RESUMEN
AIM: To propose an automated image processing routine based on free software to quantify root canal preparation outcomes in pairs of sound and instrumented roots after micro-CT scanning procedures. METHODOLOGY: Seven mesial roots of human mandibular molars with different canal configuration systems were studied: (i) Vertucci's type 1, (ii) Vertucci's type 2, (iii) two individual canals, (iv) Vertucci's type 6, canals (v) with and (vi) without debris, and (vii) canal with visible pulp calcification. All teeth were instrumented with the BioRaCe system and scanned in a Skyscan 1173 micro-CT before and after canal preparation. After reconstruction, the instrumented stack of images (IS) was registered against the preoperative sound stack of images (SS). Image processing included contrast equalization and noise filtering. Sound canal volumes were obtained by a minimum threshold. For the IS, a fixed conservative threshold was chosen as the best compromise between instrumented canal and dentine whilst avoiding debris, resulting in instrumented canal plus empty spaces. Arithmetic and logical operations between sound and instrumented stacks were used to identify debris. Noninstrumented dentine was calculated using a minimum threshold in the IS and subtracting from the SS and total debris. Removed dentine volume was obtained by subtracting SS from IS. RESULTS: Quantitative data on total debris present in the root canal space after instrumentation, noninstrumented areas and removed dentine volume were obtained for each test case, as well as three-dimensional volume renderings. CONCLUSION: After standardization of acquisition, reconstruction and image processing micro-CT images, a quantitative approach for calculation of root canal biomechanical outcomes was achieved using free software.