PavicNet-MC: A Multilabel Classification Model Applied to Lung Ultrasound

In recent years, the consequences of Covid-19 and other lung diseases have caused an increase in demand for health services, early and accurate diagnosis of these diseases is essential for the recovery of patients. This paper proposed a multilabel classification model, called PavicNet-MC. This model was developed with the aim of identifying five artifacts visible on lung ultrasound. The proposed model obtained an accuracy of 99% in all five classified artifacts. The results showed the proposed model is highly effective in detecting the visible artifacts that correlate with lung diseases, and has a relatively low complexity compared to other architectures found in the literature.