Fırat Üniversitesi Mühendislik Bilimleri Dergisi, 2022
Korona virüsün (COVID-19) hızlı bulaşması nedeniyle dünya büyük bir sağlık kriziyle karşı karşıya... more Korona virüsün (COVID-19) hızlı bulaşması nedeniyle dünya büyük bir sağlık kriziyle karşı karşıya kalmıştır. Korona virüsün yayılmasını engellemek için Dünya Sağlık Örgütüne (WHO) göre en etkili tedbir, halka açık yerlerde ve kalabalık alanlarda maske takmaktır. Ancak kalabalık ortamlarda uzun süre kalan kişilerde sıkılma, boş verme ve umursamazlık gibi nedenlerle insanlar bu kuralı ihlal edebilmektedir. Bu nedenle kalabalık alanlarda insanların izlenmesi ve gerektiğinde ilgililerin uyarılarak toplum sağlığını korumak önem arz etmektedir. Bu çalışmada maske takmayan, maskesini yanlış takan ve maskesini doğru takan kişileri belirleme sürecini otomatikleştirmek için iki derin öğrenme modeli kullanan bir robotik model geliştirilmiştir. İnternetten elde edilen veri setleri ve çevreden alınan fotoğraflar kullanılarak özgün bir veri seti oluşturulmuştur. Geliştirilen yapay zekâ modellerinin daha iyi tahmin sonuçları verebilmesi için veri seti görüntüleri üzerinde veri çoğaltma (aynalama, ...
Bu caly?mada, sa?lykly ve epilepsi hastasy olan ki?ilerden alynan EEG i?aretleri farkly oni?leme ... more Bu caly?mada, sa?lykly ve epilepsi hastasy olan ki?ilerden alynan EEG i?aretleri farkly oni?leme ve synyflandyrma yontemleri kullanylarak analiz edilmi? ve bu yontemlerin ba?ary oranlary kar?yla?tyrylmy?tyr. EEG i?aretlerinin synyflandyrylmasy veya obekle?tirilmesi icin oznitelik cykarym (Ortalama Mutlak De?er (OMD), Yule-Walker AR ve Kovaryans AR), obekle?tirme (K-Ortalama, Bulanyk C-Ortalama (BCO) ) ve synyflandyrma (Destek Vektor Analizi ve Lineer Diskriminant Analizi) metotlary kullanylmy?tyr. Elde edilen yuksek synyflandyrma basarym sonuclary kar?yla?tyrmaly olarak verilmi?tir.
INTERNATIONAL JOURNAL OF INFORMATICS TECHNOLOGIES, 2016
Tip arastirmacilari tarafindan sik kullanilan bir arama motoru olan Pubmed, MEDLINE veri tabanind... more Tip arastirmacilari tarafindan sik kullanilan bir arama motoru olan Pubmed, MEDLINE veri tabaninda uzerinde sorgulama yapmaktadir. MEDLINE medikal, biyoloji ve genetik alanindaki calismalari iceren ve surekli guncel tutulan bibliyografik bir veri tabanidir. Icerdigi yuksek hacimdeki yapisal olmayan metinler sebebiyle, MEDLINE veri tabani veya belli bolumleri uzerinde pek cok metin siniflandirma calismalari mevcuttur. Bu calismada kanser turleri hakkinda yazilmis makale ozetlerini inceleyerek makalenin hangi kanser turuyle ilgili oldugunu otomatik bulan bir metot gelistirilmistir. Metodu egitmek ve test etmek icin MEDLINE veri tabani uzerinde 25962 makale ozeti, Pubmed arama motoru uzerinden ayrica gelistirilen bir program (crawler) uzerinden toplanmistir. Elde edilen veri seti uzerinde iki ayri calisma yurutulmustur. Birinci calismada, gelistirilen metot ozellik secim yontemi uygulamadan ve Ki-Kare ve Bilgi Kazanci ozellik secim yontemlerini uygulayarak, Naif Bayes ve Destek Vektor ...
