Roger Lédée

A New Adaptive Switching Median Filter

A new Adaptive Switching Median (ASWM) filter for removing impulse noise from corrupted images is presented. The originality of ASWM is that no a priori Threshold is needed as in the case of a classical Switching Median filter. Instead, Threshold is computed locally from image pixels intensity values in a sliding window. Results show that ASWM provides better performance in terms of PSNR and MAE than many other median filter variants for random-valued impulse noise. In addition it can preserve more image details in a high noise environment.

3D registration using a new implementation of the ICP algorithm based on a comprehensive lookup matrix: Application to medical imaging

The iterative closest point (ICP) algorithm is an efficient algorithm for robust rigid registration of 3D data. Results provided by the algorithm are highly dependent upon the step of finding corresponding pairs between the two sets of 3D data before registration. In this paper, a look up matrix is introduced in the point matching step to enhance the overall ICP performance. Convergence properties and robustness are evaluated in the presence of Gaussian and impulsive noise, and under different data set sizes. The new algorithm has been evaluated on 3D medical data. It has been applied successfully to register closed surfaces acquired using different medical imaging modalities.

Comparison of Image Restoration Methods for Bioluminescence Imaging

Bioluminescence imaging is a recent modality to visualize biological effects, especially for small animals. However, the acquired images are degraded by diffusion and absorption phenomena from the tissue and by the acquisition system itself. In this paper, we use restoration methods to enhance the quality of bioluminescence images. We propose a model for image formation and an experimental determination of the PSF (Point Spread Function). Several restoration methods are compared on test images generated according to the model and on real data. This comparison is insured by using MSE (Mean Square Error) and two other quantitative criteria. Results showed that the statistical methods give more accurate restoration and are well adapted for Bioluminescence Imaging.

An image dependent stopping method for iterative denoising procedures

Iterative methods are very successful for denoising images corrupted by random valued impulse noise. However, choosing the optimal number of iterations is a difficult issue. In this letter, a stopping method is proposed: the iterative denoising process is stopped when the number of cleaned pixels is minimal. It corresponds to the moment when the denoising process tends to modify noise-free pixels. It also corresponds with a high precision to the maximum of PSNR of the restored image. The originality of the method is that no a priori iteration number is to be chosen but the method results from image information. The proposed stopping strategy is therefore an efficient and image dependent method that can be easily implemented on real data.

A generating model of realistic synthetic heart sounds for performance assessment of phonocardiogram processing algorithms

Abstract A new model which is capable of generating realistic synthetic phonocardiogram (PCG) signals is introduced based on three coupled ordinary differential equations. The new PCG model takes into account the respiratory frequency, the heart rate variability and the time splitting of first and second heart sounds. This time splitting occurs with each cardiac cycle and varies with inhalation and exhalation. Clinical PCG statistics and the close temporal relationship between events in ECG and PCG are used to deduce values of PCG model parameters. In comparison with published PCG models, the proposed model allows a larger number of known PCG features to be taken into consideration. Moreover it is able to generate both normal and abnormal realistic synthetic heart sounds. Results show that these synthetic PCG signals have the closest features to those of a conventional heart sound in both time and frequency domains. Additionally, a sound quality test carried out by eight cardiologists demonstrates that the proposed model outperforms the existing models. This new PCG model is promising and useful in assessing signal processing techniques which are developed to help clinical diagnosis based on PCG.

Quantitative CT imaging for adipose tissue analysis in mouse model of obesity

In obese humans CT imaging is a validated method for follow up studies of adipose tissue distribution and quantification of visceral and subcutaneous fat. Equivalent methods in murine models of obesity are still lacking. Current small animal micro-CT involves long-term X-ray exposure precluding longitudinal studies. We have overcome this limitation by using a human medical CT which allows very fast 3D imaging (2 sec) and minimal radiation exposure. This work presents novel methods fitted to in vivo investigations of mice model of obesity, allowing (i) automated detection of adipose tissue in abdominal regions of interest, (ii) quantification of visceral and subcutaneous fat. For each mouse, 1000 slices (100μm thickness, 160 μm resolution) were acquired in 2 sec using a Toshiba medical CT (135 kV, 400mAs). A Gaussian mixture model of the Hounsfield curve of 2D slices was computed with the Expectation Maximization algorithm. Identification of each Gaussian part allowed the automatic classification of adipose tissue voxels. The abdominal region of interest (umbilical) was automatically detected as the slice showing the highest ratio of the Gaussian proportion between adipose and lean tissues. Segmentation of visceral and subcutaneous fat compartments was achieved with 2D 1/2 level set methods. Our results show that the application of human clinical CT to mice is a promising approach for the study of obesity, allowing valuable comparison between species using the same imaging materials and software analysis.

