Modelos Acústicos HMM Multimodales para Sonidos Cardiacos y Pulmonares / Multimodal acoustic models HMM applied to heart and lung sounds

Este artículo muestra el proceso de clasificación de señales bioacústicas normales y anormales registradas sobre el tórax humano lo cual incluye los sonidos de corazón y del pulmón. La idea específica es diseñar un sistema de clasificación de señales basado en técnicas de modelado acústico empleando particularmente modelos HMM para detectar secuencias de eventos, y GMM para modelar cúmulos que corresponden a los datos de los eventos. Las modalidades para extraer las características de los datos son vectores MFCC y Octiles. Esta aproximación tiene el potencial de mejorar la clasificación de la precisión en indicadores de diagnóstico auscultatorios, esto es interesante ya que los modelos HMM han demostrado ser menos sensibles al ruido en estudios previos. Resultados preliminares demuestran una precisión del 95% en clasificación de las señales de sonido evaluadas. Esto es particularmente critico tomando en cuenta la interferencia ambiental en una variedad de consultorios médicos. Debido a que algunas frecuencias del sonido cardiaco son paralelas a los sonidos pulmonares, estas pueden ser modeladas a partir de un mismo registro. Resultados experimentales preliminares de esta aproximación demuestran que es factible el desarrollo de valoraciones de diagnóstico automatizado de pacientes mediante identificadores de diagnóstico auscultatorios en forma temprana usando tecnologías de bajo costo.

This paper demonstrates classification processes of normal and abnormal bioacoustics signals recorded over a human thorax which encompasses heart and lung sounds. The specific aim is to design a signal classification system based on acoustical modeling techniques employing particularly HMM models to detect events' sequences, and GMM to model clusters corresponding to the data events. The modalities for extracting data characteristic are the MFCC and Octile vectors. These approaches have a potential of enhancing the classification accuracy of these auscultatory diagnostic indicators as the initial studies demonstrated that the HMM based models are less sensitive to the noise. Preliminary results demonstrate over 95% accuracy in classification of the evaluated sound signals. This is particularly critical taking into account environmental interference in a variety of medical care settings. As the heart sounds frequency components parallel those of the lungs sounds, but with a different periodicity, they can be modeled with the same recording. The preliminary experimental results are supportive of this approach and demonstrate feasibility of a development of an automated early diagnostic assessment of patients' auscultatory diagnostic indicators utilizing low cost technologies.

Comprehension of Lung Sounds for Family Medicine Residents in Daegu, Gyeongbuk / 가정의학회지

BACKGROUND: Understanding lung sound is important physical examination of chest. We studied to identify the state of the understanding, the education and the self-learning experience for lung sound's interpretation during the residentship training period. METHODS: Data were collected from 45 family medicine residents who had trained in Daegu and Gyeongbuk from April 1st to May 31th 2008. Residents completed self-administered questionnaire about the education and the self-learning experience for lung sound's interpretation and had been estimated the understanding of lung sound by using the audio file of five lung sounds. RESULTS: The median score (inter-quartile range, IQR) of lung sound's interpretation was 2 (1, 3) between 0 and 5 range. The number of residents who were educated for the interpretation of lung sounds during the training period were 10 (22.2%). Twenty-four (53.3%) had the self-learning experience for lung sounds. CONCLUSION: The understanding for lung sounds of family medicine residents in Daegu, Gyeungbuk was not high and the majority of residents were not educated the interpretation of lung sounds during the training period. Directors of family medicine should consider to make the regular education program for interpretation of lung sound during the training period.

Method for automatic detection of wheezing in lung sounds

The present report describes the development of a technique for automatic wheezing recognition in digitally recorded lung sounds. This method is based on the extraction and processing of spectral information from the respiratory cycle and the use of these data for user feedback and automatic recognition. The respiratory cycle is first pre-processed, in order to normalize its spectral information, and its spectrogram is then computed. After this procedure, the spectrogram image is processed by a two-dimensional convolution filter and a half-threshold in order to increase the contrast and isolate its highest amplitude components, respectively. Thus, in order to generate more compressed data to automatic recognition, the spectral projection from the processed spectrogram is computed and stored as an array. The higher magnitude values of the array and its respective spectral values are then located and used as inputs to a multi-layer perceptron artificial neural network, which results an automatic indication about the presence of wheezes. For validation of the methodology, lung sounds recorded from three different repositories were used. The results show that the proposed technique achieves 84.82% accuracy in the detection of wheezing for an isolated respiratory cycle and 92.86% accuracy for the detection of wheezes when detection is carried out using groups of respiratory cycles obtained from the same person. Also, the system presents the original recorded sound and the post-processed spectrogram image for the user to draw his own conclusions from the data.

