PET block detector calibration using subtractive clustering algorithm and comparison with hough transform algorithm

Detector calibration plays an important role in improving of image quality and increasing performance of positron emission tomography (PET) systems. To achieve this aim, raw data coordinate event-byevent are mapped to the index of the crystal in which the particle is absorbed. We proposed and tested subtractive clustering and Hough transform algorithms to determine crystal peak position and generate an appropriate look-up table. The results show superiority of Hough transform to the subtractive clustering and other methods because this algorithm determines position of all peaks, even in irregular and unclear data. The acquired results can be beneficial for all of the medical imaging instruments such as PET and single-photon emission computed tomography detectors based on pixilated scintillators.