Individuals with chronic kidney disease (CKD) are often not aware that the medical tests they take for other purposes may contain useful information about CKD, and that this information is sometimes not used effectively to tackle the identification of the disease. Therefore, attributes of different medical tests are investigated to identify which attributes may contain useful information about CKD. A database with several attributes of healthy subjects and subjects with CKD are analyzed using different techniques. Common spatial pattern (CSP) filter and linear discriminant analysis are first used to identify the dominant attributes that could contribute in detecting CKD. Here, the CSP filter is applied to optimize a separation between CKD and nonCKD subjects. Then, classification methods are also used to identify the dominant attributes. These analyses suggest that hemoglobin, albumin, specific gravity, hypertension, and diabetes mellitus, together with serum creatinine, are the most important attributes in the early detection of CKD. Further, it suggests that in the absence of information on hypertension and diabetes mellitus, random blood glucose and blood pressure attributes may be used.
The automatic identification of different electric loads using the current waveform at the time of switching, is analysed in this paper. The time variation of the harmonics at the time of switching is modelled using the Hidden Markov Model with Gaussian Mixture Models representing the probabilities. Short Time Fourier Transform (STFT) and Wavelet Transform (WT) based features are compared at their optimum configurations. The STFT based feature gave an accuracy of 97.9% while the WT features provided an accuracy of 93.75% in a cross fold validation experiment.Index Terms-Electric load identification, harmonic analysis, Hidden Markov Model, Wavelet Transform.
Speech-based authentication system can perform remote authentication over telephone channels. However, telephone channels are restricted to a bandwidth of ∼0-4 kHz while studies on the distribution of speakerspecific information in the speech spectrum strongly suggests that useful speaker-specific information is present above 4 kHz. A method to shift a part of this speaker-specific information above 4 kHz into the telephone bandwidth in place of less speaker-specific information originally present below 4 kHz is proposed. Speaker recognition experiments conducted using the proposed method leads to ∼18.5% relative improvement on equal error rate when compared to a system using the conventional telephone band speech, as evaluated on the Intelligence Advanced Research Projects Activity (IARPA) Babel Program Tamil language collection.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.