Experience with the Lung-Sound Auscultation Simulator "Mr. Lung" / 医学教育

To evaluate the educational effects of a lung-sound auscultation simulator, “Mr. Lung”, we compared outcomes of fifth-year medical students in 2001 and 2002. From June 2001 through March 2002, we used “Mr. Lung” for small-group teaching for 100 fifth-year medical students. The following year, we opened our laboratory so that the fifth-year students could study with “Mr. Lung” by themselves for 1 month. “Mr. Lung” was then used for objective structured clinical examinations in May 2002. From June 2002 through April 2003, we used “Mr. Lung” again for small-group teaching for 91 fifth-year students. The class consisted of 90 minutes' training for the auscultation of lung sounds. At the beginning of class, auscultation tests were performed in which the students listened through their stethoscopes to 3 examples of abnormal lung sounds on “Mr. Lung” and answered questions about the location and quality of the sounds. The percentages of correct answers in 2001 and 2002, respectively, were 36.9% and 35.4% for differences between bilateral lung sounds, 52.5% and 55.8% for coarse crackles, 34.1% and 58.3%(p<0.05) for fine crackles, 69.2% and 70.8% for wheezes, 62.1% and 90.7%(p<0.01) for rhonchi, and 22.2% and 32.6% for stridor. In conclusion, 1 month's self-study with “Mr. Lung” to prepare for the objective structured clinical examinations improves auscultation skills.

Valorização do exame físico do tórax em prontuários médicos de clínica geral / Valorization of the thorax's physical examination in medical records of general practice

Os objetivos do presente trabalho são: (1) verificar como médicos residentes de um hospital de ensino empregam as várias etapas da semiotécnica respiratória; (2) comparar os termos usados para registros dos sons pulmonares com a classificação para uniformização terminológica proposta no Symposium of Lung Sounds (1985); (3) determinar a variabilidade de registros da ausculta pulmonar. A pesquisa foi observacional, com revisão de 712 prontuários de pacientes internados nas enfermarias de clínica médica do Hospital Universitário Lauro Wanderley da UFPB. O exame completo dos pulmões foi registrado em apenas 2,4% dos casos; a inspeção foi mencionada em 24%, a palpação em 9,9% e a percussão em 2,8% dos casos, enquanto que a ausculta foi registrada em 99,6% dos prontuários. Nenhum relatório apresentou referência à terminologia simplificada (sons contínuos/descontínuos) e alguns termos antigos foram mencionados (sopro tubário, broncofonia e atrito pleural). Houve concordância nos registros da ausculta em 76,3% dos casos. Conclui-se que: (1) o método de ausculta foi valorizado no exame do aparelho respiratório, porém as demais técnicas físicas foram, significativamente, omitidas; (2) a modificação na nomenclatura proposta pelo Symposium of Lung Sounds parece não ter sido incorporada à prática; (3) houve boa concordância quanto aos resultados observados à ausculta. Esses dados mostram que apenas a ausculta foi devidamente valorizada no exame clínico do tórax, havendo estabilidade razoável dos achados em dias consecutivos, mas as modificações terminológicas mais modernas não estão disseminadas na prática clínica.

The objectives of the present study are: (1) to verify how residents and interns of an university hospital use the physical diagnosis of the thorax (2) to compare the physical diagnosis register with classification for terminologic uniformization propused in the Symposium on Lung Sounds (1985); (3) to determine the variability of registers of the lung auscultation. This research was observational, with revision of the 712 relatories of internation of in-patient's on the medical clinic of the Lauro Wanderley Universitary Hospital (UFPB). The complete examination was registered in only 2,4% of the cases; the inspection was considered in 24%, palpation in 9,9% e percussion in 2,8% of the cases, while lung's auscultation was registered in 99,6% of records. No one medical registry presented referency to the simples terminology (continous/ descontinous sounds) and some names wasn't writed (bronquic breathing, broncophony and pleural atrict). There was concordance of auscultation's registers in 76,3% of the cases. The conclusions are: (1) the method of the auscultation was very used in exam of the physical diagnosis of the thotax, but the others clinical was significantly omited; (2) the modification in the nomenclature propused by the Symposium on Lung Sounds perhaps doesn't inclused to the practice; (3) there was a good concordance in the results observed at auscultation in consecutive days. This results shows that the lung's physical diagnosis is being invaluable in the clinical examination, with satisfatory reliabilith of dates in consecutive days, although moderns terminological modifications didn't spread in the clinical practice.