Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/526778
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.coverage.spatial | ||
dc.date.accessioned | 2023-11-22T04:48:20Z | - |
dc.date.available | 2023-11-22T04:48:20Z | - |
dc.identifier.uri | http://hdl.handle.net/10603/526778 | - |
dc.description.abstract | Hierarchical fuzzy expert system, an ease to use interface has been designed by using symptoms of diseases which consists of twenty one input parameters and fuzzy if-else rules are constructed by using medical expert advice in order to build very precise fuzzy inference system. Fuzzy rules and membership functions are constructed and followed by fine tuning to obtain a very efficient performance. This system can help the medical specialists for the early diagnosis of diabetic retinopathy. This system can also be used as a classifier to differentiate the types of diabetic retinopathy. The hierarchical system contains medical parameters based on the fuzzy rules designed in the system, the system give an accurate result. After comparing the fuzzy rules with health care experts, the hierarchical fuzzy system was found to be 0.99% accuracy, 0.98% sensitivity and 100% specificity respectively. newlineSecondly the fundus image positive and negative images for Non proliferative diabetic retinopathy and proliferative diabetic retinopathy recorded under standard image acquisition protocol are considered for this work. To increase the quality of the image, pre-processing is done using a image normalization, median and image embedded (Adaptive Histogram Equalization). Active Counter Model was used to perform segmentation and then feature extraction was done by using first order statistical features of the images. At the last step, classification had been done with the help of the mean values. Based on the calculations the newlinesensitivity of the system was calculated as 0.98%, with 0.97% of specificity and 0.97% of accuracy. newlineFurther, three different classification techniques namely Decision Tree (Adaptive Histogram Equalization), Naïve Bayes and Support Vector Machine (SVM) has been applied on training and test dataset to detect diabetic retinopathy. newlineThe designed system has been validated by conducting various trials in hospitals of Jalandhar region. The results have been verified by the medical specialists. | |
dc.format.extent | i-xvii, 131 | |
dc.language | English | |
dc.relation | ||
dc.rights | university | |
dc.title | Detection of Diabetic Retinopathy by using Artificial Intelligence Technique | |
dc.title.alternative | ||
dc.creator.researcher | Kumar, Krishan | |
dc.subject.keyword | Engineering | |
dc.subject.keyword | Engineering and Technology | |
dc.subject.keyword | Engineering Multidisciplinary | |
dc.description.note | ||
dc.contributor.guide | Sharma, Vikrant | |
dc.publisher.place | Hoshiarpur | |
dc.publisher.university | GNA University | |
dc.publisher.institution | Department of Electronics and Communication Engineering | |
dc.date.registered | 2018 | |
dc.date.completed | 2023 | |
dc.date.awarded | 2023 | |
dc.format.dimensions | ||
dc.format.accompanyingmaterial | DVD | |
dc.source.university | University | |
dc.type.degree | Ph.D. | |
Appears in Departments: | Department of Electronics and Communication Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf | Attached File | 185.8 kB | Adobe PDF | View/Open |
02_prelim pages.pdf | 863.57 kB | Adobe PDF | View/Open | |
03_content.pdf | 422.11 kB | Adobe PDF | View/Open | |
04_abstract.pdf | 250.33 kB | Adobe PDF | View/Open | |
05_chapter 1.pdf | 11.52 MB | Adobe PDF | View/Open | |
06_chapter 2.pdf | 11.53 MB | Adobe PDF | View/Open | |
07_chapter 3.pdf | 11.52 MB | Adobe PDF | View/Open | |
08_chapter 4.pdf | 11.54 MB | Adobe PDF | View/Open | |
09_chapter 5.pdf | 11.52 MB | Adobe PDF | View/Open | |
10_annexures.pdf | 7.68 MB | Adobe PDF | View/Open | |
80_recommendation.pdf | 122.9 kB | Adobe PDF | View/Open |
Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
Altmetric Badge: