HQF-CC: hybrid framework for automated respiratory disease detection based on quantum feature extractor and custom classifier model using chest X-rays

In recent years, humans have been affected by a variety of respiratory diseases, such as dry cough, fever, pneumonia, and COVID-19. Respiratory diseases may severely damage the respiratory system of humans. The early diagnosis of it can help enable appropriate treatment immediately to reduce the abnormalities in the healthcare monitoring systems. A chest X-ray could be used as one of the medical tools to assess the severity of respiratory diseases. In order to predict or classify respiratory diseases, many researchers have proposed different deep learning-based models. But they consume more computational resources. This article proposes a hybrid framework for respiratory disease detection based on a quantum feature extractor and custom classifier model (HQF-CC) from Chest X-rays. Here, we propose a quantum machine learning algorithm named MMS (Multi-Multi-Single) for feature extraction and a custom classifier for classification. The experiments were performed on the COVID-19 Radiography Dataset (CRD), which includes 15,153 Chest X-ray images of COVID-19, viral, and Normal diseases, respectively. The proposed model had the highest training and testing accuracy of 97.2% and 98.8% on the CRD dataset and a training and testing loss of 0.02, 0.01. Upon various experiments, the proposed model has proven to be more accurate, portable, and memory-efficient than other deep and machine learning models for respiratory disease detection.