Pub Date : 2006-04-17DOI: 10.1109/SIU.2006.1659802
F. Tokan, Nurhan Türker, T. Yildirim
Recently, artificial neural networks are widely used in medical prognosis. The goal of this work is to predict whether a patient will live at least one year after a heart attack by using neural networks as an example of prognosis. With this aim, multi layer perceptrons (MLP), radial basis function networks (RBF), probabilistic neural networks (PNN), generalized regression neural networks (GRNN) and learning vector quantization networks (LVQ) are used. To demonstrate the real performances of the networks, not only classification accuracies but also receiver operation characteristics (ROC) analysis must be investigated. For this purpose, both sensitivity-specificity values and ROC curves are evaluated for all networks
{"title":"Determination of The Neural Network Performances In The Medical Prognosis By Roc Analysis","authors":"F. Tokan, Nurhan Türker, T. Yildirim","doi":"10.1109/SIU.2006.1659802","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659802","url":null,"abstract":"Recently, artificial neural networks are widely used in medical prognosis. The goal of this work is to predict whether a patient will live at least one year after a heart attack by using neural networks as an example of prognosis. With this aim, multi layer perceptrons (MLP), radial basis function networks (RBF), probabilistic neural networks (PNN), generalized regression neural networks (GRNN) and learning vector quantization networks (LVQ) are used. To demonstrate the real performances of the networks, not only classification accuracies but also receiver operation characteristics (ROC) analysis must be investigated. For this purpose, both sensitivity-specificity values and ROC curves are evaluated for all networks","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121116804","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2006-04-17DOI: 10.1109/SIU.2006.1659792
E. Ege, A. Saranli
Significant recent advances have been achieved in the mobile robotics area partly due to technological advances in low power computational resources, new power storage technologies and biologically inspired principles in robotics. Despite these advances, primary problems of robotics are still open for contribution. One of these problems is path planning/finding for mobile robots. There are a number of different methods proposed for path planning. One of the recent alternatives is the rapidly-exploring random trees (RRT). In this study, we propose a new approach to the nearest neighbor finding section of the method which promises to increase the speed performance of the algorithm
{"title":"A New Approach to Increase Performance of Rapidly-Exploring Random Trees (RRT) in Mobile Robotics","authors":"E. Ege, A. Saranli","doi":"10.1109/SIU.2006.1659792","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659792","url":null,"abstract":"Significant recent advances have been achieved in the mobile robotics area partly due to technological advances in low power computational resources, new power storage technologies and biologically inspired principles in robotics. Despite these advances, primary problems of robotics are still open for contribution. One of these problems is path planning/finding for mobile robots. There are a number of different methods proposed for path planning. One of the recent alternatives is the rapidly-exploring random trees (RRT). In this study, we propose a new approach to the nearest neighbor finding section of the method which promises to increase the speed performance of the algorithm","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"341 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122313327","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2006-04-17DOI: 10.1109/SIU.2006.1659791
H. Kul, I. Cavdar
In recent years, digital subscriber line (DSL) technology has been gaining popularity as a high speed access technology, thanks to its capability of delivering multimedia services. The major impairment for DSL technologies is crosstalk in the telephone lines. Therefore,in order to overcome this noise effect in DSL lines, complicated error mitigation and framing techniques have to be used in DSL systems. In this work, the crosstalk model of DSL lines is given, and the effect of crosstalk on the channel capacity of ADSL services (the most widely used DSL service type) is investigated.
