- In order to analyze respiratory sounds on a computer, we developed a cost-effective and easy-to-use Algorithm that can be used with any device.
- We employed two types of machine learning algorithms; Gammatone cepstrum coefficients features in a Convolutional Neural Network and Since using GTCC and STFC features with a CNN-LSTM algorithm.
- We prepared four data sets for CNN-LSTM algorithm to classify respiratory audio: (1) healthy versus pathological classification; (2) rale, rhonchus, and normal sound classification; (3) singular respiratory sound type classification; and (4) audio type classification with all sound types.
- In existing work, the architecture of some well-known MFCC features when fed to Inception Network classifier used to classify the audio scenes.
- Furthermore, it analyzed different methods of combining these features, and also of combining information from two channels when the data is in different format.
- More recently, with appropriate changes from designing Inception Network for image analysis to taking into account speech-specific properties, the Inception Network is also found effective for speech recognition
- Existing Inception Network -based analysis requires significantly High computation Time.
- Obviously classification based on these low-level(MFCC) features alone may not be accurate.
- High complexity algorithm will result in poor performance
- In this article, one of the major objectives is to provide an automated algorithmic approach that can categorize lung sounds in a variety of diseased states.
- Another objective of this present research work is to propose a lightweight deeplearning architecture that can classify lung sounds accurately while keeping parameter size and computational complexity less.
- The majority of these respiratory diseases have almost similar kind of symptoms; therefore, it becomes difficult for the doctor to predict the actual disease just by hearing the lung sound only and requires additional tests, such as spirometry test.
- The novel contributions of the proposed framework are itemized as follows.
- 1) Designing a novel lightweight LSTM, namely, CNN_LSTM, to classify the lung sound efficiently, while keeping the architecture lightweight in terms of total trainable parameters and model storage size.
- 2) Classification of seven respiratory diseases for the first time, utilizing three publicly available lung sound databases: ICBHI 2017 challenge database and chest wall lung sound database .Engaging all the databases also ensures the robustness of the classification mechanism, as the DLM is trained with a wide variety of lung sounds.
3) Computing the ablation study and classification report containing the statistics of layers, parameters, accuracy, precision, recall, F1 score, and so on, in order to have a thorough performance/classification accuracy analysis of the proposed lightweight CNN_LSTM.
- Front End – Anaconda IDE
- Backend – SQL
- Language – Python 3.8
- •Hard Disk: Greater than 500 GB
- •RAM: Greater than 4 GB
- •Processor: I3 and Above
- * Base Paper
- * Complete Source Code
- * Complete Documentation
- * Complete Presentation Slides
- * Flow Diagram
- * Database File
- * Screenshots
- * Execution Procedure
- * Readme File
- * Addons
- * Video Tutorials
- * Supporting Softwares
- * 24/7 Support * Ticketing System
- * Voice Conference
- * Video On Demand
- * Remote Connectivity
- * Code Customization
- * Document Customization
- * Live Chat Support
lung sound recognition algorithm based on vggish-bigru, Lung sound classification github, Vesicular breath sounds, Lung sounds auscultation, Respiratory sound classification, COPD dataset Kaggle, Respiratory disease dataset, Kaggle Respiratory Sound Database, COPD lung sounds wheeze, machine learning based lung sound classification, lung sound recognition, Machine Learning Projects 2023, artificial intelligence projects 2023 2024, asthma detection using machine learning, tuberculosis detection using machine learning