Umit Ozsandikcioglu - Academia.edu (original) (raw)

Uploads

Papers by Umit Ozsandikcioglu

Turkish Journal of Electrical Engineering and Computer Sciences

Lung cancer has the highest death rates among all types of cancer worldwide. Detection of lung ca... more Lung cancer has the highest death rates among all types of cancer worldwide. Detection of lung cancer in its early stages significantly increases the survival rate. In this study, the aim is to improve the lung cancer detection performance of electronic noses (e-noses) with breath analysis by using two different types of gas sensor-based e-nose. The developed e-nose system consists of 14 quartz crystal microbalance (QCM) sensors and 8 metal oxide semiconductor (MOS) sensors. Breath samples were collected from a total of 100 volunteers, including 60 patients with lung cancer, 20 healthy nonsmokers, and 20 healthy smokers, and were classified using decision tree (DT), support vector machine (SVM), k-nearest neighbour (kNN), and random forest (RF) algorithms. Principal component analysis (PCA) and linear discriminant analysis (LDA) algorithms were used for dimension reduction. A classification accuracy of 86.34% and 75.48% was obtained using MOS and QCM sensor data, respectively. The overall results have shown that combining the sensor data increases the accuracy to 88.54%. Additionally, it can be indicated that the PCA and LDA algorithms have a positive effect on the performance. By using PCA and LDA algorithms, the accuracy increased up to 92.67% and 95.36%, respectively.

Karadeniz Teknik Üniversitesi, 2016

In this work an Electronic Nose system with low cost is developed in order to analyze human breat... more In this work an Electronic Nose system with low cost is developed in order to analyze human breath and this system's success is tested on diagnosing lung cancer. In this Electronic Nose system, Quartz Crystal Microbalance and Metal Oxide Semiconductor gas sensors are used. The sensors that are sensed to volatile organic compounds found in the breaths of patient lung cancer patients are selected. Breath examples of lung cancer patient and healthy people are analysed with this system. Data acquired from system are preprocessed and dimension of these data are reduced by Principal Component Analysis method. After this processed, features from data are extracted. Classification of data is examined with k-Nearest Neighbors, Support Vector Machines and Artificial Neural Networks algorithms. Maximum success rates obtained with these algorithms are 91.4%, 85.7%, 91.4% respectively.

Brain computer interface is a structure that allows systems to be controlled with signals from th... more Brain computer interface is a structure that allows systems to be controlled with signals from the brain. In this study, we investigated the features that could best represent the computer interface systems, different dimension reduction methods were applied to the feature matrices and the best classification method was chosen. EEG signals were taken from the data set III of the preparation of the “BCI III Competition” competition. Linear Discriminant Analysis, Stochastic Neighbor Embedding and Maximally Collapsing Metric Learning algorithms were applied to feature matrices as dimension reduction method. Extracted features are classified by k-Nearest Neighbor and Support Vector Machine methods. As a result, the size was reduced by the Linear Discriminant Analysis algorithm and the highest success rate was obtained as 90.2% from the EEG data classified by k-Nearest Neighbor algorithm.

2018 26th Signal Processing and Communications Applications Conference (SIU)

2017 25th Signal Processing and Communications Applications Conference (SIU), 2017

In this study; Electroencephalogram signals which have caused in the brain of imagining right and... more In this study; Electroencephalogram signals which have caused in the brain of imagining right and left hand movements of human are classified with k-Nearest Neighbor Algorithm, Multi Layer Perceptron and Support Vector Machines algorithms. By acquiring different features (mean value, standard deviation, variance, maximum, minimum, absolute value and Fourier transform) from this electroencephalogram data, the best representing features of this data have been investigated. When the results of the used features and classification algorithms are examined, the features obtained from Fourier Transform and k-Nearest Neighbor Algorithm have shown the most successful results in classifying electroencephalogram data.

In this work, by using an Electronic Nose, human breath samples were analyzed. Breaths samples we... more In this work, by using an Electronic Nose, human breath samples were analyzed. Breaths samples were collected from three different groups. These are lung cancer patients, healthy people and healthy smokers. In structure of developed E-Nose, Quartz Crystal Microbalance and Metal Oxide Semiconductor gas sensors were used. Data acquired from system were preprocessed and dimension of these data were reduced with Linear Discriminant Analysis algorithm. Classification of data was performed with k-Nearest Neighbors, Support Vector Machines algorithms. For both algorithms, maximum classification success rates were obtained as 94.1% with 5 Fold Cross Validation.

2018 IEEE International Symposium on Medical Measurements and Applications (MeMeA)

Volatile Organic Compounds in human breath can be indicator of certain diseases. In this work an ... more Volatile Organic Compounds in human breath can be indicator of certain diseases. In this work an Electronic Nose system was developed in order to analyze human breath about diagnose whether people with lung cancer or not. In this Electronic Nose system, S Metal Oxide Semiconductor gas sensors and 14 Quartz Crystal Microbalance sensors were used. All sensors used in this system are sensitive to Volatile Organic Compounds related with lung cancer. Data obtained by using this system were first preprocessed. Then features were extracted from data for increase success of classification. In order to remove unnecessary features, Principal Component Analysis was used. Last, extracted features were classified with k-Nearest Neighbours and Support Vector Machines methods.