Electroensefalogram (EEG) isaretleri beyin yuzeyinden algilanan dusuk genlikli biyoelektrik isare... more Electroensefalogram (EEG) isaretleri beyin yuzeyinden algilanan dusuk genlikli biyoelektrik isaretler olup, beyin fonksiyonlari hakkinda cok miktarda bilgi icermektedir. Bu sebeple, EEG isaretleri ozellikle tip’ta bir cok beyin rahatsizliklari nin teshisinde yaygin olarak kullanilir. Bu calismada, saglikli ve hasta (epilepsi sikayeti olan) kisilerden toplanan EEG isaretlerinin spektrumlarinin incelenmesinde, altuzay yontemlerinden MUSIC (Multiple Signal Classification) ve ozvektor (Eigenvektor:EV) yontemleri kullanilmistir. Bir karsilastirma yapabilmek icin elde edilen spektrumlar, klasik yontemlerden Welch yontemi ile frekans cozunurlugu ve frekans iceriginin belirlenmesini kolaylastiran etkiler acisindan karsilastirilmistir. Net bir sonuc alabilmek amaciyla yontemler once gercek EEG isaretlerine uygulanmis daha sonra da frekans icerigi bilinen simule isaretlere uygulanarak karsilastirma yapilmistir. Elde edilen spektrumlar incelendiginde EEG’de frekans analizi yapilmasinda alt uza...
Bilgisayar destekli tanı (BDT) sistemleri, çeşitli hasta bilgilerini kullanarak doktora yardımcı ... more Bilgisayar destekli tanı (BDT) sistemleri, çeşitli hasta bilgilerini kullanarak doktora yardımcı karar destek sistemi oluşturmak amacıyla son yıllarda sıklıkla kullanılmaktadır. Bu çalışmada BDT sistemine yönelik yapılan karar ağaçları algoritması uygulamasıyla, Wechsler Çocuklar için Zekâ Ölçeği (WISC-R) profillerinin, sınır zekâ (SZ), hafif ve orta düzeyde zihinsel yetersizlik (ZY) teşhisindeki sınıflandırma başarısının karşılaştırılması amaçlanmıştır. Çalışmanın veri seti DSMV’ e göre tanı konan 50 SZ, 61 hafif düzeyde ZY ve 21 orta düzeyde ZY olmak üzere toplam 132 hastanın WISC-R testi sonuç raporları kullanılarak oluşturulmuştur. WISC-R puanlarının sonuca etkisinin karşılaştırılabilmesi için 132 hastanın test puanları: toplam, sözel ve performans zekâ bölümü puanları; sözel ve performans zekâ bölümü alt ölçek puanları ve bu ikisinden oluşan 3 ayrı veri seti oluşturulmuştur. WISC-R testinin bütün puan türlerini içeren veri setinde, ilk iki düğüm toplam zekâ bölümü puanı seçilmi...
Concurrency and Computation: Practice and Experience, 2022
Diabetic retinopathy (DR) is a pathology occurring in the optic nerve due to an excessive blood s... more Diabetic retinopathy (DR) is a pathology occurring in the optic nerve due to an excessive blood sugar level in human body. It is one of the major reasons for visual impairment in the developed and developing countries. Patients with DR usually suffer from visual damages due to a high blood sugar level in retinal blood vessel walls. These damages may also leak into other retinal layers of the eye within time. As a result of these leakages and nutritional disorders, a number of lesions such as excudate, edema, microaneurysm, and hemorrhage may occur. In this respect, an accurate and effective detection of these lesions in earlier stages of DR plays an important role in the progression of the disease. In the proposed study, exudate and hemorrhages, which are important clinical findings for DR, were automatically detected from low contrast colored fundus images. Exudate and hemorrhages are lesions with different characteristics. However, in this study, high performance was achieved by making a three‐class semantic segmentation. In addition, a color space transformation was performed and the classical U‐Net algorithm was provided to achieve stable high performance in low contrast images. Finally, lesion images which were manually detected by a physician were matched with automatically segmented excudate and hemorrhage images using the proposed method. Thus, both segmentation and lesion detection performances of the proposed method were measured. The findings demonstrated that Dice and Jaccard similarity indexes were calculated nearly as 0.95 for the segmentation performance. A sensitivity of 98% and specificity value of 91% were measured for detection performance. It can be inferred from these figures that the proposed method can be effectively used as a supporting system by physicians for the detection and classification of lesions in the color fundus images for the diagnosis of DR.
