ABSTRACT
In recent years, a variety of deep learning techniques and methods have been adopted to provide AI solutions to issues within the medical field, with one specific area being audio-based classification of medical datasets. This research aims to create a novel deep learning architecture for this purpose, with a variety of different layer structures implemented for undertaking audio classification. Specifically, bidirectional Long Short-Term Memory (BiLSTM) and Gated Recurrent Units (GRU) networks in conjunction with an attention mechanism, are implemented in this research for chronic and non-chronic lung disease and COVID-19 diagnosis. We employ two audio datasets, i.e. the Respiratory Sound and the Coswara datasets, to evaluate the proposed model architectures pertaining to lung disease classification. The Respiratory Sound Database contains audio data with respect to lung conditions such as Chronic Obstructive Pulmonary Disease (COPD) and asthma, while the Coswara dataset contains coughing audio samples associated with COVID-19. After a comprehensive evaluation and experimentation process, as the most performant architecture, the proposed attention BiLSTM network (A-BiLSTM) achieves accuracy rates of 96.2% and 96.8% for the Respiratory Sound and the Coswara datasets, respectively. Our research indicates that the implementation of the BiLSTM and attention mechanism was effective in improving performance for undertaking audio classification with respect to various lung condition diagnoses.
ABSTRACT
Purpose: Online learning is essential in today’s world. The COVID-19 has resulted in shutting down all the universities across the globe. Countries like India and Turkey (lower-income countries) are suffering a lot in giving the best classroom practice to their students through online mode. The entire way of teaching-learning has changed drastically, and it is a need of an hour. Research suggests that online learning has been shown to increase retention of information, and take less time, meaning the changes coronavirus have caused might be here to stay. It is therefore important to understand from student’s perspectives about learning online. The paper systematically surveys the perception of learning online for Indian and Turkan students. Design/methodology/approach: To achieve this goal, 594 samples of students (from India and Turkey country) have been taken into considerations, and through statistical measures, the results were analyzed. The set of four research questions comprising of effect of study on COVID-19 pandemic, perception of learning online in COVID-19 pandemic, perception of different genders in learning online and perception of Indians over Turkan students in learning online were analyzed through statistical measures such as mean, standard deviation and so on. Findings: The descriptive statistics of various responses across various dimensions (gender, country) reveals that there is no effect in learning online as compared to classroom-based teaching. On the other hand, there is no significant difference in gender and country in learning online. Originality/value: Online learning has become crucial in higher education as far as pandemic situation is concerned. Many higher education institutions across different countries are suffering various problems from student point of view. Middle-income countries who are with limited assets and less advancements in higher education need to adhere to certain guidelines in online learning. This empirical study will help to understand the perception of students in online learning across India and Turkey. © 2021, Emerald Publishing Limited.