Theory and Practice of Risk Assessment, Springer Proceedings in Mathematics and Statistics
The use of image recognition and classification of objects according to images is becoming extremely
popular, especially in the field of medicine. A mathematical procedure allows us, not only to evaluate the amount of data per se, but also ensures that each image is processed similarly. Our study has two focal points: The first one is the automated data entry and the second one is the evaluation in a manageable way. We propose the use of mathematical procedures to support the applicants in their evaluation of magnetic resonance images (MRI) of renal tumours. Therapy of renal tumours in childhood based on therapy optimizing SIOP (Society of Pediatric Oncology and Hematology)-study protocols in Europe. The most frequent tumour is the nephroblastoma (over 80 %). Other tumour entities in the retroperitoneum are clear cell sarcoma, renal cell carcinoma and extrarenal tumours, especially neuroblastoma. Radiological diagnosis is produced with the help of cross sectional imaging methods (computer tomography CT or Magnetic Resonance Images MRI). Our research is the first mathematical approach on MRI of retroperitoneal tumours for transversal images (of 40 patients). We use MRI in 3 planes and evaluate their potential to differentiate other types of tumours. We determine the key points or three dimensional landmarks of retroperitoneal tumours in childhood by using the edges of the platonic body (C60) and test the difference between the groups (nephroblastoma versus non-nephroblastoma). The size is not eliminated like in former studies. All objects are comparable. For other important references see [1, 4, 10].