AuD-Former: A Hierarchical Transformer Network for Multimodal Audio-Based Disease Prediction

Audio-based disease prediction is emerging as a promising supplement to traditional medical diagnosis methods, facilitating early, convenient, and non-invasive disease detection and prevention. Multimodal fusion, which integrates features from various domains within or across bio-acoustic modalities, has proven effective in enhancing diagnostic performance. However, most existing methods in the field employ unilateral fusion strategies that focus solely on either intra-modal or inter-modal fusion. This approach limits the full exploitation of the complementary nature of diverse acoustic feature domains and bio-acoustic modalities. Additionally, the inadequate and isolated exploration of latent dependencies within modality-specific and modality-shared spaces curtails their capacity to manage the inherent heterogeneity in multimodal data. To fill these gaps, we propose AuD-Former, a hierarchical transformer network designed for general multimodal audio-based disease prediction. Specifically, we seamlessly integrate intra-modal and inter-modal fusion in a hierarchical manner and proficiently encode the necessary intra-modal and inter-modal complementary correlations, respectively. Comprehensive experiments demonstrate that AuD-Former achieves state-of-the-art performance in predicting three diseases: COVID-19, Parkinson's disease, and pathological dysarthria, showcasing its promising potential in a broad context of audio-based disease prediction tasks. Additionally, extensive ablation studies and qualitative analyses highlight the significant benefits of each main component within our model.

Discovering COVID-19 Coughing and Breathing Patterns from Unlabeled Data Using Contrastive Learning with Varying Pre-Training Domains

Rapid discovery of new diseases, such as COVID-19 can enable a timely epidemic response, preventing the large-scale spread and protecting public health. However, limited research efforts have been taken on this problem. In this paper, we propose a contrastive learning-based modeling approach for COVID-19 coughing and breathing pattern discovery from non-COVID coughs. To validate our models, extensive experiments have been conducted using four large audio datasets and one image dataset. We further explore the effects of different factors, such as domain relevance and augmentation order on the pre-trained models. Our results show that the proposed model can effectively distinguish COVID-19 coughing and breathing from unlabeled data and labeled non-COVID coughs with an accuracy of up to 0.81 and 0.86, respectively. Findings from this work will guide future research to detect an outbreak of a new disease early.