2017 International Artificial Intelligence and Data Processing Symposium (IDAP), 2017
Detection of optic disc in retinal images is an important step in disease diagnosis and patient f... more Detection of optic disc in retinal images is an important step in disease diagnosis and patient follow-up. Optic disc detection is a Preprocess step for the diagnosis of many diseases, such as glaucoma and DP (Diabetic Retinopathy), which are vital for the eye. In this study, a hybrid approach is proposed for fully automatic segmentation of Optic Disc (OD). The first phase of the study consists of OD localization. The OD localization coordinate obtained in the study was used as an input to the method of region segmentation, which is a semi-automatic segmentation method, used as an initial point for OD segmentation. In this way, fully automatic segmentation is done. The success of the hybrid approach was evaluated according to both localization and segmentation. Performance of the study was evaluated with 30 images according to Dice and Jaccard Similarity. As a result of the evaluation have been obtained respectively % 92 and % 87 success.
2017 International Artificial Intelligence and Data Processing Symposium (IDAP), 2017
Developing technological infrastructure has enabled the development of computer based biomedical ... more Developing technological infrastructure has enabled the development of computer based biomedical systems as in many areas in the field of medicine. One of these systems are biomedical imaging systems. Many studies are being conducted for a new image processing techniques to improve the performance of this systems. Noise reduction is one of the important steps in biomedical image processing. In this study, noise reduction was performed to emphasize coronary arteries on x-ray heart angiography images. For this purpose, the original angiography images were smoothed using non-local averages. Thus, insignificant groups of pixels in the image described as noise has been removed. Then, the boundaries of coronary arteries are sharpened with first and second derivatives of image based combined enhancement method. It is seen that the mean square error values obtained by the proposed method are more successful when compared with the noise reduction results obtained using the Wiener filter. These results show that the non-local means method can be used as a successful pre-processing method for noise reduction in angiography images.
In this study, EMG signals taken from the skin surface as a result of muscles' contraction ar... more In this study, EMG signals taken from the skin surface as a result of muscles' contraction are classified. Studied EMG signals include 400 different patterns relating to four different movements. Each pattern is obtained by adding EMG signals one after another, which are recorded synchronously from two different muscles relating to one movement. Support Vector Machine (SVM) classifier, a supervised method, is used to classify these pattterns. But signals need to be preprocessed before being used in SVM classifier. To this end, spectral methods are consulted. In this way, feature vectors which are more significant than raw data and are composed of coefficients are achieved. Four different methods are used for preprocessing and feature vectors obtained are classified by SVM. Success of SVM classifier is tested and performances of preprocessing methods are compared. Best achievement is 94.25%. Keywords: EMG; Spectral Methods; Autoregressive (AR); SVM Classifier.
PV modules are one of the most important alternative energy tools that can directly convert solar... more PV modules are one of the most important alternative energy tools that can directly convert solar energy into electrical energy. Rapidly developing PV technology has been set up to process of integration PV systems into everyday life of people. Obtaining of a general mathematical model of solar cell is an important event for designing tools that run with solar energy. In this study, a general mathematical model of solar cells has been obtained and Matlab/Simulink software based simulation of this model has been visually programmed.. Proposed model and PVSYSTEM toolbox can be used with other hybrid systems to develop solar cell simulations for training to engineers, designers and students.
