(454 days)
Delta4 Insight is a software intended to provide quality assurance of radiotherapy treatment dose calculation.
Delta4 Insight is not a treatment planning system or a radiation delivery device. Information provided by Delta4 Insight shall not be used to directly modify or influence radiation treatments. Delta4 Insight is to be used radiation oncology personnel for quality assurance purposes.
Delta4 Insight is software specifically design for quality assurance of radiotherapy treatment plans generated by a treatment planning system. The device calculates a secondary dose calculation via an independent Monte-Carlo based dose calculation software and compares this to the treatment planning system dose. The device is used as a secondary check for the results of a TPS and not for comparison with a measurement.
Delta4 Insight is a software module within the General Delta4 software. The software module is independent of all other software modules within the Delta4 software. The device is NOT a treatment planning system or a radiation therapy delivery device. It is only used by trained radiation therapy oncology personnel for the purposes of quality assurance in a hospital setting.
Insight supports treatments with MV photons. No other radiation type is supported.
Here's a breakdown of the acceptance criteria and the study details for the Delta4 Insight device, based on the provided FDA 510(k) summary:
1. Table of Acceptance Criteria and Reported Device Performance
| Acceptance Criteria (Used by Predicate) | Reported Delta4 Insight Performance |
|---|---|
| 97% of dose voxels with >5% of maximum dose pass a gamma criteria of 2%/2mm when comparing patient plan dose calculation results to a reference algorithm (Acuros). [Hoffman et al, MedPhys 2018] | Similar gamma pass rate to that of the predicate when comparing patient plan dose calculation results to a reference algorithm (Monaco, Acuros). |
2. Sample Size Used for the Test Set and Data Provenance
- Sample Size: "Test plans were chosen to span the space of test parameters for the types of treatment types/modalities/energies/TPS/machines that are supported. The variety of plans tested are reflective of common clinical treatments and tumor sites and span many field sizes and tissue depths."
- Interpretation/Clarification: The exact number of test plans used is not specified. However, the description implies a diverse set of clinically relevant plans.
- Data Provenance: "The dose from anonymized DICOM patient treatment plans from chosen reference algorithms were re-calculated with Insight and the dose compared."
- Interpretation/Clarification: The data consists of retrospective, anonymized DICOM patient treatment plans. The country of origin is not specified but is likely from European or North American oncology departments given the context of FDA submission and common practices.
3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications
- The document does not explicitly state that human experts established a "ground truth" for the test set in the traditional sense of clinical assessment.
- Instead, the comparison is made against "reference algorithms" (Acuros, Monaco) which themselves are established dose calculation engines. The accuracy of these reference algorithms is implicit, and one (Acuros) is cited with a publication (Hoffman et al, MedPhys 2018).
- The study focuses on the agreement between software calculations, not expert visual assessment.
4. Adjudication Method for the Test Set
- The concept of "adjudication" (e.g., 2+1, 3+1 expert review) is not applicable to this type of study.
- The comparison method used is a gamma comparison between the Delta4 Insight dose results and the original reference TPS dose. Gamma analysis is a common method in radiation oncology for quantitatively comparing dose distributions.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done
- No, a MRMC comparative effectiveness study was not done.
- This device is a "secondary check QA software" designed for an algorithm-to-algorithm comparison (Delta4 Insight's calculation vs. Treatment Planning System's calculation), not for human reader performance evaluation. Therefore, the effect size of human readers improving with AI vs. without AI assistance is not relevant to this submission.
6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done
- Yes, a standalone performance evaluation was indeed done.
- The study explicitly describes comparing "the dose from anonymized DICOM patient treatment plans from chosen reference algorithms were re-calculated with Insight and the dose compared." This is a direct measure of the algorithm's performance in matching established reference calculations.
7. The Type of Ground Truth Used
- The "ground truth" in this context is the dose calculation results from established "reference algorithms" (Monaco, Acuros) within existing Treatment Planning Systems (TPS). It is not pathology, outcomes data, or human expert consensus in the diagnostic imaging sense. It is a comparison against other validated computational models for dose calculation.
8. The Sample Size for the Training Set
- The document does not specify a separate training set size.
- The description focuses on "validation testing involved testing of clinical treatment plans" and "test plans were chosen to span the space of test parameters." This suggests that the device's algorithms were likely developed and refined using internal data, but the 510(k) summary only describes the validation/test set and its performance.
9. How the Ground Truth for the Training Set Was Established
- Since a training set is not explicitly mentioned with its own ground truth establishment in this document, this information is not provided.
- Typically, for a dose calculation algorithm, "ground truth" during development (training) would involve comparing the algorithm's output against physical measurements (e.g., phantom measurements, ionization chamber data) and/or highly accurate, gold-standard Monte Carlo simulations for a variety of beam configurations and patient geometries. However, this 510(k) summary focuses on the comparative performance against other clinical TPS algorithms for validation.
§ 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.