Volume 14, Issue 2, January 2016, Pages 297–303
Abdelouahed Ait Ider1, Malika Echajia2, Constant TCHEKA3, Abdelkrim Merbouha4, and Mohamed Mbarki5
1 Department of chemistry and environment, Sultan Moulay Slimane University, Faculty of Science and Technology, Transdisciplinary Team of Analytical Science for Sustainable Development, PB 523, Beni Mellal, Morocco
2 Department of chemistry and environment, Sultan Moulay Slimane University, Faculty of Science and Technology, Transdisciplinary Team of Analytical Science for Sustainable Development, PB 523, Béni Mellal, Morocco
3 Laboratory of Physical and Theoretical Chemistry, Department of Inorganic Chemistry, Faculty of Science, P.O. Box 812, Yaoundé, Cameroon
4 Faculty of Sciences and Technology, Sultan Moulay Slimane University, Beni Mellal, Morocco
5 Department of chemistry and environment, Sultan Moulay Slimane University, Faculty of Science and Technology, Transdisciplinary Team of Analytical Science for Sustainable Development, PB 523, Beni Mellal, Morocco
Original language: English
Copyright © 2016 ISSR Journals. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
The goal of this study is to develop a simple intelligent urolithiasis diagnosis system. The accuracy of the system was determined by comparing the recognition rates of the Artificial Neural Networks (ANNs)-, k-nearest Neighbor (kNN)-, and Support Vector Machines (SVM) algorithms. The results showed that the ANN model was superior to SVM and KNN models in prediction. We aimed through this work to classify the subjects in three classes according to the chemical concentrations of variables (Ca, Ox, pCaOx, Ca/ Ox) using and according to their clinical status. The ANN model, used to determine the first class that contains the subjects presenting their urine a calcium oxalate monohydrate (CaC2O4,H2O : whewellite (Wh)) crystal type. This ANN model reached a correct prediction rate of 85.3%. Using SVM- and KNN model the correct prediction rate reached 82.6% and 65.55% respectively. The second class contains the subjects presenting a calcium oxalate dihydrate (CaC2O4,2H2O wedellite (Wd)) crystal type. The ANN-, SVM- and KNN model reached a 93.4%-, 94.2%- and 77.25% correct prediction rate, respectively. In third class that corresponds to the subjects who have negative crystalluria (NC), ANN-, SVM- and KNN model reached a 91.7%-, 87.8%- and 69.77% correct prediction rate, respectively. Compared to SVM- and KNN models, the developed system using ANN model has allowed us to discriminate the subjects. This system is important in clinical laboratories since it could be a helpful tool for provide information about the development, formation of urinary stones crystals and the determination of their crystal type.
Author Keywords: Urinary Lithiasis, Whewellite, Wedellite, Artificial neural network, k-nearest neighbor, Support Vector Machines, Intelligent Diagnosis System.
Abdelouahed Ait Ider1, Malika Echajia2, Constant TCHEKA3, Abdelkrim Merbouha4, and Mohamed Mbarki5
1 Department of chemistry and environment, Sultan Moulay Slimane University, Faculty of Science and Technology, Transdisciplinary Team of Analytical Science for Sustainable Development, PB 523, Beni Mellal, Morocco
2 Department of chemistry and environment, Sultan Moulay Slimane University, Faculty of Science and Technology, Transdisciplinary Team of Analytical Science for Sustainable Development, PB 523, Béni Mellal, Morocco
3 Laboratory of Physical and Theoretical Chemistry, Department of Inorganic Chemistry, Faculty of Science, P.O. Box 812, Yaoundé, Cameroon
4 Faculty of Sciences and Technology, Sultan Moulay Slimane University, Beni Mellal, Morocco
5 Department of chemistry and environment, Sultan Moulay Slimane University, Faculty of Science and Technology, Transdisciplinary Team of Analytical Science for Sustainable Development, PB 523, Beni Mellal, Morocco
Original language: English
Copyright © 2016 ISSR Journals. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Abstract
The goal of this study is to develop a simple intelligent urolithiasis diagnosis system. The accuracy of the system was determined by comparing the recognition rates of the Artificial Neural Networks (ANNs)-, k-nearest Neighbor (kNN)-, and Support Vector Machines (SVM) algorithms. The results showed that the ANN model was superior to SVM and KNN models in prediction. We aimed through this work to classify the subjects in three classes according to the chemical concentrations of variables (Ca, Ox, pCaOx, Ca/ Ox) using and according to their clinical status. The ANN model, used to determine the first class that contains the subjects presenting their urine a calcium oxalate monohydrate (CaC2O4,H2O : whewellite (Wh)) crystal type. This ANN model reached a correct prediction rate of 85.3%. Using SVM- and KNN model the correct prediction rate reached 82.6% and 65.55% respectively. The second class contains the subjects presenting a calcium oxalate dihydrate (CaC2O4,2H2O wedellite (Wd)) crystal type. The ANN-, SVM- and KNN model reached a 93.4%-, 94.2%- and 77.25% correct prediction rate, respectively. In third class that corresponds to the subjects who have negative crystalluria (NC), ANN-, SVM- and KNN model reached a 91.7%-, 87.8%- and 69.77% correct prediction rate, respectively. Compared to SVM- and KNN models, the developed system using ANN model has allowed us to discriminate the subjects. This system is important in clinical laboratories since it could be a helpful tool for provide information about the development, formation of urinary stones crystals and the determination of their crystal type.
Author Keywords: Urinary Lithiasis, Whewellite, Wedellite, Artificial neural network, k-nearest neighbor, Support Vector Machines, Intelligent Diagnosis System.
How to Cite this Article
Abdelouahed Ait Ider, Malika Echajia, Constant TCHEKA, Abdelkrim Merbouha, and Mohamed Mbarki, “Contribution to a diagnosis system of the Whewellite- and Weddellite crystals in human urinary using data minig algorithms,” International Journal of Innovation and Applied Studies, vol. 14, no. 2, pp. 297–303, January 2016.
