کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
870768 | 910020 | 2015 | 9 صفحه PDF | دانلود رایگان |
![عکس صفحه اول مقاله: Quantification of DNA damage by the analysis of silver stained comet assay images Quantification of DNA damage by the analysis of silver stained comet assay images](/preview/png/870768.png)
1. ObjectivesThe main objective is to develop a fully automated system for Deoxyribo Nucleic Acid (DNA) damage analysis using silver stained comet assay images for clinical applications.2. Materials and methodsSilver stained comet assay images of lymphocytes obtained from confirmed cases of Schizophrenia were employed for DNA damage analysis using Single Cell Gel Electrophoresis (SCGE) or Comet assay. The image analysis consists of three stages: (1) Comet segmentation by incorporating shading correction, pre-processing stages for image enhancement, noise filtering and thresholding; (2) Comet partitioning by fuzzy clustering and (3) Comet quantification stage in which the limitations of fuzzy clustering are eliminated.3. ResultsThis study involves 40 cases of Schizophrenia. Results of the proposed method are compared with those of OpenComet and CometScore software. Comet segmentation result is validated with a confusion matrix which shows better values for all performance indices. The proposed technique gave 87.12% Positive Predictive Value (PPV) and 90.16% sensitivity. It shows a large improvement in PPV and sensitivity, over the most recent competitive method, OpenComet. Regression analysis is carried out to validate the quantification results. Visual scores evaluated by the expert and DNA (%) in tail are used for finding the coefficient of determination, r2r2 statistic. The proposed method gives a better r2r2 value of 0.9. The performance of the proposed method is evaluated by a clinical expert.4. ConclusionThe proposed method is found to be very efficient for DNA damage analysis of silver stained comet assay images in the area of clinical research.
Journal: IRBM - Volume 36, Issue 5, October 2015, Pages 306–314