(160 days)
Intuition Image's indications for use is to prepare and present patient and image data based on CT, MR, Angiographic and other imaging sources including
- image preparation -
- image localization -
- image fusion -
- image segmentation -
- isocenter handling -
- plan review and approval where the result is used for stereotactic radiation treatment planning that is intended for use in stereotactic, conformal, computer planned, LINAC based radiation treatment of cranial, head and neck and extracranial lesions.
Intuition Image is a software tool running on a standard, standalone computer workstation or being accessible via the intranet connection for pre-planning of treatments based on stereotactic systems.
The system provides e.g. tools for the automatic or manual segmentation of anatomical structures, which helps the user such as the radiologist or the neurosurgeon to quickly achieve the desired segmentation results through a variety of automatic and/or manual re-segmentations. Additionally anatomical and functional structures and segmentations of the human brain as defined and described by Talairach/Tournoux and/or Schaltenbrand/Wahren brain atlases can be correlated with the patient's brain data.
The created treatment plans of Intuition Image can be used on its own or in conjunction with other BrainLAB treatment planning systems such as Intuition Dose (to be developed) or BrainSCAN (K994413) for further planning of parameters, which are relevant for Radiotherapy/Radiosurgery. Intuition Image may also serve as a pre-planning station for various third party treatment planning systems.
The provided 510(k) summary for Intuition Image (K032511) does not contain detailed information about specific acceptance criteria or an explicit study proving the device meets these criteria in the way a modern AI/ML device submission would.
The approval is based on "substantial equivalence" to predicate devices (K020631 and K994413), and the provided text highlights general validation and verification according to BrainLAB's procedures. This pre-dates the current rigorous expectations for AI/ML device performance validation.
Therefore, many of the requested details cannot be extracted from the given text. Below is an attempt to answer what can be inferred or explicitly stated, with clear indications where information is missing.
Acceptance Criteria and Device Performance
1. A table of acceptance criteria and the reported device performance:
The document does not provide a table of specific acceptance criteria (e.g., minimum accuracy, sensitivity, specificity, or segmentation precision) or quantified performance metrics for Intuition Image. The primary "performance" reported is that the device "has been verified and validated according to BrainLAB's procedures for product design and development" and that this "validation proves the safety and effectiveness of the system."
| Metric / Feature | Acceptance Criteria (Not explicitly stated) | Reported Device Performance | Comments / Interpretation |
|---|---|---|---|
| Overall Functionality | Device performs intended functions for image preparation, localization, fusion, segmentation, isocenter handling, and plan review/approval. | "Verified and validated according to BrainLAB's procedures for product design and development." | Implies software functions as designed and fulfills its stated intended use. |
| Safety | No undue risk to patients or users. | "validation proves the safety and effectiveness of the system." | General assurance of safety. |
| Effectiveness | Device achieves intended clinical purpose. | "validation proves the safety and effectiveness of the system." | General assurance of effectiveness. |
| Segmentation Accuracy | (Not specified) | Tools for "automatic or manual segmentation of anatomical structures, which helps the user such as the radiologist or the neurosurgeon to quickly achieve the desired segmentation results through a variety of automatic and/or manual re-segmentations." | Implies the tool facilitates accurate segmentation by the user, but no inherent automated accuracy metric is provided for the device itself. |
| Equivalence to Predicates | Functionally similar and equally safe/effective as iPlan! (K020631) and BrainSCAN (K994413). | "found to be substantially equivalent with the predicate devices." | The core of the 510(k) approval process for this device. |
Study Details (Based on available information)
2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective):
- Sample Size: Not specified.
- Data Provenance: Not specified. The document states BrainLAB AG is in Germany, but this does not confirm the origin of any test data.
- Retrospective or Prospective: Not specified.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience):
- Number of Experts: Not specified.
- Qualifications of Experts: Not specified. It mentions the "user such as the radiologist or the neurosurgeon" for segmentation, implying these are the intended experts for using the tool, but not necessarily for establishing ground truth in a formal test.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set:
- Not specified.
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 MRMC comparative effectiveness study is described or implied in the provided text. The device is a "pre-planning station" and software tool, not an AI-assisted diagnostic aid in the modern sense that would typically require such a study for its primary claim (though the "automatic segmentation" could be seen as an early form of AI-assistance).
6. If a standalone (i.e. algorithm only without human-in-the loop performance) was done:
- The device description states it provides "tools for the automatic or manual segmentation." While "automatic segmentation" implies a standalone algorithmic component, no specific standalone performance metrics for this automatic function are reported. The focus is on the user leveraging these tools to achieve results.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc):
- Not specified. Given the nature of a planning system, it's likely that ground truth for segmentation or localization would involve expert anatomical delineation on imaging, potentially with cross-validation from multiple modalities or surgical confirmation contextually, but this is not detailed.
8. The sample size for the training set:
- This device predates the common terminology and strict requirements for "training sets" as understood in modern AI/ML submissions. Therefore, a training set size is not mentioned, as the validation would have focused on a more traditional software verification and validation approach rather than an AI model training and testing paradigm.
9. How the ground truth for the training set was established:
- As no training set is mentioned, the method for establishing its ground truth is also not specified.
Summary of Missing Information:
The provided 510(k) summary is typical for its era (2003) and for device types that are primarily software tools aiding clinicians, rather than standalone diagnostic AI algorithms. It emphasizes "substantial equivalence" to predicate devices and general verification/validation of software functionality. The granular details now expected for AI/ML device submissions, such as specific performance metrics, detailed study designs, ground truth establishment methods, and training/test set sizes, are largely absent.
§ 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.