In Vivo Assessment of Tumor Vascularity Using Confocal Laser Endomicroscopy in Murine Models of Colon Cancer

Machine Learning-Based Classification of the Health State of Mice Colon in Cancer Study from Confocal Laser Endomicroscopy

In this article, we address the problem of the classification of the health state of the colon's wall of mice, possibly injured by cancer with machine learning approaches. This problem is essential for translational research on cancer and is a priori challenging since the amount of data is usually limited in all preclinical studies for practical and ethical reasons. Three states considered including cancer, health, and inflammatory on tissues. Fully automated machine learning-based methods are proposed, including deep learning, transfer learning, and shallow learning with SVM. These methods addressed different training strategies corresponding to clinical questions such as the automatic clinical state prediction on unseen data using a pre-trained model, or in an alternative setting, real-time estimation of the clinical state of individual tissue samples during the examination. Experimental results show the best performance of 99.93% correct recognition rate obtained for the second strategy as well as the performance of 98.49% which were achieved for the more difficult first case.

In vivo assessment of tumour angiogenesis in colorectal cancer: the role of confocal laser endomicroscopy

AIM

Tumour neoangiogenesis is a key factor in tumour progression and metastatic spread and the possibility to assess tumour angiogenesis might provide prognostic information. The aim of this study was to establish the role of probe-based confocal laser endomicroscopy (p-CLE) in the identification of vascular architecture and specific morphological patterns in normal colorectal mucosa and malignant lesions during routine endoscopy.

METHOD

Fourteen consecutive patients with colorectal cancer were included. The following features were identified and then compared between normal and neoplastic mucosa on p-CLE images: vessel shape (straight vs irregular) vessel diameter the 'branching patterns' vessel permeability (fluorescein leakage) and blood flow (normal vs defective flux). Immunohistochemistry was used to confirm the presence and to study the morphology of vascular structures (CD-34 staining) and 'neo-vessels' (WT-1 staining) on tumour and normal mucosal sections.

RESULTS

Tumour vessels appeared as irregular, ectatic and with a highly variable calibre and branching patterns on p-CLE images. The mean diameter of tumour vessels was significantly larger than those in normal mucosa (weighted mean difference 3.38, 95% CI 2.65-4.11, P = 0.01). Similarly, 'vessel branching' (OR 2.74, 95% CI 1.23-6.14, P = 0.01), fluorescent dye 'extravasation' (OR 3.46, 95% CI 1.39-8.57, P = 0.01) were significantly more frequent in colorectal cancer than in normal colorectal mucosa. Immunohistochemistry corroborated the p-CLE findings, showing higher vascularity in tumour sections due to neoformed vessels, presenting irregular patterns.

CONCLUSION

Probe-based confocal laser endomicroscopy provides a noninvasive characterization of the microvascular architecture of colonic mucosa. Different morphological patterns have been described, discriminating normal and malignant microvascular networks in colorectal mucosa.