{"title":"Modelling XDSL Lines under Crosstalk Effect","authors":"H. Kul, I. Cavdar","doi":"10.1109/SIU.2006.1659791","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659791","url":null,"abstract":"In recent years, digital subscriber line (DSL) technology has been gaining popularity as a high speed access technology, thanks to its capability of delivering multimedia services. The major impairment for DSL technologies is crosstalk in the telephone lines. Therefore,in order to overcome this noise effect in DSL lines, complicated error mitigation and framing techniques have to be used in DSL systems. In this work, the crosstalk model of DSL lines is given, and the effect of crosstalk on the channel capacity of ADSL services (the most widely used DSL service type) is investigated.","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116463372","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2006-04-17DOI: 10.1109/SIU.2006.1659682
A. Demirkol
In this paper, an alternative target density function (TDF) is proposed to image the radar targets. While the target density function is developed by a new technique in direction density function (DDF) form, it is obtained by considering a novel range and scanning angle plane different from the conventional methods. Although the imaging technique is obtained via the linear phased array radars, the problem associated with beamforming is bypassed in this algorithm
{"title":"Dogrultu Yogunluk Fonksiyonu Ile Radar Goruntu Isleme","authors":"A. Demirkol","doi":"10.1109/SIU.2006.1659682","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659682","url":null,"abstract":"In this paper, an alternative target density function (TDF) is proposed to image the radar targets. While the target density function is developed by a new technique in direction density function (DDF) form, it is obtained by considering a novel range and scanning angle plane different from the conventional methods. Although the imaging technique is obtained via the linear phased array radars, the problem associated with beamforming is bypassed in this algorithm","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125800160","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2006-04-17DOI: 10.1109/SIU.2006.1659742
M. Cavusoglu, Y. Serinağaoğlu, O. Eroğul
Several studies have done in order to determine the relationship between snoring and obstructive sleep apnea syndrome (OSAS). One of the common problem that is faced during the medical treatment of the apnea is the undetermination of the efficiency of the applied treatment in terms of objective criteria. It is needed to automatically detect each snoring episode in order to estimate the spectral features and determine the snoring sound intensity. In this study, an automatic detection system of acoustic snoring signals has been designed, to work with long duration respiratory sound recordings. The system was designed to select snoring episodes from simple snorers and OSAS patients and to reject the undesired waveforms. The sound recordings were taken from patients that are suspected of OSAS pathology while they were connected to the polysomnography in Gulhane Military Medical Academy (GMMA) Sleep Studies Laboratory. In order to validate the system, 500 snores were analysed taken from 30 patients with different apnea/hypopnea index (AHI) . Results were compared with manual annotations done by a medical doctor and the average sensitivity of the system is determined as 86%
{"title":"Analysing Snoring Sounds For Obstructive Sleep Apnea (OSA) Patients","authors":"M. Cavusoglu, Y. Serinağaoğlu, O. Eroğul","doi":"10.1109/SIU.2006.1659742","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659742","url":null,"abstract":"Several studies have done in order to determine the relationship between snoring and obstructive sleep apnea syndrome (OSAS). One of the common problem that is faced during the medical treatment of the apnea is the undetermination of the efficiency of the applied treatment in terms of objective criteria. It is needed to automatically detect each snoring episode in order to estimate the spectral features and determine the snoring sound intensity. In this study, an automatic detection system of acoustic snoring signals has been designed, to work with long duration respiratory sound recordings. The system was designed to select snoring episodes from simple snorers and OSAS patients and to reject the undesired waveforms. The sound recordings were taken from patients that are suspected of OSAS pathology while they were connected to the polysomnography in Gulhane Military Medical Academy (GMMA) Sleep Studies Laboratory. In order to validate the system, 500 snores were analysed taken from 30 patients with different apnea/hypopnea index (AHI) . Results were compared with manual annotations done by a medical doctor and the average sensitivity of the system is determined as 86%","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125268843","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2006-04-17DOI: 10.1109/SIU.2006.1659779
Ali Orkan Bayer, Tolga Q7iloglut, Meltem Turhan, Yondem Bilgisayar, Miihendisligi B6liimii Telektrik Ve Elektronik, Miihendisligi B6liimii
