Verification of Normality in the Source Intensity Detection Rate for Batch Assay Using a Well-Type Ionization Chamber

The American Association of Physicists in Medicine (AAPM) advocates for the importance of source intensity verification at the facility of use. At our facility, we have continuously performed source intensity verification on all blister packs delivered since initiating 125I seed brachytherapy. The single-seed assay recommended by the AAPM is difficult to implement domestically. As an alternative method, batch assays using an ionization chamber or a well-type ionization chamber to measure multiple seeds simultaneously have been proposed. Our facility calculates the source strength detection rate (detection rate) from measurements using a well-type ionization chamber to verify the accuracy of source strength. Understanding the characteristics of these measurements is crucial for their proper handling. Previous studies have reported verifications for the model STM1251 (C.R. Bard, Murray Hill, NJ) and the OncoSeed model 6711 (General Electric Healthcare, Barrington, IL). However, to our knowledge, there are no reports verifying whether the detection rate for the TheraAgX100 (Theragenics Corporation, Buford, GA) used at our facility is normally distributed. The purpose of this study is to verify the normality of detection rates based on data accumulated to date. The measurement subjects were the five blister packs with the highest number of deliveries among the delivered TheraAgX100 blister packs, totaling 262 cases. For all of these, the detection rate was calculated from the measurements obtained using the CRC-15R well-type ionization chamber (Capintec Corp., Ramsey, NJ). Normality was verified using Statistical Product and Service Solutions (SPSS, version 29; IBM SPSS Statistics for Windows, Armonk, NY). The Shapiro-Wilk test was used for the test, with a significance level of 0.05. The results of the normality test showed a Shapiro-Wilk test significance probability of 0.623, confirming that the detection rate follows a normal distribution. For batch assays, the nominal value must be calculated from the accumulated measurements at the facility. In other words, without confirming the normality of the detection rate, it is unclear whether the mean or median should be adopted as the nominal value. This study demonstrated that, for the combination of radiation sources and dosimeters at this facility, the detection rate follows a normal distribution. Therefore, it is valid to use either the mean or the median of the detection rate as the nominal value. The guidelines describe the response when batch assay results deviate from the nominal value. Specific control limits, such as tolerance limits and intervention limits, are indicated. For batch assays, the nominal value must be calculated based on the cumulative detection rate at the facility. If the detection rate does not follow a normal distribution, using the mean or median from the batch assay as the nominal value is not appropriate. This study demonstrated the validity of using the mean or median of the cumulative detection rate as the nominal value. We plan to verify this approach for blister packs containing different numbers of units. Furthermore, since the shape of the radiation source varies by product, re-evaluation is necessary when new products are introduced. Understanding the characteristics of the acquired measurements and applying this knowledge to clinical practice is a critical responsibility of medical physicists in charge of quality assurance.