JavaScript is disabled for your browser. Some features of this site may not work without it.
| dc.contributor.author | YAWUKAL, ASHAGRIE ASAYE | |
| dc.date.accessioned | 2023-07-03T06:41:01Z | |
| dc.date.available | 2023-07-03T06:41:01Z | |
| dc.date.issued | 2023-03 | |
| dc.identifier.uri | http://ir.bdu.edu.et/handle/123456789/15431 | |
| dc.description.abstract | Nephrolithiasis is a prevalent cause of chronic renal diseases which is extremely costly to treat. The diagnosis of nephrolithiasis is difficult since there aren't enough radiologist-interpreters to interpret pictures from imaging devices and make a decision. Machine Learning (ML) algorithms are currently used for the detection or diagnosis of kidney stones, with the major drawbacks of limited data, ionizing radiation from scanning devices, ex-vivo techniques, and cost. The ultimate objective of this thesis is to analyze the performance of ML algorithms on local datasets and estimate the size of kidney stones for the purposes of characterization. In this thesis, ultrasound images were collected from different hospitals and annotated by radiographers or experts. Preprocessing mainly focused on filtering using Gabor filters to remove speckle noise. Image segmentation techniques were used for further analysis, using thresholding image segmentation mechanisms for feature extraction and stone size estimation. Entropy and Grey Level Co-occurrence Matrix (GLCM) feature descriptors were extracted from the segmented image. Support Vector Classifiers (SVC), Decision Trees (DT), K-Nearest Neighbors (KNN), and Random Forest (RF) algorithms were developed, using the train-test split cross validation technique. KNN and RF models were outperformed in the provided datasets with performance metrics of accuracy, precision, recall, and AUC; 98.4%, 0.97, 1.0, and 0.98, respectively, for KNN and 95.1%, 0.94, 0.97, and 0.9896, respectively, for RF. And estimation of stone size with the major axis length of 10.2235 mm was obtained for the actual stone size of 11.9 mm, as annotated by the expert. Hence, these developed ML algorithms can enhance and improve the diagnosis and detection of kidney stones (renal calculi) from ultrasound images, which are noninvasive, available, and affordable without any ionizing radiation to improve the quality of life of the patients. Keywords: Detection, Entropy, Feature extraction, Gabor filter, GLCM features, KNN, ML algorithms, Nephrolithiasis, Performance metrics, RF, Stone size estimation, Thresholding based segmentation. | en_US |
| dc.language.iso | en_US | en_US |
| dc.subject | Electrical and Computer Engineering | en_US |
| dc.title | PERFORMANCE ANALYSIS ON DETECTION OF NEPHROLITHIASIS USING MACHINE LEARNING ALGORITHMS | en_US |
| dc.type | Thesis | en_US |
