Project

Slide Imaging is a research project supported by Das srl and developed by Università Campus Bio-Medico (Medical Informatics & Computer Science Laboratory) and Das srl. It aims at providing a Computer-Aided Diagnosis (CAD) system for autoimmune diseases through the analysis of Indirect ImmunoFluorescence (IIF) images. Indeed, IIF is considered a powerful, sensitive, and comprehensive test for antinuclear autoantibodies (ANA) analysis. Furthermore, it is one of the most effective and widely-used diagnostic screening assay able to detect in a timely manner some pathologies whose incidence has been constantly growing in the last few years.

IIF slides are examined at the fluorescence microscope, and their diagnosis requires both the estimation of fluorescence intensity and the description of staining pattern. The former is scored semi-quantitatively with respect to both positive and negative controls contained in each slide. The latter suggests the localization of reactive nuclear antigens and may help clinicians in differential diagnosis.

However, IIF method has some disadvantages. The major ones are: the low level of standardization, the interobserver variability which limits the reproducibility of IIF readings; the lack of resources and adequately trained personnel; the photobleaching effect, which bleaches significantly the tissues in a few seconds. Such drawbacks affect the diagnosis repeatability, therefore limiting the procedure reliability. In literature the intra-laboratory variability has been estimated equal to 7-10%. Indeed, humans are limited in their ability to detect and diagnose disease during image interpretation due to their non-systematic search patterns and to the presence of noise. In addition, the vast amount of image data that is generated makes the detection of potential disease a burdensome task and may cause oversight errors. Another problem is that similar characteristics of some abnormal and normal structure may cause interpretational errors.

Automation may offer a solution to the growing demand of diagnostic tests for systemic autoimmune diseases, as in other areas of medicine.

Being able to automatically determine the presence of autoantibodies in IIF would enable easier, faster and more reliable tests.

Hence, an evident medical demand is the development of a Computer-Aided Diagnosis (CAD) system, which may support the physician's decision and overcome current method limitations. Indeed CAD methods, which have definitely been proven effective in other contexts, (i)allow to perform a pre-selection of the cases to be examined, enabling the physician to focus his/her attention only on relevant cases, making it easier to carry out mass screening campaigns, (ii) serve as a second reader, thus augmenting the physician’s capabilities and reducing errors, (iii) aid the physician while he/she carries out the diagnosis, (iv) work as a tool for training and education of specialized medical personnel.

Besides providing image acquisition and traditional image post-processing tools, the main functionality of a CAD regards the automatic classification of the images.

The analysis of the literature in the field of ANAs detection reveals that a comprehensive CAD system in IIF is not available yet.

Therefore, this research project aims at developing a system that addresses the limitations reported above. It should improve both data management and standardization level, with particular reference to image acquisition and classification.

As first step of the project we have validated the use of digital images in IIF practice both in manual and assisted diagnosis. The outcomes are the following. First, three classes of fluorescence intensity (positive, negative and border zone) allow to identify the presence/absence of a disease. Second, five staining pattern classes (homogeneous, speckled, rim, nucleolar and no pattern) help the clinicians to choose the most appropriate second level autoantibody serum tests.

Actually, the computer tools developed so far permit to recognize three classes of fluorescence intensity, whereas in a near future they will be able to classify all five staining patterns. It will also include an error-reject option that makes the classification system more flexible, since it can vary its operating point. Therefore, it can be suited to different operating scenarios.

In this respect, SlideImagingOnWeb is the web application of this project. It has been developed for testing and research purposes among the international community. It offers support for IIF fluorescence intensity and staining pattern classification of HEp-2 images prepared at the 1:80 titer.

We hope that the use of these tools will improve the standardization level of IIF procedure, will enable an easier sharing of the information in the research environment, and will be employed as an effective learning tool for training the medical personnel.