Large vocabulary continuous speech recognition can be performed with high accuracy for languages like English that do not have a rich morphological structure. However, the performance of these systems for agglutinative languages is very low. The major reason for that is, the language models that are built on the words do not perform well for agglutinative languages. In this study, three different language models that consider the structure of the agglutinative languages are investigated. Two of the models consider the subword units as the units of language modeling. The first one uses only the stem of the words as units, and the other one uses stems and endings of the words separately as the units. The third model, firstly, places the words into certain classes by using the co-occurrences of the words, and then uses these classes as the units of the language model. The performance of the models are tested by using two stage decoding; in the first stage, lattices are formed by using bi-gram models and then tri-gram models are used for recognition over these lattices. In this study, it is shown that the vocabulary coverage of the system seriously affects the recognition performance. For this reason, models that use stems and endings as the modeling unit perform better since their coverage of the vocabulary is higher. In addition to that, a single-pass decoder that can perform single pass decoding over these models is believed to increase the recognition performance
{"title":"Investigation of Different Language Models for Turkish Speech Recognition","authors":"Ali Orkan Bayer, Tolga Q7iloglut, Meltem Turhan, Yondem Bilgisayar, Miihendisligi B6liimii Telektrik Ve Elektronik, Miihendisligi B6liimii","doi":"10.1109/SIU.2006.1659779","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659779","url":null,"abstract":"Large vocabulary continuous speech recognition can be performed with high accuracy for languages like English that do not have a rich morphological structure. However, the performance of these systems for agglutinative languages is very low. The major reason for that is, the language models that are built on the words do not perform well for agglutinative languages. In this study, three different language models that consider the structure of the agglutinative languages are investigated. Two of the models consider the subword units as the units of language modeling. The first one uses only the stem of the words as units, and the other one uses stems and endings of the words separately as the units. The third model, firstly, places the words into certain classes by using the co-occurrences of the words, and then uses these classes as the units of the language model. The performance of the models are tested by using two stage decoding; in the first stage, lattices are formed by using bi-gram models and then tri-gram models are used for recognition over these lattices. In this study, it is shown that the vocabulary coverage of the system seriously affects the recognition performance. For this reason, models that use stems and endings as the modeling unit perform better since their coverage of the vocabulary is higher. In addition to that, a single-pass decoder that can perform single pass decoding over these models is believed to increase the recognition performance","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"157 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121482932","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2006-04-17DOI: 10.1109/SIU.2006.1659689
A. Kaya, E. Y. Yuksel
Impedance matching networks are used extensively in radio communications. In this study Pi matching circuit has been used. This circuit impedance region has been investigated analytically under the variable load condition. The rectangular microstrip antenna was designed at 2.5 GHz frequency by using microstrip feed line. Better return loss level has been obtained for this antenna using pi matching circuit. It is shown that theoretical, simulation and laboratory results are good agreement
{"title":"The Performance Analysis of the Antenna System Designed by Using The PI Matching Circuit","authors":"A. Kaya, E. Y. Yuksel","doi":"10.1109/SIU.2006.1659689","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659689","url":null,"abstract":"Impedance matching networks are used extensively in radio communications. In this study Pi matching circuit has been used. This circuit impedance region has been investigated analytically under the variable load condition. The rectangular microstrip antenna was designed at 2.5 GHz frequency by using microstrip feed line. Better return loss level has been obtained for this antenna using pi matching circuit. It is shown that theoretical, simulation and laboratory results are good agreement","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132759067","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2006-04-17DOI: 10.1109/SIU.2006.1659684
S. Altunkaya, M. Bayrak
The aim of this study is to help to improve new diagnosis methods for cardiovascular disease by designing a computer program and a hardware which realizes the recording of Korotkoff sounds produced during the measurement of blood pressure and basic analysis process. Korotkoff sounds have been traditionally used to measure blood pressure noninvasively by employing the auscultatory method. The recording system developed in this study can be used to produce a large database by recording some important information such as weight, age, and disease name in addition to Korotkoff sounds. Basic analysis processes like filtering and finding frequency component of recorded data is accomplished by means of the designed software. Thus, the necessity of using different computer programs for analyzing and recording data has been eliminated
{"title":"Recording System For Korotkoff Sound Signals","authors":"S. Altunkaya, M. Bayrak","doi":"10.1109/SIU.2006.1659684","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659684","url":null,"abstract":"The aim of this study is to help to improve new diagnosis methods for cardiovascular disease by designing a computer program and a hardware which realizes the recording of Korotkoff sounds produced during the measurement of blood pressure and basic analysis process. Korotkoff sounds have been traditionally used to measure blood pressure noninvasively by employing the auscultatory method. The recording system developed in this study can be used to produce a large database by recording some important information such as weight, age, and disease name in addition to Korotkoff sounds. Basic analysis processes like filtering and finding frequency component of recorded data is accomplished by means of the designed software. Thus, the necessity of using different computer programs for analyzing and recording data has been eliminated","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134010889","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2006-04-17DOI: 10.1109/SIU.2006.1659796
