Team Asclepios

Overall Objectives
Scientific Foundations
New Results
Contracts and Grants with Industry
Other Grants and Activities

Section: New Results

Biological Image Analysis

Imaging Study of the Ovarian Function from Histological Images

Keywords : Histology, Registration, 3D Reconstruction, Segmentation.

Participants : Johan Debayle, Grégoire Malandain.

This work takes place in the Cooperative Research Initiative (CRI) named REGulation of Ovulation (REGLO).

This CRI aims to study the follicular development in mammals, and proposes mathematical models that allow to follow the granulosa cell population, and then to predict the outcome of the follicular development (ovulation or degeneration) with respect to the hormonal environment.

To study the ovarian function, we have to extract, from ovary images (in ewe), quantitative parameters that will allow to better understand and assess those mathematical models. These parameters will be either morphological: temporal evolution of the follicle volume, vascular density around follicles; or functional: vascular perfusion, following the imaging modalities. At this time, we have histological images (after staining procedures) which give access to microscopic reference images.

So, we have to reconstruct 3-D volumes from those histological sections and then to extract quantitative parameters from those images.

Figure 9. After registration and 3D reconstruction from the histological sections [a], the granulosa cells and the vascularization [c] (superimposed in blue and red respectively) of each selected follicle [b] are extracted.
(a) histological section(b) zoom on a selected follicle(c) extraction of the parameters

After the 3-D reconstruction, we have to segment the granulosa cells and the vascularization (Fig. 9 ) of each follicle within the ovary to be studied. Those regions are segmented by thresholding and/or model deformation. Quantitative parameters are then computed and discussed with the biologists involved in the REGLO project, before validation of the results.

Mosaicing of Confocal Microscopic In Vivo Soft Tissue Video Sequences

Keywords : Mosaicing, In Vivo fibered confocal microscopy, multi-image registration.

Participants : Tom Vercauteren, Xavier Pennec, Aymeric Perchant [ MKT ] , Grégoire Malandain, Nicholas Ayache.

This work is done in collaboration with Mauna Kea Technologies, Paris, France,

Fibered confocal microscopy (FCM) is a potential tool for in vivo and in situ optical biopsy. FCM is based on the principle of confocal microscopy which is the ability to reject light from out-of-focus planes and provide a clear in-focus image of a thin section within the sample. This optical sectioning property is what makes the confocal microscope ideal for imaging thick biological samples.

Figure 10. Left: The Deckel-Maho milling machine holding the flexible microprobe. This micrometric precision machine was used to evaluate the accuracy of our mosaicing algorithm. Right: Mosaic of a normal bronchial area, right upper lobe carina. Courtesy of L. Thiberville, Rouen University Hospital, France.

In 2005, we proposed an algorithm using image sequence mosaicing techniques to widen the field of view (FOV) and enhance the possibilities offered by FCM. In [57] , [61] , we further developed this algorithm to compensate for the motion distortions arising from the laser scanning.

In the field of biomedical imaging, the issue of validation for image processing tasks is essential. In order to get some ground truth data to evaluate the accuracy of our mosaicing algorithm, we used the micrometric precision computer numerical control milling machine shown in Fig. 10 . In [57] , we showed that our algorithm was able to automatically recover the motion imposed by the milling machine. This paper was awarded the Best MICCAI Paper published in Medical Image Analysis journal in 2006.

Because of its small size, the fiberoptic probe can be introduced into the working channel of a flexible bronchoscope. This makes it possible to use FCM to study the bronchial autofluorescence at the microscopic level in vivo. In [56] , our mosaicing was successfully used in this clinical setting as shown in Fig. 10 .


Logo Inria