A new detector for switching median filter

In this paper, a new detector for switching median filter is presented. The originality of this approach is that no a priori threshold is to be given. Instead, it is automatically computed from image pixels and based on the weighted mean and the weighted variance in a selected sliding window. The weights are inversely proportional to the grey levels difference between each pixel of the considered window and the mean value of this window. Results show that this new algorithm provides better performances in terms of PSNR and MAE than many other variants of switching median filter. It suppresses noise and preserves details. In addition, psycho visual results are of high quality.

The comprehensive ICP (CICP) algorithm and its application to the multidmoal registration of 3D surfaces of the heart

Image registration is a valuable technique for medical diagnosis and treatment. In this paper, we present an enhanced implementation of the popular iterative closest point (ICP) algorithm developed for the registration of 3D free-form closed surfaces, based on the use of a look up matrix for finding the best correspondence pairs. The algorithm, called comprehensive ICP (CICP) algorithm, is then successfully applied for comparing two sequences of 3D surfaces of the left ventricle of the heart, obtained from two different modalities: 4D Echography and Gated SPECT. The results show a good correspondence between the reconstructed sequence from 4D echography and the reference gated SPECT sequence. This study can be extended to comparing different medical reconstructions of sphere-like shaped organs collected from different modalities.

Segmentation of mice cerebral structures: application in Trisomy 21

In this paper, a semi automatic method is proposed for the segmentation of mice cerebral structures (brain, cerebellum and hippocampus) in MR images. First, a Chan-Vese method is applied on the axial images to segment the brain volume. The method takes into account the special shape of the brain mice. Second, variational atlases are constructed by manual segmentation of various MRI brain images of reference and Trisomy 21 mice. These atlases are then registered on true data to assist the Chan-Vese segmentation of the cerebellum and the hippocampus. This semi automatic method makes that human intervention is limited and the tedious manual handling is greatly reduced. Results have shown that the brain volumes estimated by the method are identical to expert manually estimated volumes. The new method was used in the analysis of the cerebral malformations linked to Trisomy 21: no significant difference of the cerebral structures between Trisomy 21 mice and the control ones was found.

3D visualization and quantification of bone and teeth mineralization for the study of osteo/dentinogenesis in mice models

Research on bone and teeth mineralization in animal models is critical for understanding human pathologies. Genetically modified mice represent highly valuable models for the study of osteo/dentinogenesis defects and osteoporosis. Current investigations on mice dental and skeletal phenotype use destructive and time consuming methods such as histology and scanning microscopy. Micro-CT imaging is quicker and provides high resolution qualitative phenotypic description. However reliable quantification of mineralization processes in mouse bone and teeth are still lacking. We have established novel CT imaging-based software for accurate qualitative and quantitative analysis of mouse mandibular bone and molars. Data were obtained from mandibles of mice lacking the Fibromodulin gene which is involved in mineralization processes. Mandibles were imaged with a micro-CT originally devoted to industrial applications (Viscom, X8060 NDT). 3D advanced visualization was performed using the VoxBox software (UsefulProgress) with ray casting algorithms. Comparison between control and defective mice mandibles was made by applying the same transfer function for each 3D data, thus allowing to detect shape, colour and density discrepencies. The 2D images of transverse slices of mandible and teeth were similar and even more accurate than those obtained with scanning electron microscopy. Image processing of the molars allowed the 3D reconstruction of the pulp chamber, providing a unique tool for the quantitative evaluation of dentinogenesis. This new method is highly powerful for the study of oro-facial mineralizations defects in mice models, complementary and even competitive to current histological and scanning microscopy appoaches.