Fırat Üniversitesi Mühendislik Bilimleri Dergisi, 2022
Korona virüsün (COVID-19) hızlı bulaşması nedeniyle dünya büyük bir sağlık kriziyle karşı karşıya... more Korona virüsün (COVID-19) hızlı bulaşması nedeniyle dünya büyük bir sağlık kriziyle karşı karşıya kalmıştır. Korona virüsün yayılmasını engellemek için Dünya Sağlık Örgütüne (WHO) göre en etkili tedbir, halka açık yerlerde ve kalabalık alanlarda maske takmaktır. Ancak kalabalık ortamlarda uzun süre kalan kişilerde sıkılma, boş verme ve umursamazlık gibi nedenlerle insanlar bu kuralı ihlal edebilmektedir. Bu nedenle kalabalık alanlarda insanların izlenmesi ve gerektiğinde ilgililerin uyarılarak toplum sağlığını korumak önem arz etmektedir. Bu çalışmada maske takmayan, maskesini yanlış takan ve maskesini doğru takan kişileri belirleme sürecini otomatikleştirmek için iki derin öğrenme modeli kullanan bir robotik model geliştirilmiştir. İnternetten elde edilen veri setleri ve çevreden alınan fotoğraflar kullanılarak özgün bir veri seti oluşturulmuştur. Geliştirilen yapay zekâ modellerinin daha iyi tahmin sonuçları verebilmesi için veri seti görüntüleri üzerinde veri çoğaltma (aynalama, ...
Bu caly?mada, sa?lykly ve epilepsi hastasy olan ki?ilerden alynan EEG i?aretleri farkly oni?leme ... more Bu caly?mada, sa?lykly ve epilepsi hastasy olan ki?ilerden alynan EEG i?aretleri farkly oni?leme ve synyflandyrma yontemleri kullanylarak analiz edilmi? ve bu yontemlerin ba?ary oranlary kar?yla?tyrylmy?tyr. EEG i?aretlerinin synyflandyrylmasy veya obekle?tirilmesi icin oznitelik cykarym (Ortalama Mutlak De?er (OMD), Yule-Walker AR ve Kovaryans AR), obekle?tirme (K-Ortalama, Bulanyk C-Ortalama (BCO) ) ve synyflandyrma (Destek Vektor Analizi ve Lineer Diskriminant Analizi) metotlary kullanylmy?tyr. Elde edilen yuksek synyflandyrma basarym sonuclary kar?yla?tyrmaly olarak verilmi?tir.
INTERNATIONAL JOURNAL OF INFORMATICS TECHNOLOGIES, 2016
Tip arastirmacilari tarafindan sik kullanilan bir arama motoru olan Pubmed, MEDLINE veri tabanind... more Tip arastirmacilari tarafindan sik kullanilan bir arama motoru olan Pubmed, MEDLINE veri tabaninda uzerinde sorgulama yapmaktadir. MEDLINE medikal, biyoloji ve genetik alanindaki calismalari iceren ve surekli guncel tutulan bibliyografik bir veri tabanidir. Icerdigi yuksek hacimdeki yapisal olmayan metinler sebebiyle, MEDLINE veri tabani veya belli bolumleri uzerinde pek cok metin siniflandirma calismalari mevcuttur. Bu calismada kanser turleri hakkinda yazilmis makale ozetlerini inceleyerek makalenin hangi kanser turuyle ilgili oldugunu otomatik bulan bir metot gelistirilmistir. Metodu egitmek ve test etmek icin MEDLINE veri tabani uzerinde 25962 makale ozeti, Pubmed arama motoru uzerinden ayrica gelistirilen bir program (crawler) uzerinden toplanmistir. Elde edilen veri seti uzerinde iki ayri calisma yurutulmustur. Birinci calismada, gelistirilen metot ozellik secim yontemi uygulamadan ve Ki-Kare ve Bilgi Kazanci ozellik secim yontemlerini uygulayarak, Naif Bayes ve Destek Vektor ...
