Image Processing for Cytology
Development and Creation of Computer Analyser of Images for Cytologic Differential Diagnosis of Thyroid Diseases
Tech Area / Field
- MED-DID/Diagnostics & Devices/Medicine
- INF-OTH/Other/Information and Communications
8 Project completed
Senior Project Manager
Gremyakova T A
Belorussian State Medical University, Belarus, Minsk
- National Academy of Sciences of the Republic of Belarus / Institute of Informatics Problems, Belarus, Minsk
- University of Surrey, UK, Guildford\nTechnical University of Aachen / Institute of Medical Informatics and Biometry, Germany, Aachen
Project summaryThe aim of the proposed project is to create the computer analyser of images, methodical support and software for differential cytologic diagnosis of thyroid diseases.
This work is necessitated by the fact that in differential cytologic diagnosis of thyroid tumors, the role of generally accepted criteria of malignancy is limited. According to the WHO cytologic classification, during the assessment of a thyroid neoplasm, a cytologist can “only state the presence of a tumor of follicular structure, but the question about the malignancy can only be settled after the histological examination of operative material” (WHO. Cytologic classification of tumors, Geneva, 1982). In the case of puncturing, insufficient amount or absence of specialized cells in punctate smears is being added to this limitation. Consequently, it leads to low informativity of punctate examinations with thyroid cancer (less than 50%). In Belarus, this problem acquired a special significance after the Chernobyl accident which lead to a sharp increase in thyroid carcinomas. Low rate of cancer identification at early stages demands the development of new effective methods of oncological diagnosis. In this respect, most perspective is considered to be the transformation of qualitative indices of pathological changes in thyroid tissue cells into the quantitative form with the help of morphometry. Previously, during the morphometric study of postoperative material, it was shown by us that the character of distribution of parameters reflecting the pathological changes in thyrocytes depended on the nature (malignant-benign) of a thyroid disease (V.Kirillov et al., 1994; V.Kirillov et al., 1996 a) which makes it possible to diagnose thyroid carcinoma by only one parameter. Later it has been ascertained that the share of small lymphocytes in the general population of thyroid tissue lymphoid cells is considerably smaller with cancer than in norm and with benign pathology (V.Kirillov et al., 1996 b). Such morphological assessment of nonspecialized cells allows to increase considerably the accuracy and informativity of oncological diagnosis. We have taken out 10 patents for the modes of thyroid carcinoma diagnosis at postoperative stages. In the given project , it is envisaged to develop with the help of morphometry quite a new approach to cytologic diagnosis of thyroid carcinoma and its regional metastases at preoperative stages.
To solve the problem of differential diagnosis of the main forms of thyroid diseases (papillary and follicular cancer, adenoma, autoimmune thyroiditis, nodular colloid and diffuse toxic goiter), it is proposed to make the cytologic differential diagnosis of thyroid diseases by means of comparison of two images at the computer monitor – that of the studied sample with standard images from the electronic atlas where they are systematized by qualitative parameters reflecting the pathological changes in thyroid tissue cells with different forms of thyroid diseases. The automatic analysis of images is envisaged with the elements of identification and subsequent verification of diagnosis.
Specialized analyser of images created on the basis of personal computer, light microscope, colour telecamera and frame-grabber, apart from high technical characteristics, will have a powerful software in order to solve the raised problems. System’s software working in Windows environment, will include software packages aimed at: automated processing and binarization of colour images; automatic raster-to-vector transformation and formation of objects in terms of cytologic diagnosis; automated analysis of images with the elements of biological object identification; automation of operations for morphometric assessment of biological objects by quantitative parameters; statistical processing of the results; realization of modes of differential diagnosis of thyroid diseases; electronic atlas.
The project will be finalized with the creation of computer analyser aimed at solving the tasks of cytologic diagnosis, a new approach to solving the problem of oncological and differential diagnosis of thyroid diseases, electronic atlas of thyroid pathology and software package for cell morphometry, image processing and verification of diagnosis. Actually, this system will be universal which will allow to use it, with the availability of corresponding methodical elaboration, in cytologic diagnosis of diseases of different organs and tissues. There are no such analogues at present. Therefore, this system, both as a whole and its separate parts, will be much in demand at the world market. The use of project results in medical practice will allow to raise the accuracy and efficiency of verification of diagnosis at preoperative stage which is most important when the mode of operative treatment is being chosen.
Foreign collaborators will be involved in the development of working plan, exchange of information, consultations, conduction of joint workshops and local technical inspection of project implementation in accordance with ISTC procedures.
The International Science and Technology Center (ISTC) is an intergovernmental organization connecting scientists from Kazakhstan, Armenia, Tajikistan, Kyrgyzstan, and Georgia with their peers and research organizations in the EU, Japan, Republic of Korea, Norway and the United States.
ISTC facilitates international science projects and assists the global scientific and business community to source and engage with CIS and Georgian institutes that develop or possess an excellence of scientific know-how.