COVID-19 Related Pneumonia Detection in Lung Ultrasound
| Autor: | Michael Stiven Ramirez Campos, Sandra Liliana Cancino Suárez, Santiago Saavedra Bautista, Juan Manuel López López, Jose Vicente Alzate Guerrero |
|---|---|
| Rok vydání: | 2021 |
| Předmět: | |
| Zdroj: | Lecture Notes in Computer Science ISBN: 9783030770037 MCPR |
| Popis: | Accurate diagnosis plays an important role in the current public health situation caused by the Covid-19 outbreak. Ultrasound images offer some advantages over other imaging techniques due to their lowers costs; however, to the authors’ knowledge, these type of images have not received as much attention as the other methods. This article describes a set of novel features for Covid-19 detection from lung ultrasound scans, obtained from the Pocovid database described in [3]. Two simultaneous approaches were considered: analysis and segmentation of the pleura, and highlighting of information from frame sequences through PCA and ICA. The proposed features were tested using machine learning models, achieving an average accuracy of 0.9, which considering the interpretability of the features and the complexity of the classification models used, is a good result. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|