CNN Networks to Classify Cardiopulmonary Signals

At the moment, the world lives in a pandemic situation of COVID-19 and related variants, driving urgent needs for expanded assessments. A complementary support of related healthcare can be based on an intelligent system that can diagnose early onset of respiratory disorders. The convolutional neural networks (CNN) were implemented utilizing image data, reflecting bidimensional signals. Specifically, CNN has shown to be powerful tool in the context of cardiopulmonary sounds evaluation. The configurations of CNN contain convolutional layers to extract feature maps and fully connected layers to classify indicators of interest. Even though, learning algorithms use parameters like learning rate which can determine and attain CNN configuration less complex, with excellent results as reflected in the experiments we carried out, and which focused on achieved configuration of CNN with excellent results classifying heart sounds (HS) and lung sounds (LS).

Lightweight Skip Connections With Efficient Feature Stacking for Respiratory Sound Classification

As the number of deaths from respiratory diseases due to COVID-19 and infectious diseases increases, early diagnosis is necessary. In general, the diagnosis of diseases is based on imaging devices (e.g., computed tomography and magnetic resonance imaging) as well as the patient's underlying disease information. However, these examinations are time-consuming, incur considerable costs, and in a situation like the ongoing pandemic, face-to-face examinations are difficult to conduct. Therefore, we propose a lung disease classification model based on deep learning using non-contact auscultation. In this study, two respiratory specialists collected normal respiratory sounds and five types of abnormal sounds associated with lung disease, including those associated with four lung lesions in the left and right anterior chest and left and right posterior chest. For preprocessing and feature extraction, the noise was removed using three pass filters (low, band, and high), and respiratory sound features were extracted using the Log-Mel Spectrogram-Mel Frequency Cepstral Coefficient followed by feature stacking. Then, we propose a lung disease classification model of dense lightweight convolutional neural network-bidirectional gated recurrent unit skip connections using depthwise separable convolution based on the extracted respiratory sound information. The performance of the classification model was compared with both the baseline and the lightweight models. The results indicate that the proposed model achieves high performance and has an accuracy of 92.3%, sensitivity of 92.1%, specificity of 98.5%, and f1-score of 91.9%. Using the proposed model, we aim to contribute to the early detection of diseases during the COVID-19 pandemic.

Employing deep learning for lung sounds classification

Respiratory diseases indicate severe medical problems. They cause death for more than three million people annually according to the World Health Organization (WHO). Recently, with coronavirus disease 19 (COVID-19) spreading the situation has become extremely serious. Thus, early detection of infected people is very vital in limiting the spread of respiratory diseases and COVID-19. In this paper, we have examined two different models using convolution neural networks. Firstly, we proposed and build a convolution neural network (CNN) model from scratch for classification the lung breath sounds. Secondly, we employed transfer learning using the pre-trained network AlexNet applying on the similar dataset. Our proposed model achieved an accuracy of 0.91 whereas the transfer learning model performing much better with an accuracy of 0.94. © 2022 Institute of Advanced Engineering and Science. All rights reserved.

A System for Remote Monitoring of Patient Body Parameters

In today's world, life without technology is not possible. Continuous advancement in patient health monitoring techniques, medical equipment's or machines and other enhancing technologies is ongoing as per recent trends in specifically healthcare sector in order to reduce human efforts. Taking into consideration the serious nature of the above aforementioned problem, it is necessary to make some major improvements in the communication devices and systems with application-based technology in order to enhance their performance thereby saving medical costs and achieve other major advantages. The principal objective of this paper is to provide a system for remote and secure monitoring of healthcare information of patient suffering from virus and utilizing a mobile device as per the patient requirements. In this paper, a proposed model measure the temperature of the body, respiratory system especially lung sound and breathing activity, which are the main source of symptoms to understand the actual health condition of a person. And data sensed by the IoT sensor device used for measuring the real-time data body temperature, lung sounds, respiratory data, pulse rate and heartbeat. © 2021 IEEE.

Performance Evaluation of an Intelligent Lung Sound Classifier Based on an Enhanced MFCC Model