T. Baykaş, A. Yongaçoğlu
In this paper, we investigate the performance of several advanced discrete-time RAKE receivers under the effect of channel statistics estimation errors. The name of the first receiver is the decorrelating discrete-time RAKE receiver (D-DTR), which improves the performance in the presence of channel estimation errors in diffuse channels. The second receiver is the discrete-time version of the generalized RAKE (G-DTR) receiver. The G-DTR has been proposed for correlated interference mitigation. The last system is the generalized decorrelating discrete-time RAKE receiver (GD-DTR), which combines the benefits of the D-DTR and the G-DTR. In the literature, it is shown that the D-DTR is sensitive to the estimation of the channel covariance matrix. In this work, we reduced this sensitivity and according to our results in worst case scenario the performance of the D-DTR is equal to a conventional discrete-time RAKE receiver (C-DTR). For the case of the GD-DTR, we introduce a very simple method for the estimation of the noise plus interference statistics and we modify the channel statistics estimation method of the D-DTR to be valid for the GD-DTR. Our simulations show that the GD-DTR needs an estimation quality above some threshold for a performance gain compared to the C-DTR
{"title":"Performance of Advanced Discrete-Time RAKE Receivers","authors":"T. Baykaş, A. Yongaçoğlu","doi":"10.1109/SIU.2006.1659796","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659796","url":null,"abstract":"In this paper, we investigate the performance of several advanced discrete-time RAKE receivers under the effect of channel statistics estimation errors. The name of the first receiver is the decorrelating discrete-time RAKE receiver (D-DTR), which improves the performance in the presence of channel estimation errors in diffuse channels. The second receiver is the discrete-time version of the generalized RAKE (G-DTR) receiver. The G-DTR has been proposed for correlated interference mitigation. The last system is the generalized decorrelating discrete-time RAKE receiver (GD-DTR), which combines the benefits of the D-DTR and the G-DTR. In the literature, it is shown that the D-DTR is sensitive to the estimation of the channel covariance matrix. In this work, we reduced this sensitivity and according to our results in worst case scenario the performance of the D-DTR is equal to a conventional discrete-time RAKE receiver (C-DTR). For the case of the GD-DTR, we introduce a very simple method for the estimation of the noise plus interference statistics and we modify the channel statistics estimation method of the D-DTR to be valid for the GD-DTR. Our simulations show that the GD-DTR needs an estimation quality above some threshold for a performance gain compared to the C-DTR","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126998699","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2006-04-17DOI: 10.1109/SIU.2006.1659821
M. Bastan, Pinar Duygulu
We propose a new approach to object recognition problem motivated by the availability of large annotated image and video collections. Similar to translation from one language to another, this approach considers the object recognition problem as the translation of visual elements to words. The visual elements represented in feature space are first categorized into a finite set of blobs. Then, the correspondences between the blobs and the words are learned using a method adapted from statistical machine translation. Finally, the correspondences, in the form of a probability table, are used to predict words for particular image regions (region naming), for entire images (auto-annotation), or to associate the automatically generated speech transcript text with the correct video frames (video alignment). Experimental results are presented on TRECVID 2004 data set, which consists of about 150 hours of news videos associated with manual annotations and speech transcript text.
{"title":"Object Recognition and Auto-annotation In News Videos","authors":"M. Bastan, Pinar Duygulu","doi":"10.1109/SIU.2006.1659821","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659821","url":null,"abstract":"We propose a new approach to object recognition problem motivated by the availability of large annotated image and video collections. Similar to translation from one language to another, this approach considers the object recognition problem as the translation of visual elements to words. The visual elements represented in feature space are first categorized into a finite set of blobs. Then, the correspondences between the blobs and the words are learned using a method adapted from statistical machine translation. Finally, the correspondences, in the form of a probability table, are used to predict words for particular image regions (region naming), for entire images (auto-annotation), or to associate the automatically generated speech transcript text with the correct video frames (video alignment). Experimental results are presented on TRECVID 2004 data set, which consists of about 150 hours of news videos associated with manual annotations and speech transcript text.","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"462 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131796204","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：