Differential Diagnosis of Lung Disease through Quantitative Lung Ultrasound Spectroscopy, an in vivo clinical study over 114 patients

Lung ultrasound (LUS) represents an important imaging tool to evaluate the state of the lung surface. However, it is mainly based on the visual interpretation of imaging artifacts, especially the vertical ones. This subjective interpretation leads to poorly reproducible approaches, and, thus, qualitative diagnoses. Therefore, in this study, we present a quantitative multifrequency analysis based on the parameterization of vertical artifacts. Specifically, radiofrequency (RF) data were acquired from 114 patients affected by different diseases using a linear and a convex probe, connected to the ULtrasound Advanced Open Platform (ULA-OP). The vertical artifacts observed in patients affected by pneumonia, cardiogenic pulmonary edema (CPE), and pulmonary fibrosis (PF) were extracted and analyzed. Specifically, the significance of three features (native frequency, bandwidth, and total intensity) in discriminating these diseases was evaluated. The results show how the total intensity seems si...

Lung ultrasound (LUS) represents an important imaging tool to evaluate the state of the lung surface. However, it is mainly based on the visual interpretation of imaging artifacts, especially the vertical ones. This subjective interpretation leads to poorly reproducible approaches, and, thus, qualitative diagnoses. Therefore, in this study, we present a quantitative multifrequency analysis based on the parameterization of vertical artifacts. Specifically, radiofrequency (RF) data were acquired from 114 patients affected by different diseases using a linear and a convex probe, connected to the ULtrasound Advanced Open Platform (ULA-OP). The vertical artifacts observed in patients affected by pneumonia, cardiogenic pulmonary edema (CPE), and pulmonary fibrosis (PF) were extracted and analyzed. Specifically, the significance of three features (native frequency, bandwidth, and total intensity) in discriminating these diseases was evaluated. The results show how the total intensity seems significant in differentiating PF from the other two diseases (p-value<0.05).