A JavaScript-based computational method for predicting decay activity of 131I: development, benchmark- and phantom-based assessment of an open-access research tool

This study investigated the use of JavaScript (JS) to develop and test an open-access method, provided as a research resource, for predicting the decay and cumulative activity of iodine-131 (131I). A client-side JS engine was implemented to simulate decay using mono-exponential and piecewise multi-segmented models. Built-in JS functions were used to perform logarithmic and exponential operations. The tool's performance was validated against ion-chamber measurements from a thyroid phantom containing a pre-calibrated 131I capsule; moreover, measurement reliability was assessed. The phantom test showed a mean absolute error of 0.010 μSv/h for the mono-exponential model and 0.008 μSv/h for the multi-segmented model. The intraclass correlation coefficient (ICC) demonstrated excellent inter-observer agreement (ICC range: 0.998-0.999), while the mean intra-observer difference was 0.3% (SD = 0.2%; range: 0.1%-0.6%). By quantifying instrument-related error in advance, future studies may more confidently isolate biological variability, thereby supporting the robustness of patient-level interpretation and dose-prediction models.