The new SARS-CoV2 2019 has become a critical public health problem worldwide and is approaching 238,557,111 cases worldwide, according to the October 12, 2021 Johns Hopkins infographic.
Lung ultrasound may be an alternative for screening patients with or suspected of having COVID-19. A prospective study conducted at AP-HP evaluated the correlation between a pulmonary ultrasound severity score and the patient clinical outcome under 28 days.
The pulmonary ultrasound score consisted of scoring lesions in 8 thoracic zones from 0 to 3, defining a global severity score (GS) ranging from 0 to 24. This study concludes that the GS score is a simple and useful tool that can be used in the emergency department to predict an unfavorable evaluation of COVID-19 patients at 28 days.
In teams, you will need to use a Machine Learning approach on data from this study to construct a global predictor of the patient's clinical outcome. The GS score will be used as a baseline. Particular attention will be paid to the medical rationale behind this global indicator. All your work will be released under an open-source license and documented in a GitHub library.
The dataset at your disposal concerns the scoring of 8 loops of pulmonary ultrasound in 327 patients included in 3 hospitals of the AP-HP. These data, in tabular format, are characterized by different indicators.the winning team will benefit from a scientific publication support in a medical journal