Mitochondria are sub-cellular components which are mainly responsible for synthesis of adenosine tri-phosphate (ATP) and involved in the regulation of several cellular activities such as apoptosis. The relation between some common diseases of aging and morphological structure of mitochondria is gaining strength by an increasing number of studies. Electron microscope tomography (EMT) provides high-resolution images of the 3D structure and internal arrangement of mitochondria. Studies that aim to reveal the correlation between mitochondrial structure and its function require the aid of special software tools for manual segmentation of mitochondria from EMT images. Automated detection and segmentation of mitochondria is a challenging problem due to the variety of mitochondrial structures, the presence of noise, artifacts and other sub-cellular structures. Segmentation methods reported in the literature require human interaction to initialize the algorithms. In our previous study, we focused on 2D detection and segmentation of mitochondria using an ellipse detection method. In this study, we propose a new approach for automatic detection of mitochondria from EMT images. First, a preprocessing step was applied in order to reduce the effect of nonmitochondrial sub-cellular structures. Then, a curve fitting approach was presented using a Hessian-based ridge detector to extract membrane-like structures and a curve-growing scheme. Finally, an automatic algorithm was employed to detect mitochondria which are represented by a subset of the detected curves. The results show that the proposed method is more robust in detection of mitochondria in consecutive EMT slices as compared with our previous automatic method.
Cortical renal (kidney) scintigraphy images are 2D images (256x256) acquired in three projection angles (posterior,
right-posterior-oblique and left-posterior-oblique). These images are used by nuclear medicine specialists to examine
the functional morphology of kidney parenchyma. The main visual features examined in reading the images are: size,
location, shape and activity distribution (pixel intensity distribution within the boundary of each kidney). Among the
above features, activity distribution (in finding scars if any) was found to have the least interobserver reproducibility.
Therefore, in this study, we developed an image-based retrieval (IBR) and a computer-based diagnosis (CAD) system,
focused on this feature in particular. The developed IBR and CAD algorithms start with automatic segmentation,
boundary and landmark detection. Then, shape and activity distribution features are computed. Activity distribution
feature is obtained using the acquired image and image set statistics of the normal patients. Active Shape Model (ASM)
technique is used for more accurate kidney segmentation. In the training step of ASM, normal patient images are used.
Retrieval performance is evaluated by calculating precision and recall. CAD performance is evaluated by specificity and
sensitivity. To our knowledge, this paper is the first IBR or CAD system reported in the literature on renal cortical