------------------
D. EXTERNAL ELECTRON BEAM
For external electron beam planning, the predicates for the APEX are again both the CAD Plan and the Render-Plan 3-D devices. All three use the same physics data entry to create the dose lookup tables, and use pencil beam algorithms based on those created by Hogstrom et al to calculate dose. The APEX is able to perform both 2-D and 3-D electron dose calculation. All three systems support 2 - 20 MeV of electron energy, and use the same dose analysis tools (dose volume histogram and beam weight optimization.)
BRACHYTHERAPY E.
For brachytherapy planning, the APEX system is substantially equivalent to the ADAC XL CAD Plan System in terms of dose calculation methods. Both systems provide intracavitary and interstitial planning for the same types of radiation sources, including seeds, tubes and ribbons. The same Sievert Integral-equivalent table lookup method is used to compute dosage. Both the APEX and CAD Plan provide planar display of isodose distribution. Parallel testing was conducted to demonstrate the equivalence of brachytherapy planning in these two systems, and is provided in the 510(k).
The APEX also contains two additional features: calculation for High Dose Rate (HDR) therapy and volumetric dose display. The CAD Plan is capable of performing HDR calculation but this claim was not included in the prior 510(k) submission. Instead, the Render-Plan 3-D radiation treatment planning software (Precision Therapy Inc., K894722) which includes high dose rate as a claim in its labeling, serves as the predicate for this feature and for volumetric dose display. The HDR function for the APEX and Render-Plan contains the same dose algorithm capable of calculating at the higher dose rate.
{3}------------------------------------------------
PERFORMANCE TESTING F.
The performance testing, including bench testing, parallel testing, and beta site testing will be thoroughly performed and verified prior to release of the product. The protocols and results of these tests are provided in Appendix D of the 510(k).
SOFTWARE DEVELOPMENT AND VERIFICATION G.
Software development, validation and verification of the APEX system has been conducted according to the policies and procedures discussed in the 510(k), which includes engineering diagrams, fault tree analyses, the Verification and Validation documentation, and all pertinent reference articles. These documents can be found in Appendix E.
H. LABELING
All draft labeling associated with the APEX has been presented in the 510(k) document, including the draft User's Guide, the Physics Guide, ,an outline of the training program, and promotional literature. The labeling available for the predicate devices -- the Pinnacle3, the CAD Plan, and the Render-Plan -- are also provided.
§ 892.5050 Medical charged-particle radiation therapy system.
(a)
Identification. A medical charged-particle radiation therapy system is a device that produces by acceleration high energy charged particles (e.g., electrons and protons) intended for use in radiation therapy. This generic type of device may include signal analysis and display equipment, patient and equipment supports, treatment planning computer programs, component parts, and accessories.(b)
Classification. Class II. When intended for use as a quality control system, the film dosimetry system (film scanning system) included as an accessory to the device described in paragraph (a) of this section, is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 892.9.