(106 days)
DoseLab Pro is quality assurance software intended to be used as part of a dosimetry verification system for linear accelerators. It can be used to import radiation-exposed images from scanned film, other measurement devices, and treatment planning systems to display differences between measured and calculated dose distributions.
DoseLab Pro is a software-only device that uses image analysis to perform radiation oncology quality assurance (QA) as part of a dosimetry verification system. Images are useful in radiation oncology QA because they can be analyzed qualitatively by viewing them and quantitatively using mathematical routines on the data that composes them. A variety of data sets can be analyzed as images in DoseLab Pro. They include radiation dose distributions calculated by treatment planning systems, measured dose distributions from arrays (diode and ion chamber), and radiation-exposed film images.
DoseLab Pro uses numerous built-in image analysis routines that have been developed to perform the tests and meet the standards of the medical physics QA community. In particular, these tools were designed to specifically support completing dose comparisons. Dose comparisons are made between two, two-dimensional images containing radiation dose and spatial information. The first image is exported from the dose calculation of a patient-specific computed treatment plan from treatment planning software, while the second image is the measured dose from delivery of that plan captured by film or a measurement array. DoseLab Pro assists in aligning the images spatially before performing several different comparisons including Gamma analysis and normalization. DoseLab Pro additionally contains tools for image editing, film image import, and film calibration.
It is important to note that while DoseLab Pro operates in the field of radiation therapy, it is neither a radiation delivery device (e.q. a linear accelerator), nor is it a treatment planning system (TPS). DoseLab Pro is an analysis tool meant solely for quality assurance (QA) purposes when used by trained medical professionals. Being a software-only QA tool, DoseLab Pro never comes into contact with patients.
The provided text is a 510(k) summary for the DoseLab Pro device, a quality assurance software for linear accelerators. It describes the device, its intended use, and claims substantial equivalence to a predicate device based on non-clinical performance data.
Here's a breakdown of the requested information based on the provided text:
1. Table of Acceptance Criteria and Reported Device Performance
The document does not explicitly present a table of acceptance criteria with corresponding performance metrics. Instead, it states:
| Criterion | Reported Performance |
|---|---|
| Software functionality as per design | All tests passed without defect. |
| Expected behavior and output | Confirmed with known good data for inputs. |
| Device performs quality assurance comparisons between TPS images and measured dose images | Stated as the principal technological characteristic and demonstrated through testing. |
2. Sample size used for the test set and the data provenance
The document does not specify a numerical sample size for the test set. It mentions "known good data for inputs," but the number of such data points is not quantified.
- Sample size for test set: Not specified numerically.
- Data provenance: Not explicitly stated, given it's referred to as "known good data for inputs." It implies synthesized or carefully selected internal data rather than real-world retrospective or prospective clinical data from a specific country. However, the context of "dosimetry verification system for linear accelerators" suggests that this "known good data" would be representative of radiation oncology QA scenarios.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts
The document does not mention the involvement of external experts to establish ground truth for the test set. The validation testing was performed "manually on the fully compiled software," implying an internal validation process based on the software's design specifications.
4. Adjudication method for the test set
No adjudication method (e.g., 2+1, 3+1) is mentioned in the document. The testing appears to be a direct comparison of software output against expected output based on "known good data," rather than a consensus-based ground truth establishment.
5. If a multi-reader multi-case (MRMC) comparative effectiveness study was done, and 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 done. DoseLab Pro is described as an image analysis tool for quality assurance, not a diagnostic or decision-support AI intended to be used with human readers to improve their performance in the traditional sense of medical image interpretation (e.g., radiologists interpreting images for disease detection). It performs calculations and displays differences between measured and calculated dose distributions, aiding medical physics QA, but not in a way that would typically involve an MRMC study comparing human reader performance.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
Yes, a standalone performance evaluation was done. The "Non-Clinical Performance Data" section describes "Testing involved the use of known good data for inputs into DoseLab Pro and execution of tests designed to confirm expected behavior and expected output." This is a standalone evaluation of the software's functionality and accuracy in performing its intended QA tasks. The software itself is the "algorithm only" in this context, as it's not designed to be a human-in-the-loop diagnostic AI that modifies human reader performance. Its role is to independently carry out calculations and comparisons.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)
The type of ground truth used was based on "known good data" and "expected behavior and expected output." This implies a set of pre-defined, analytically derived, or previously validated outputs for specific inputs, against which the software's results were compared. It's an internal validation against known correct values, not an external ground truth like pathology or expert consensus on patient outcomes.
8. The sample size for the training set
The document does not mention a "training set." DoseLab Pro is described as software with "numerous built-in image analysis routines that have been developed to perform the tests and meet the standards of the medical physics QA community." This suggests that it is rule-based or algorithm-driven software, rather than a machine learning or AI model that requires a distinct training phase with a labeled dataset. Therefore, the concept of a "training set" as typically understood in machine learning is not applicable here.
9. How the ground truth for the training set was established
As there is no mention of a training set, the establishment of ground truth for a training set is not applicable. The software's design and built-in routines are based on established medical physics QA standards.
§ 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.