Kalogeropoulou Christina, Zampakis Petros, Tsimara Maria, Panagopoulos Konstantinos, Hatzilygeroudis Ioannis and Koutsojannis Constantinos
Background: According to previous studies, CT scan exhibits the highest sensitivity and specificity rates among the imaging methods for parathyroid glands. The purpose of this work is the development and evaluation of an intelligent decision support system for the appropriate surgical intervention of parathyroid gland based on the evaluation of CT images for preoperative mapping. additionally to the traditional technique in distinguishing mono/multiglandular parathyroid pathology assisting precise preoperative localization of adenomas or hyperplasia and offering the best choice for minimal or extensive surgical exploration. Methods: Based on simple clinical and CT imaging information and criteria such as CT diagnosis, location of enlarged parathyroid gland, density in ΗU before IV contrast enhancement, increase of enhancement, volume of pathological parathyroid and final surgical results, data mining algorithms were used to develop the knowledge base for decision making. As training data the records from 118 histologically proven PHPT patients referred to our Hospital University Department of Radiology-Computer Tomography were used on a retrospective chart analysis. As test database the records from histologically proven PHPT patients were used. Results: The final developed intelligent system produced succeeded to predict an acceptable outcome for all participants who emerged surgical therapy and compared to the surgical results revealed minimal false positive rate and excellent positive and negative likelihood ratios. Conclusion: The final intelligent system has provided better performance than the commonly used imaging criteria in predicting the surgical of multi- or single-glandular disease.
English 
Spanish 
Chinese 
Russian 
German 
French 
Japanese 
Portuguese 
Hindi