RESUMO
In this study, feature extraction methods used in the classification of single-channel lung sounds obtained by automatic identification of respiratory cycles were examined in detail in order to extract distinctive features at the lowest size. In this way, it will be possible to design a system for the detection of lung diseases, completely autonomously. In the study, automatic separation and classification of 400 respiratory cycles were performed from the single-channel common lung sounds obtained from 94 people. Leave one out cross validation (LOOCV) was used for the calibration and validation of the classification model. The Mel frequency cepstrum coefficients (MFCC), time domain features, frequency domain features, and linear predictive coding (LPC) were used for classification. The performance of the features was tested using linear discriminant analysis (LDA), k-nearest neighbors (k-NN), support vector machines (SVM), and naive Bayes (NB) classification algorithms. The success of combinations of features was explored and enhanced using the sequential forward selection (SFS). As a result, the best accuracy (90.14% in the training set and 90.63% in the test set) was acquired using the k-NN for the triple combination, which included the standard deviation of LPC and the standard deviation and the mean of MFCC.
Assuntos
Sons Respiratórios , Máquina de Vetores de Suporte , Algoritmos , Teorema de Bayes , Análise Discriminante , HumanosRESUMO
INTRODUCTION: A century ago, Kraepelin stated that the distinctive feature of schizophrenia was progressive deterioration. Kraepelin criteria for schizophrenia are: (1) continuous hospitalization or complete dependence on others for obtaining basic necessities of life, (2) unemployment and (3) no remission for the past five years. We aimed to determine the clinical appearance and structural biological features of Kraepelinian schizophrenia. METHODS: The sample consisted of 17 Kraepelinian patients, 30 non-Kraepelinian schizophrenic patients and 43 healthy controls. The Clinical Global Impressions (CGI) and the Positive and Negative Syndrome Scales (PANSS) were used for clinical assessment. The Frontal Assessment Battery (FAB) and the Verbal Fluency and Color Trail Test (CTT) were included in the cognitive battery. Brain magnetic resonance imaging and dermatoglyphic measurements were performed for structural features. RESULT: Duration of illness, hospitalization, suicide attempts, admission type, presence of a stressor and treatment choice were similar between the two patient groups. Treatment resistance and family history of schizophrenia were more common in Kraepelinian patients. PANSS and CGI subscales scores were also higher in this group. Only the category fluency and CTT-I were different in Kraepelinian patients in comparison to the other patient group. Structural findings were not different between the three groups. CONCLUSION: Category fluency, which was lower in Kraepelinian patients, is an important marker of a degenerative process. The collection of severe clinical symptoms, family history of psychiatric illness and nonresponse to treatment in this particular group of patients points to the need to conduct further studies including cluster analysis in methodology.