Semiautomatic detection of villi in confocal endoscopy for the evaluation of celiac disease.

Celiac Disease (CD) is an immune-mediated enteropathy, diagnosed in the clinical practice by intestinal biopsy and the concomitant presence of a positive celiac serology. Confocal Laser Endomicroscopy (CLE) allows skilled and trained experts to potentially perform in vivo virtual histology of small-bowel mucosa. In particular, it allows the qualitative evaluation of mucosa alteration such as a decrease in goblet cells density, presence of villous atrophy or crypt hypertrophy. We present a semi-automatic method for villi detection from confocal endoscopy images, whose appearance change in case of villous atrophy. Starting from a set of manual seeds, a first rough segmentation of the villi is obtained by means of mathematical morphology operations. A merge and split procedure is then performed, to ensure that each seed originates a different region in the final segmentation. A border refinement process is finally performed, evolving the shape of each region according to local gradient intensities. Mean and median Dice coefficients for 290 villi originating from 66 images when compared to manually obtained ground truth are 80.71% and 87.96% respectively.

Detection and density estimation of goblet cells in confocal endoscopy for the evaluation of celiac disease.

Celiac Disease (CD) is an immune-mediated enteropathy, diagnosed in the clinical practice by intestinal biopsy and the concomitant presence of a positive celiac serology. Confocal Laser Endomicroscopy (CLE) allows skilled and trained experts to potentially perform in vivo virtual histology of small-bowel mucosa. In particular, it allows the qualitative evaluation of mucosa alteration such as a decrease in goblet cells density, presence of villous atrophy or crypt hypertrophy. We present a semi-automatic computer-based method for the detection of goblet cells from confocal endoscopy images, whose density changes in case of pathological tissue. After a manual selection of a suitable region of interest, the candidate columnar and goblet cells' centers are first detected and the cellular architecture is estimated from their position using a Voronoi diagram. The region within each Voronoi cell is then analyzed and classified as goblet cell or other. The results suggest that our method is able to detect and label goblet cells immersed in a columnar epithelium in a fast, reliable and automatic way. Accepting 0.44 false positives per image, we obtain a sensitivity value of 90.3%. Furthermore, estimated and real goblet cell densities are comparable (error: 9.7 ± 16.9%, correlation: 87.2%, R(2) = 76%).

Contrast-enhanced ultrasound findings in soft-tissue lesions: preliminary results.

OBJECTIVES: There is currently no widely available, minimally invasive first-level examination that allows physicians to identify soft-tissue lesions that are likely to be malignant. The aim of this pilot study was to explore the potential suitability of dynamic contrast-enhanced ultrasound (DCE-US) for this purpose. MATERIALS AND METHODS: 23 patients were referred to the Veneto Oncological Institute for work-up of superficial soft-tissue lesions. Fourteen lesions were examined with CEUS and enhancement kinetics was analyzed. Subsequently, all lesions were surgically removed and subjected to histological analysis. RESULTS: The 14 lesions included in the study were histologically classified as malignant (n = 7) or benign (n = 7, including 3 schwannomas). A statistically significant difference between benign and malignant lesions was found in terms of mean times to peak enhancement intensity (p = 0.03) but not mean filling times (FT). When schwannomas were analyzed as a separate group, their mean FT was found to be significantly different from that of the other benign lesions (p = 0.001) and from that of the group comprising other benign lesions as well as malignant lesions (p < 0.005). CONCLUSIONS: CEUS with analysis of contrast-enhancement kinetics is a relatively low-cost, minimally invasive imaging technique, which appears to be a potentially effective first-level method for identifying suspicious soft-tissue masses.