Electroensefalogram (EEG) isaretleri beyin yuzeyinden algilanan dusuk genlikli biyoelektrik isare... more Electroensefalogram (EEG) isaretleri beyin yuzeyinden algilanan dusuk genlikli biyoelektrik isaretler olup, beyin fonksiyonlari hakkinda cok miktarda bilgi icermektedir. Bu sebeple, EEG isaretleri ozellikle tip’ta bir cok beyin rahatsizliklari nin teshisinde yaygin olarak kullanilir. Bu calismada, saglikli ve hasta (epilepsi sikayeti olan) kisilerden toplanan EEG isaretlerinin spektrumlarinin incelenmesinde, altuzay yontemlerinden MUSIC (Multiple Signal Classification) ve ozvektor (Eigenvektor:EV) yontemleri kullanilmistir. Bir karsilastirma yapabilmek icin elde edilen spektrumlar, klasik yontemlerden Welch yontemi ile frekans cozunurlugu ve frekans iceriginin belirlenmesini kolaylastiran etkiler acisindan karsilastirilmistir. Net bir sonuc alabilmek amaciyla yontemler once gercek EEG isaretlerine uygulanmis daha sonra da frekans icerigi bilinen simule isaretlere uygulanarak karsilastirma yapilmistir. Elde edilen spektrumlar incelendiginde EEG’de frekans analizi yapilmasinda alt uza...
Bilgisayar destekli tanı (BDT) sistemleri, çeşitli hasta bilgilerini kullanarak doktora yardımcı ... more Bilgisayar destekli tanı (BDT) sistemleri, çeşitli hasta bilgilerini kullanarak doktora yardımcı karar destek sistemi oluşturmak amacıyla son yıllarda sıklıkla kullanılmaktadır. Bu çalışmada BDT sistemine yönelik yapılan karar ağaçları algoritması uygulamasıyla, Wechsler Çocuklar için Zekâ Ölçeği (WISC-R) profillerinin, sınır zekâ (SZ), hafif ve orta düzeyde zihinsel yetersizlik (ZY) teşhisindeki sınıflandırma başarısının karşılaştırılması amaçlanmıştır. Çalışmanın veri seti DSMV’ e göre tanı konan 50 SZ, 61 hafif düzeyde ZY ve 21 orta düzeyde ZY olmak üzere toplam 132 hastanın WISC-R testi sonuç raporları kullanılarak oluşturulmuştur. WISC-R puanlarının sonuca etkisinin karşılaştırılabilmesi için 132 hastanın test puanları: toplam, sözel ve performans zekâ bölümü puanları; sözel ve performans zekâ bölümü alt ölçek puanları ve bu ikisinden oluşan 3 ayrı veri seti oluşturulmuştur. WISC-R testinin bütün puan türlerini içeren veri setinde, ilk iki düğüm toplam zekâ bölümü puanı seçilmi...
Concurrency and Computation: Practice and Experience, 2022
Diabetic retinopathy (DR) is a pathology occurring in the optic nerve due to an excessive blood s... more Diabetic retinopathy (DR) is a pathology occurring in the optic nerve due to an excessive blood sugar level in human body. It is one of the major reasons for visual impairment in the developed and developing countries. Patients with DR usually suffer from visual damages due to a high blood sugar level in retinal blood vessel walls. These damages may also leak into other retinal layers of the eye within time. As a result of these leakages and nutritional disorders, a number of lesions such as excudate, edema, microaneurysm, and hemorrhage may occur. In this respect, an accurate and effective detection of these lesions in earlier stages of DR plays an important role in the progression of the disease. In the proposed study, exudate and hemorrhages, which are important clinical findings for DR, were automatically detected from low contrast colored fundus images. Exudate and hemorrhages are lesions with different characteristics. However, in this study, high performance was achieved by making a three‐class semantic segmentation. In addition, a color space transformation was performed and the classical U‐Net algorithm was provided to achieve stable high performance in low contrast images. Finally, lesion images which were manually detected by a physician were matched with automatically segmented excudate and hemorrhage images using the proposed method. Thus, both segmentation and lesion detection performances of the proposed method were measured. The findings demonstrated that Dice and Jaccard similarity indexes were calculated nearly as 0.95 for the segmentation performance. A sensitivity of 98% and specificity value of 91% were measured for detection performance. It can be inferred from these figures that the proposed method can be effectively used as a supporting system by physicians for the detection and classification of lesions in the color fundus images for the diagnosis of DR.
