Result window" shows the cartilage information with gray (or color) intensity over the 3D bone shape. The developed program can display and save color DICOM MRI as shown in Fig. 2, and provide the statistical information (Mean, Min, Max of thickness and Volume) of the local VOI of "Lateral Condyle" and "Medial Condyle" as shown in the figures. Conclusions: This paper proposes fully automatic cartilage segmentation and measurement method without any training process. The proposed method can visualize the whole cartilage distribution with the 3D bone shape. The experimental results demonstrate that it can be used for inspecting cartilage damage or loss directly. The developed software is therefore applicable in clinical knee OA diagnosis systems. Future studies include more tests with new MRI data for confirming the accuracy of the measurements.