The rate at which technology grew in the past years is unbelievably fast and astounding. However, chronic illnesses like respiratory diseases remains a common and widely experienced problem globally. The emergence of infectious respiratory health issues such as the coronavirus (COVID-19) had only made this enigma more harmful, causing an increase in the number of death due to respiratory illnesses. Hence, the development of modern and accurate methods to improve medical diagnosis is one of the simple step's humans can perform to overcome such problems. In this study, the researchers proposed an enhanced model for lung sound classification using Mel Frequency Cepstral Coefficient (MFCC). The design will classify four different lung sounds, with data input taken and classified one at a time. The goal of which is to augment human intelligence and not to replace the existing lung sound classification methods. The pre-recorded lung sounds were characterized, and the researcher proposed four enhanced MFCC models with three varying designs. The data collected from feature extraction and data mining were evaluated by the machine learning algorithms Support Vector Machine (SVM) and K-Nearest Neighbor (KNN). Measures like sensitivity, specificity, and accuracy were tested to determine which model was superior. Results showed that in terms of performance metrics, KNN performed better than SVM in classifying lung sounds. Tested in three designs where the pre-emphasis was removed, and the original 44.1kHz data resampled. Model 3 using KNN sampled at a frequency of 12000Hz has reached an average accuracy of 96.92% and a blind-data accuracy of 93.33%. A specificity of 97.94% and a sensitivity of 93.83%, achieving a performance that is comparable with existing studies on lung sound classification. © 2021 IEEE.

CoCross: An ICT Platform Enabling Monitoring Recording and Fusion of Clinical Information Chest Sounds and Imaging of COVID-19 ICU Patients.

Monitoring and treatment of severely ill COVID-19 patients in the ICU poses many challenges. The effort to understand the pathophysiology and progress of the disease requires high-quality annotated multi-parameter databases. We present CoCross, a platform that enables the monitoring and fusion of clinical information from in-ICU COVID-19 patients into an annotated database. CoCross consists of three components: (1) The CoCross4Pros native android application, a modular application, managing the interaction with portable medical devices, (2) the cloud-based data management services built-upon HL7 FHIR and ontologies, (3) the web-based application for intensivists, providing real-time review and analytics of the acquired measurements and auscultations. The platform has been successfully deployed since June 2020 in two ICUs in Greece resulting in a dynamic unified annotated database integrating clinical information with chest sounds and diagnostic imaging. Until today multisource data from 176 ICU patients were acquired and imported in the CoCross database, corresponding to a five-day average monitoring period including a dataset with 3477 distinct auscultations. The platform is well accepted and positively rated by the users regarding the overall experience.

VECTOR: An algorithm for the detection of COVID-19 pneumonia from velcro-like lung sounds.

The coronavirus disease 2019 (COVID-19) has severely stressed the sanitary systems of all countries in the world. One of the main issues that physicians are called to tackle is represented by the monitoring of pauci-symptomatic COVID-19 patients at home and, generally speaking, everyone the access to the hospital might or should be severely reduced. Indeed, the early detection of interstitial pneumonia is particularly relevant for the survival of these patients. Recent studies on rheumatoid arthritis and interstitial lung diseases have shown that pathological pulmonary sounds can be automatically detected by suitably developed algorithms. The scope of this preliminary work consists of proving that the pathological lung sounds evidenced in patients affected by COVID-19 pneumonia can be automatically detected as well by the same class of algorithms. In particular the software VECTOR, suitably devised for interstitial lung diseases, has been employed to process the lung sounds of 28 patient recorded in the emergency room at the university hospital of Modena (Italy) during December 2020. The performance of VECTOR has been compared with diagnostic techniques based on imaging, namely lung ultrasound, chest X-ray and high resolution computed tomography, which have been assumed as ground truth. The results have evidenced a surprising overall diagnostic accuracy of 75% even if the staff of the emergency room has not been suitably trained for lung auscultation and the parameters of the software have not been optimized to detect interstitial pneumonia. These results pave the way to a new approach for monitoring the pulmonary implication in pauci-symptomatic COVID-19 patients.

Identification of Soft Tissue-Mimicking Materials and Application in the Characterization of Sensors for Lung Sounds

Early diagnosis of pulmonary implications is fundamental for the treatment of several diseases, such as idiopathic pulmonary fibrosis, rheumatoid arthritis, connective tissue diseases and interstitial pneumonia secondary to COVID-19 among the many. Recent studies prove that a wide class of pulmonary diseases can be early detected by auscultation and suitably developed algorithms for the analysis of lung sounds. Indeed, the technical characteristics of sensors have an impact on the quality of the acquired lung sounds. The availability of a fair and quantitative approach to sensors’comparison is a prerequisite for the development of new diagnostic tools. In this work the problem of a fair comparison between sensors for lung sounds is decoupled into two steps. The first part of this study is devoted to the identification of a synthetic material capable of mimicking the acoustic behavior of human soft tissues;this material is then adopted as a reference. In the second part, the standard skin is exploited to quantitatively compare several types of sensors in terms of noise floor and sensitivity. The proposed methodology leads to reproducible results and allows to consider sensors of different nature, e.g. laryngophone, electret microphone, digital MEMS microphone, mechanical phonendoscope and electronic phonendoscope. Finally, the experimental results are interpreted under the new perspective of equivalent sensitivity and some important guidelines for the design of new sensors are provided. These guidelines could represent the starting point for improving the devices for acquisition of lung sounds. IEEE