2017 International Artificial Intelligence and Data Processing Symposium (IDAP), 2017
Detection of optic disc in retinal images is an important step in disease diagnosis and patient f... more Detection of optic disc in retinal images is an important step in disease diagnosis and patient follow-up. Optic disc detection is a Preprocess step for the diagnosis of many diseases, such as glaucoma and DP (Diabetic Retinopathy), which are vital for the eye. In this study, a hybrid approach is proposed for fully automatic segmentation of Optic Disc (OD). The first phase of the study consists of OD localization. The OD localization coordinate obtained in the study was used as an input to the method of region segmentation, which is a semi-automatic segmentation method, used as an initial point for OD segmentation. In this way, fully automatic segmentation is done. The success of the hybrid approach was evaluated according to both localization and segmentation. Performance of the study was evaluated with 30 images according to Dice and Jaccard Similarity. As a result of the evaluation have been obtained respectively % 92 and % 87 success.
2017 International Artificial Intelligence and Data Processing Symposium (IDAP), 2017
Developing technological infrastructure has enabled the development of computer based biomedical ... more Developing technological infrastructure has enabled the development of computer based biomedical systems as in many areas in the field of medicine. One of these systems are biomedical imaging systems. Many studies are being conducted for a new image processing techniques to improve the performance of this systems. Noise reduction is one of the important steps in biomedical image processing. In this study, noise reduction was performed to emphasize coronary arteries on x-ray heart angiography images. For this purpose, the original angiography images were smoothed using non-local averages. Thus, insignificant groups of pixels in the image described as noise has been removed. Then, the boundaries of coronary arteries are sharpened with first and second derivatives of image based combined enhancement method. It is seen that the mean square error values obtained by the proposed method are more successful when compared with the noise reduction results obtained using the Wiener filter. These results show that the non-local means method can be used as a successful pre-processing method for noise reduction in angiography images.
In this study, EMG signals taken from the skin surface as a result of muscles' contraction ar... more In this study, EMG signals taken from the skin surface as a result of muscles' contraction are classified. Studied EMG signals include 400 different patterns relating to four different movements. Each pattern is obtained by adding EMG signals one after another, which are recorded synchronously from two different muscles relating to one movement. Support Vector Machine (SVM) classifier, a supervised method, is used to classify these pattterns. But signals need to be preprocessed before being used in SVM classifier. To this end, spectral methods are consulted. In this way, feature vectors which are more significant than raw data and are composed of coefficients are achieved. Four different methods are used for preprocessing and feature vectors obtained are classified by SVM. Success of SVM classifier is tested and performances of preprocessing methods are compared. Best achievement is 94.25%. Keywords: EMG; Spectral Methods; Autoregressive (AR); SVM Classifier.
PV modules are one of the most important alternative energy tools that can directly convert solar... more PV modules are one of the most important alternative energy tools that can directly convert solar energy into electrical energy. Rapidly developing PV technology has been set up to process of integration PV systems into everyday life of people. Obtaining of a general mathematical model of solar cell is an important event for designing tools that run with solar energy. In this study, a general mathematical model of solar cells has been obtained and Matlab/Simulink software based simulation of this model has been visually programmed.. Proposed model and PVSYSTEM toolbox can be used with other hybrid systems to develop solar cell simulations for training to engineers, designers and students.
Uploads
Papers by ahmet alkan