2015 9th International Conference on Electrical and Electronics Engineering (ELECO), 2015

2015 23nd Signal Processing and Communications Applications Conference (SIU), 2015

2015 International Symposium on Innovations in Intelligent SysTems and Applications (INISTA), 2015

Turkish Journal of Electrical Engineering and Computer Sciences

Lung cancer has the highest death rates among all types of cancer worldwide. Detection of lung ca... more Lung cancer has the highest death rates among all types of cancer worldwide. Detection of lung cancer in its early stages significantly increases the survival rate. In this study, the aim is to improve the lung cancer detection performance of electronic noses (e-noses) with breath analysis by using two different types of gas sensor-based e-nose. The developed e-nose system consists of 14 quartz crystal microbalance (QCM) sensors and 8 metal oxide semiconductor (MOS) sensors. Breath samples were collected from a total of 100 volunteers, including 60 patients with lung cancer, 20 healthy nonsmokers, and 20 healthy smokers, and were classified using decision tree (DT), support vector machine (SVM), k-nearest neighbour (kNN), and random forest (RF) algorithms. Principal component analysis (PCA) and linear discriminant analysis (LDA) algorithms were used for dimension reduction. A classification accuracy of 86.34% and 75.48% was obtained using MOS and QCM sensor data, respectively. The overall results have shown that combining the sensor data increases the accuracy to 88.54%. Additionally, it can be indicated that the PCA and LDA algorithms have a positive effect on the performance. By using PCA and LDA algorithms, the accuracy increased up to 92.67% and 95.36%, respectively.

Karadeniz Teknik Üniversitesi, 2016

In this work an Electronic Nose system with low cost is developed in order to analyze human breat... more In this work an Electronic Nose system with low cost is developed in order to analyze human breath and this system's success is tested on diagnosing lung cancer. In this Electronic Nose system, Quartz Crystal Microbalance and Metal Oxide Semiconductor gas sensors are used. The sensors that are sensed to volatile organic compounds found in the breaths of patient lung cancer patients are selected. Breath examples of lung cancer patient and healthy people are analysed with this system. Data acquired from system are preprocessed and dimension of these data are reduced by Principal Component Analysis method. After this processed, features from data are extracted. Classification of data is examined with k-Nearest Neighbors, Support Vector Machines and Artificial Neural Networks algorithms. Maximum success rates obtained with these algorithms are 91.4%, 85.7%, 91.4% respectively.

Brain computer interface is a structure that allows systems to be controlled with signals from th... more Brain computer interface is a structure that allows systems to be controlled with signals from the brain. In this study, we investigated the features that could best represent the computer interface systems, different dimension reduction methods were applied to the feature matrices and the best classification method was chosen. EEG signals were taken from the data set III of the preparation of the “BCI III Competition” competition. Linear Discriminant Analysis, Stochastic Neighbor Embedding and Maximally Collapsing Metric Learning algorithms were applied to feature matrices as dimension reduction method. Extracted features are classified by k-Nearest Neighbor and Support Vector Machine methods. As a result, the size was reduced by the Linear Discriminant Analysis algorithm and the highest success rate was obtained as 90.2% from the EEG data classified by k-Nearest Neighbor algorithm.

2018 26th Signal Processing and Communications Applications Conference (SIU)

2017 25th Signal Processing and Communications Applications Conference (SIU), 2017

In this study; Electroencephalogram signals which have caused in the brain of imagining right and... more In this study; Electroencephalogram signals which have caused in the brain of imagining right and left hand movements of human are classified with k-Nearest Neighbor Algorithm, Multi Layer Perceptron and Support Vector Machines algorithms. By acquiring different features (mean value, standard deviation, variance, maximum, minimum, absolute value and Fourier transform) from this electroencephalogram data, the best representing features of this data have been investigated. When the results of the used features and classification algorithms are examined, the features obtained from Fourier Transform and k-Nearest Neighbor Algorithm have shown the most successful results in classifying electroencephalogram data.

In this work, by using an Electronic Nose, human breath samples were analyzed. Breaths samples we... more In this work, by using an Electronic Nose, human breath samples were analyzed. Breaths samples were collected from three different groups. These are lung cancer patients, healthy people and healthy smokers. In structure of developed E-Nose, Quartz Crystal Microbalance and Metal Oxide Semiconductor gas sensors were used. Data acquired from system were preprocessed and dimension of these data were reduced with Linear Discriminant Analysis algorithm. Classification of data was performed with k-Nearest Neighbors, Support Vector Machines algorithms. For both algorithms, maximum classification success rates were obtained as 94.1% with 5 Fold Cross Validation.

2018 IEEE International Symposium on Medical Measurements and Applications (MeMeA)

Volatile Organic Compounds in human breath can be indicator of certain diseases. In this work an ... more Volatile Organic Compounds in human breath can be indicator of certain diseases. In this work an Electronic Nose system was developed in order to analyze human breath about diagnose whether people with lung cancer or not. In this Electronic Nose system, S Metal Oxide Semiconductor gas sensors and 14 Quartz Crystal Microbalance sensors were used. All sensors used in this system are sensitive to Volatile Organic Compounds related with lung cancer. Data obtained by using this system were first preprocessed. Then features were extracted from data for increase success of classification. In order to remove unnecessary features, Principal Component Analysis was used. Last, extracted features were classified with k-Nearest Neighbours and Support Vector Machines methods.

2015 9th International Conference on Electrical and Electronics Engineering (ELECO), 2015

2015 23nd Signal Processing and Communications Applications Conference (SIU), 2015

2015 International Symposium on Innovations in Intelligent SysTems and Applications (INISTA), 2015