A feature selection based framework for histology image classification using global and local heterogeneity quantification.

Biopsy remains the gold standard for the diagnosis of chronic liver diseases. However, the concordance between readers is subject to variability causing an increasing need of objective tissue description methods. A complete framework has been implemented to analyze histological images from any kind of tissue. Based on the feature selection approach, it computes the most relevant subset of descriptors in terms of classification from a wide initial list of local and global descriptors. In comparison with equivalent methods, this implementation is able to find lists of descriptors which are significantly shorter for an equivalent accuracy and furthermore it enables the classification of slides using combinations of global and local measurements. The results have pointed that it could reach an accuracy of 82.8% in a human liver fibrosis grading approach by selecting 6 descriptors from an initial set of 258 global and local descriptors.

Optimization of a quantitative method for muscle histology assessment.

Optical microscopy offers the simplest way to obtain magnified images of biological tissues. The assessment of the muscle destructuration level can be performed by a method called Meat Destruction Indicator (MDI), which combines optical microscopy and image analysis. MDI can be used for evaluation of food quality and for considering mechanically separated meat (meat raw material with an MDI value above 58.1% contained muscle fibres sufficiently destructured). This paper is particularly focused on the metrological optimization of a quantitative image analysis method around the example of MDI measurement by microscopy, especially on the digital acquisition calibration focusing and analysis work-flow. Ten different samples (45 sections) were examined with variable settings of microscope and camera to define the optimal configuration. The tests were performed with different observers to define rules and criteria for results validation. Based on the obtained results, we suggest choosing objective rules to set the light and colour of the camera and the microscope focus. To control the results of the automatic segmentation emerged also as a key step, and objective rules for observers to select or discard wrong segmented images should be defined. The adjusted MDI measurement by microscope can be used as a reliable method with good repeatability, thanks to this metrological assessment, which could and should be applied to all image analysis applications whatever the application.