Search Results

DoseCalc is a program utilized in a radiation therapy department for the determination of monitor units. The Monitor Unit is a quantity used by a treatment machine for determining the length of time that it should remain on in order to deliver a prescribed amount of radiation to a point. Most of the time, a radiation treatment planning system is used to define the location and dose distribution of the radiation. Radiation therapy planning systems typically calculate the monitor units needed to deliver the desired amount of reduction to a point of reference and this produce the desired dose distribution within the patient. On occasion, a hand calculation will be performed in order to determine the number of monitor units needed to deliver the prescribed amount of radiation. DoseCalc is a program which allows the user to input data by hand through the use of the mouse or keyboard or electronically from the primary radiation therapy planning system. From this data, it will then determine the number of monitor units needed to be given to a patient. The process of calculating the monitor units involves DoseCalc automatically looking up the data parameters from previously inputed data and then calculating the monitor units from these values. This process greatly increases the speed at which a calculation can be performed and also eliminates many errors that occur from manually looking up the data. DoseCalc's monitor unit calculation can then be used to validate the monitor units previously determined by hand or by the primary radiation therapy planning system. It is not the intention of DoseCalc to replace the calculation performed by the primary radiation therapy planning computer but to validate its calculation as a means of quality assurance. The practice of performing a secondary check is recommended by the American Association of Physicists in Medicine (AAPM) Task Group 40 as part of good quality assurance program. This practice is an important aspect in providing quality patient care. DoseCalc is not only being submitted to perform this secondary function but to also be used as the primary means of calculating monitor units in situations where the physician does not order the use of a radiation therapy treatment plan. For this situation, it is important to accurately determine the monitor units needed for a patient's treatment. DoseCalc provides this operation. It has many built in checks that will check for many common errors that occur when calculating monitor units as well as checking that the inputted parameters are within predefined limits for the treatment machine. The use of DoseCalc in this manner provides a means for accurately determining the monitor units. A physicist can then visually examine the inputted data for accuracy and verify the computed parameters to be sure that they are correct.

DoseCalc also allows for the transfer of the treatment planning data from the primary radiation therapy planning computer to DoseCalc and then to the facility's Verify and Record system. This will reduce the number of errors that occur as a result of manually inputting this data. This feature merely transfers information from one system to another without performing any calculations.

DoseCalc is a software program that is designed to operate on a PC in a Windows environment on either a stand alone PC or on a server. It does not control any radiation hardware device but does interface with the primary radiation therapy planning software and verify and record software. The device performs monitor unit calculations for photon beams which can be used to validate monitor units calculated by the primary radiation therapy planning system or to simply provide the monitor units needed to treat a patient when a radiation therapy plan is not prescribed by the physician. DoseCalc determines the monitor units through the process looking data up from previously inputted tables.

The provided text describes the DoseCalc v1.02 software, a program for calculating monitor units in radiation therapy. However, it does not contain specific acceptance criteria or a detailed study report with quantitative results to fill in the table and answer all questions comprehensively. The document focuses on demonstrating substantial equivalence to a predicate device (muCheck) rather than presenting a formal efficacy study with defined acceptance criteria and performance metrics.

Here's an attempt to extract relevant information and note what is missing based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

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

• Test Set Sample Size: Not explicitly stated. The "Non-Clinical Tests" involved "numerous monitor unit calculations under various situations." "Beta Testing" involved "numerous copies of actual calculations."
• Data Provenance:
  • Non-Clinical Tests: "Hand calculations for the same situations" (presumably synthetic or simulated scenarios).
  • Beta Testing: "Actual calculations" from East Texas Medical Center in Tyler, Texas. These were compared to calculations from ADAC's Pinnacle3 APEX (K951581) or "hand calculations." This implies a mix of prospective (real patient data processed during the beta test) and potentially retrospective (past patient cases replicated) data.

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

• Number of Experts: Not explicitly stated.
• Qualifications of Experts: The ground truth for "hand calculations" would presumably be established by qualified medical physicists. The submitter, Craig A. Laughton, is a Medical Physicist with a Master's Degrees and 4 years of clinical experience, working with a PhD Physicist with over 25 years of experience. These individuals would be considered experts in setting up and verifying such calculations. The predicate device's calculations and "primary radiation therapy planning system" (ADAC's Pinnacle3 APEX) also serve as a form of "ground truth" or reference.

4. Adjudication method for the test set

• Adjudication Method: Not explicitly detailed. The comparison involved DoseCalc's calculations against "hand calculations" and/or calculations from "the primary radiation therapy planning system." The process seems to be a direct comparison, rather than an adjudicated consensus between multiple independent experts on each case.

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

• MRMC Study: No, an MRMC comparative effectiveness study was not performed. The device is a software program for calculating monitor units, primarily for validation or as a primary calculation tool, not an AI for human interpretation of images or data where "human readers improve with AI vs without AI assistance" would be a relevant metric.

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

• Standalone Performance Study: Yes, the described "Non-Clinical Tests" and "Beta Testing" sections primarily evaluate the algorithm's performance in a standalone capacity (i.e., its ability to calculate monitor units accurately when given inputs), with the resulting calculations then being compared to references (hand calculations, other TPS). While a physicist "can then visually examine the inputted data for accuracy and verify the computed parameters," this is a quality control step, not part of the primary performance evaluation of the algorithm's calculation capability itself.

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

• Type of Ground Truth: The ground truth was primarily established through:
  • Expert Hand Calculations: Manual calculations performed by qualified physicists.
  • Reference System Calculations: Output from an established "primary radiation therapy planning system" (ADAC's Pinnacle3 APEX) and the predicate device (muCheck).

8. The sample size for the training set

• Training Set Sample Size: Not explicitly stated. The document mentions that DoseCalc looks up data from "previously inputted tables" and has "many built in checks that will check for many common errors that occur when calculating monitor units as well as checking that the inputted parameters are within predefined limits for the treatment machine." This implies that the system was developed and configured based on established physics principles and possibly a substantial amount of data (e.g., machine-specific data, historical error patterns), but a formal "training set" in the context of machine learning (where this question typically applies) is not described.

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

• Ground Truth for Training Set: Since a formal machine learning "training set" isn't explicitly described, the concept of "ground truth for training" as typically understood in AI is not applicable here. The "training" of this system would be more akin to populating its lookup tables and programming its calculation rules based on established medical physics principles, machine calibration data, and potentially historical data on common errors. This "ground truth" would be derived from:
  • Medical Physics Principles and Formulas: The underlying equations and methodologies for calculating monitor units.
  • Machine-Specific Data: Data related to the specific treatment machines (e.g., beam profiles, output factors) that are "previously inputted tables."
  • Expert Knowledge: The expertise of the medical physicists developing the software, incorporating their knowledge of errors and limits.

Ask a specific question about this device