Pulse Lung Insight Engine: A Data-Driven AI Framework for Early Cardiopulmonary Condition Detection

Abstract

Cardio-respiratory diseases remain a major global health challenge, accounting for a significant proportion of mortality, particularly in India where they contribute to nearly 30% of total deaths. Accurate and timely diagnosis is critical; however, traditional auscultation using acoustic stethoscopes relies heavily on physician expertise, often leading to subjective and inconsistent interpretations. To address this limitation, this study proposes an automated cardio-respiratory sound classification system that enhances diagnostic reliability through machine learning and deep learning techniques. A novel dataset was developed using a digital stethoscope to capture high-quality heart and lung sounds from a clinical manikin. This dataset uniquely includes both individual and combined (mixed) cardiorespiratory recordings, enabling dual classification tasks. The system incorporates a user-friendly graphical interface with role-based access, allowing administrators to train models and users to perform predictions efficiently. Audio signals are processed using Librosa for feature extraction, including MelFrequency Cepstral Coefficients (MFCC), Chroma features, and Mel spectrograms. Multiple traditional machine learning models Quadratic Discriminant Analysis (QDA), Gradient Boosting Classifier (GBC), Naive Bayes Classifier (NBC), and Logistic Regression Classifier (LRC) are implemented and evaluated. In addition, a hybrid deep learning architecture combining a Bi-directional Convolutional Neural Network (BiCNN) and a Bi-directional Gated Recurrent Unit (BiGRU) is proposed to capture both spatial and temporal characteristics of audio signals. The system supports simultaneous classification of heart and lung sounds from mixed recordings and achieves strong performance across metrics such as accuracy, precision, recall, F1-score, and ROC-AUC, demonstrating its potential for reliable, non-invasive early diagnosis.