Search

We present in this paper some of the existing machine learning and sound processing techniques involved in the medical process and show how they can be applied in the context of a vocal disease detection task. The machine learning techniques are the usual, ranging from decision trees to neural networks, and when it comes to the sound processing approaches some scientists probably are familiar with them from the speech processing or music analysis areas. However, the techniques are adapted to the particularities of the medical field -- data collection, verification, etc. Through the concrete example of the vocal disease detection task we worked on, we made some interesting observations, both on the classification ability of the model and on the influence of the data.

This paper addresses the critical issue of cardiovascular disease (CVD), the leading cause of global mortality, emphasizing the imperative for effective and early detection to mitigate CVD-related deaths. The research problem underscores the urgency of developing advanced diagnostic tools to identify heart abnormalities promptly. The primary objective is to create a Support Vector Machine (SVM) algorithm for accurate classification of different heart conditions, namely Normal heart, Mitral Stenosis, and Mitral Regurgitation. To achieve this objective, the study utilizes a dataset of heart sounds available online using a 10-fold cross-validation method. The focus is on evaluating the efficacy of various kernel functions within the SVM framework for heart sound classification. The findings demonstrate that the linear kernel exhibits superior accuracy and robustness in effectively classifying heart conditions. Notably, the proposed classification method attains an impressive 96% accuracy, highlighting its potential as a reliable tool for early detection of cardiovascular diseases. This research contributes to the ongoing efforts to enhance diagnostic capabilities and ultimately reduce the global burden of CVD-related fatalities.

The purpose of the article is to analyze the methods and means of processing cough sounds to detect lung diseases, as well as to describe the developed system for classifying and detecting cough sounds based on a deep neural network. Four types of machine learning and the use of convolutional neural network (CNN) are considered. Hypermarkets of CNN are given. Varieties of machine learning based on the CNN are discussed. The analysis of works on the methodology and means of processing cough sounds based on the CNN with the reduction of the means used and the accuracy of recognition is carried out. Details of machine learning using the environmental sound classification 50 (ESC-50) dataset are discussed. To recognize COVID-19 cough, a classifier was analyzed using CNN as a machine learning model. The proposed CNN system is designed to classify and detect cough sounds based on ESC-50. After selecting a set of sound classification data, four stages are described: extraction of features from audio files, labeling, training, testing. The ESC-50 used for the study was downloaded from the Kaggle website. Python libraries and modules related to deep learning and data science were used to implement the project: NumPy, Librosa, Matplotlib, Hickle, Sci-Kit Learn, Keras. The implemented network used a stochastic gradient algorithm. Several volunteers recorded their voices while coughing using their smartphones and it was assured to record their voices in a public environment to introduce noise to the sounds, in addition to some audio files that were downloaded online. The results showed an average accuracy of 85.37 %, precision of 78.8 % and a recall record of 91.9 %.

A child's success in school relies on their ability to quickly grasp language and reading skills, the foundations of which are acquired even before entering a formal classroom setting. Previous studies in preschoolers have begun to establish relationships linking beat synchronization, preliteracy skills, and auditory processing. Beat synchronization involves the integration of sensorimotor systems with auditory and cognitive circuits and, therefore calls on many of the same neural networks as language. Using a drumming task, we analyzed the relationship between children's ability to maintain an isochronous beat with preliteracy skills and frequency following responses (FFRs) in over 150 preschoolers. We show that preschoolers who performed well on the beat synchronization task outscored their peers on all preliteracy measures and had more robust FFRs. Furthermore, the good synchronizers experienced less degradation of certain FFR measures when listening in noise. Together, our results are consistent with the view that rhythm, preliteracy, and auditory processing are interconnected during early childhood.

Inhibition along the auditory pathway is crucial for processing of acoustic information. Within the auditory thalamus, a key region in the central auditory pathway, inhibition is provided by the thalamic reticular nucleus (TRN), comprised of two large classes of inhibitory neurons, parvalbumin (PV TRN ) and somatostatin (SST TRN ) positive. In the auditory cortex, PV and SST neurons differentially shape auditory processing. We found that the ventral MGB, the thalamic region in the direct ascending auditory pathway, receives inputs predominantly from PV TRN neurons, whereas SST TRN neurons project to the dorso-medial regions of MGB. Consistently, inactivating PV TRN neurons increased sound-evoked activity in over a third of neurons in the vMGB, with another large fraction of neurons being suppressed. By contrast, inactivating SST TRN neuronal activity largely reduced tone-evoked activity in vMGB neurons. Cell type-specific computational models revealed candidate circuit mechanisms for generating the bi-directional effects of TRN inactivation on MGB sound responses. These differential inhibitory pathways within the auditory thalamus suggest a cell-specific role for thalamic inhibition in auditory computation and behavior.

Objective: Cochlear implants (CIs) are auditory prostheses for individuals with severe to profound hearing loss, offering substantial but incomplete restoration of hearing function by stimulating the auditory nerve using electrodes. However, progress in CI performance and innovation has been constrained by the inability to rapidly test multiple sound processing strategies. Current research interfaces provided by major CI manufacturers have limitations in supporting a wide range of auditory experiments due to portability, programming difficulties, and the lack of direct comparison between sound processing algorithms. To address these limitations, we present the CompHEAR research platform, designed specifically for the Cochlear Implant Hackathon, enabling researchers to conduct diverse auditory experiments on a large scale., Study Design: Quasi-experimental., Setting: Virtual., Methods: CompHEAR is an open-source, user-friendly platform which offers flexibility and ease of customization, allowing researchers to set up a broad set of auditory experiments. CompHEAR employs a vocoder to simulate novel sound coding strategies for CIs. It facilitates even distribution of listening tasks among participants and delivers real-time metrics for evaluation. The software architecture underlies the platform's flexibility in experimental design and its wide range of applications in sound processing research., Results: Performance testing of the CompHEAR platform ensured that it could support at least 10,000 concurrent users. The CompHEAR platform was successfully implemented during the COVID-19 pandemic and enabled global collaboration for the CI Hackathon (www.cihackathon.com)., Conclusion: The CompHEAR platform is a useful research tool that permits comparing diverse signal processing strategies across a variety of auditory tasks with crowdsourced judging. Its versatility, scalability, and ease of use can enable further research with the goal of promoting advancements in cochlear implant performance and improved patient outcomes., Competing Interests: There are no conflicts of interest, financial, or otherwise.