Abstract
New research on Artificial Intelligenc e is the subject of a report. According to news reporting from Heidelberg, Germa ny, by NewsRx journalists, research stated, "Target volume delineation is routin ely performed in postoperative radiotherapy (RT) for breast cancer patients, but it is a timeconsuming process. The aim of the present study was to validate th e quality, clinical usability and institutional-specific implementation of diffe rent auto-segmentation tools into clinical routine." The news correspondents obtained a quote from the research from Heidelberg Unive rsity Hospital, "Three different commercially available, artificial intelligence -, ESTRO-guideline-based segmentation models (M1-3) were applied to fifty consec utive reference patients who received postoperative local RT including regional nodal irradiation for breast cancer for the delineation of clinical target volum es: the residual breast, implant or chestwall, axilla levels 1 and 2, the infra- and supraclavicular regions, the interpectoral and internal mammary nodes. Obje ctive evaluation metrics of the created structures were conducted with the Dice similarity index (DICE) and the Hausdorff distance, and a manual evaluation of u sability. The resulting geometries of the segmentation models were compared to t he reference volumes for each patient and required no or only minor corrections in 72 % (M1), 64 % (M2) and 78 % (M3) of the cases. The median DICE and Hausdorff values for the resulting planning ta rget volumes were 0.87-0.88 and 2.96-3.55, respectively. Clinical usability was significantly correlated with the DICE index, with calculated cut-off values use d to define no or minor adjustments of 0.82-0.86. Right or left sided target and breathing method (deep inspiration breath hold vs. free breathing) did not impa ct the quality of the resulting structures."
NETL
NSTL