HEURISTIC ALGORITHM FOR PET IMAGES’ SEGMENTATION USING ARTIFICIAL INTELIGENCE TECHNIQUES

The overall quantity of nuclear medicine procedures has increased remarkably in recent years, making them a daily tool capable reaching wide sectors the population. Regarding therapeutic applications, it is worth noting that there an increasing demand novel techniques and greater variety radioisotopes requiring accurate patient-specific dosimetry aimed at evaluating lethal damage to tumor while maintaining acceptable dose levels healthy tissues. Image-guided internal appears as particularly suitable for theranostics procedures, which allow joint implementation diagnose treatment. In this case, correct segmentation images critical identification different tissues organs. On other hand, modern tools based on data science artificial intelligence have spread several fields, digital image processing. use machine learning models processing promising opportunity complement clinical analysis by experts. This paper reports about unsupervised heuristic algorithm using clustering together, two algorithms: K-Means HDBSCAN. results obtained highlight capacity automatic means algorithms, becoming useful assist clinician experts shorten times.