Hand bone radiograph image segmentation with ROI merging
ACC'11/MMACTEE'11 Proceedings of the 13th IASME/WSEAS international conference on Mathematical Methods and Computational Techniques in Electrical Engineering conference on Applied Computing
Hi-index | 0.00 |
This paper details a epiphyseal/metaphyseal region of interest (EMROI) extraction method used for automated TW2 RUS. Several methods for automated bone age assessment using image processing have been developed but they have some limitations in assisting physicians efficiently. Also, a method is required that can extract the feature region robustly to gray value variation of background and the position and orientation of the hand. In order to overcome the limitation of previous methods and achieve robust feature region extraction, we propose an EMROI extraction method which contains the procedures from background removal to EMROI extraction. Each procedure is designed to resolve problems which may occur by variation of the position and orientation of the hand and bones. Through experiments, the proposed method resulted in over 91% correct extraction for TW2 RUS from 296 left hand radiograph data.