Volumetry of subdural hematomas in computed tomography images: ABC methods versus an intelligent computational technique

Abstract

This work evaluates the performance of some methods orientedtowards the generation of the volume of four subduralhematomas (SDH), present in multi-layer computed tomographyimages. To do this, firstly, a reference volume is specified:the volume obtained by a neurosurgeon using the manualplanimetric method (MPM); which allows the generation ofmanual segmentations of space-occupying lesions. In thiscase, these volumes are matched with the SDH. In parallel,the volumetry of the 4 SDHs is obtained, considering both theoriginal version of the ABC/2 method and two of its variants,identified in this paper as ABC/3 method and 2ABC/3 method.The ABC methods allow the calculation of the volume ofthe hematoma under the assumption that the SDH has anellipsoidal shape. In third place, SDH’s are studied throughan intelligent automatic technique (SAT) that generates thethree-dimensional segmentation of each SDH. Finally, thepercentage relative error is calculated as a metric to evaluatethe methodologies considered. The results show that the SATmethod exhibits the best performance generating an averagepercentage error of less than 5%.