Implementation of a classification strategy of Raman data collected in different clinical conditions: application to the diagnosis of chronic lymphocytic leukemia.

The literature is rich in proof of concept studies demonstrating the potential of Raman spectroscopy for disease diagnosis. However, few studies are conducted in a clinical context to demonstrate its applicability in current clinical practice and workflow. Indeed, this translational research remains far from the patient's bedside for several reasons. First, samples are often cultured cell lines. Second, they are prepared on non-standard substrates for clinical routine. Third, a unique supervised classification model is usually constructed using inadequate cross-validation strategy. Finally, the implemented models maximize classification accuracy without taking into account the clinician's needs. In this paper, we address these issues through a diagnosis problem in real clinical conditions, i.e., the diagnosis of chronic lymphocytic leukemia from fresh unstained blood smears spread on glass slides. From Raman data acquired in different experimental conditions, a repeated double cross-validation strategy was combined with different cross-validation approaches, a consensus label strategy and adaptive thresholds able to adapt to the clinician's needs. Combined with validation at the patient level, classification results were improved compared to traditional strategies.

Ontology-based lymphocyte population description using mathematical morphology on colour blood images.

Despite modern technologies (immunophenotyping, molecular probing, etc.) cytomorphologic examination of stained peripheral blood smears by microscopy remains the main way of diagnosis in a large variety of diseases (e.g. leukaemic disorders). Using tools from mathematical morphology for processing peripheral blood colour images, we have developed an image-based approach, to provide an objective and understandable description of lymphocyte populations according to a specifically designed ontology. This ontology-based framework needs a conceptualisation of the problem from a morphological viewpoint, the introduction of an adapted language, the generation of representative image databases, the development of image processing and data classification algorithms to automate the procedure and the validation of the system by human expertise. In this paper we present the main concepts, algorithms and some results to illustrate the high-performance of the approach. The aim of our work is to reconcile the automatisation with the medical expertise, so that they can reinforce each other.

A European de facto standard for image folders applied to telepathology and teaching.

Since 1980, French pathologists at ADICAP (Association pour le Développement de l'Informatique en Cytologie et en Anatomie Pathologique) have created a common language code allowing the use of computers for routine applications. This code permitted the production of an associated exhaustive image bank of approximately 30,000 images. This task involved many specialists necessitating the definition of specific processes for security and simplicity of data handling. In particular, it has been necessary to develop image communication. To achieve that goal, it was necessary to define a folder, associating textual information to images. That was done through several industrial software providers contribution. Consequently, this folder, using a common packaging standard, allowed any pathologist access to images, codified data and clinical information. Accessing folders has been made easy by launching a Web server at CRIHAN under the supervision of ADICAP. An ADICAP software user may not only browse through the folder but may also import them into their own system and produce new folders. Today more than a hundred users in France and in foreign countries are able to provide diagnostic advice and also referential products useful for further education and quality control. The next challenge is the development of this preliminary de facto approach toward an internationally admitted standard suited for morphological image exchange.