Quantitative image analysis of synovial tissue.

Quantitative image analysis is a form of imaging that includes microscopic histological quantification, video microscopy, image analysis, and image processing. Hallmarks are the generation of reliable, reproducible, and efficient measurements via strict calibration and step-by-step control of the acquisition, storage and evaluation of images with dedicated hardware and software. Major advantages of quantitative image analysis over traditional techniques include sophisticated calibration systems, interaction, speed, and control of inter- and intraobserver variation. This results in a well controlled environment, which is essential for quality control and reproducibility, and helps to optimize sensitivity and specificity. To achieve this, an optimal quantitative image analysis system combines solid software engineering with easy interactivity with the operator. Moreover, the system also needs to be as transparent as possible in generating the data because a "black box design" will deliver uncontrollable results. In addition to these more general aspects, specifically for the analysis of synovial tissue the necessity of interactivity is highlighted by the added value of identification and quantification of information as present in areas such as the intimal lining layer, blood vessels, and lymphocyte aggregates. Speed is another important aspect of digital cytometry. Currently, rapidly increasing numbers of samples, together with accumulation of a variety of markers and detection techniques has made the use of traditional analysis techniques such as manual quantification and semi-quantitative analysis unpractical. It can be anticipated that the development of even more powerful computer systems with sophisticated software will further facilitate reliable analysis at high speed.

Digital image analysis for the evaluation of the inflammatory infiltrate in psoriasis.

Traditionally the evaluation of the cellular infiltrate and protein expression in skin tissue sections is done by manual quantification. However, for reliable evaluation of histology in the development of new anti-psoriatic treatments there is a need for a more time-efficient and reproducible method. To test the use of digital image analysis (DIA) in this situation we compared the assessment of immunohistochemically stained skin sections with the more traditional manual quantification (MQ) and semi-quantitative analysis (SQA). The number of CD3+ T cells and the expression of E-selectin were evaluated in stained paired skin biopsies from 11 patients with chronic plaque psoriasis before and after initiation of anti-psoriasis therapy. We observed significant correlations between MQ and DIA for the number of T cells (epidermis: r=0.88, P< or =0.01, dermis r=0.87, P< or =0.01). Both DIA and MQ were equally effective in detecting reductions of T-cell numbers in active-treated patients. MQ took 20 h, compared to 6 h for DIA. We also observed significant correlations between SQA and DIA for the expression of E-selectin (r=0.88, P< or =0.01), although DIA was more sensitive than SQA to detect (early) changes. SQA took 10 h, compared to 4 h for DIA. In conclusion, the quantification of the inflammatory infiltrate in psoriatic lesional skin by DIA generated similar results as MQ and SQA in a reliable, reproducible and higher time efficient fashion.