Lung Ultrasound Features and Risk Stratification in 80 COVID-19 Patients

From March 6 to April 3, 2020, this dataset includes lung ultrasonography (LUS) features and clinical outcomes for 80 PCR-confirmed COVID-19 patients from Lausanne University Hospital. It was created by Thomas Brahier and colleagues for a prospective observational cohort study. The data likely contains LUS scores and patterns used to stratify patients into self-resolving outpatient, hospitalized, and intubated/death groups.

Use Cases

• Predicting hospitalization risk in COVID-19 patients based on lung ultrasound score and pattern features.
• Training classifiers to discriminate between outpatient and severe outcome groups using LUS zone involvement percentages.
• Analyzing the correlation between specific LUS patterns (e.g., pleural line irregularities, B lines) and 7-day clinical outcomes.

Strengths

• Includes data from 80 patients with PCR-confirmed COVID-19.
• LUS images were reviewed by two experts blinded to patient outcome, suggesting a level of quality control.
• The LUS score demonstrated good discrimination (AUC 0.80) between outpatient and more severe groups.

Limitations

• Row count and column-level documentation are absent; field semantics must be inferred after download.
• Data is from a single hospital center over a one-month period, which may limit generalizability.
• Last update date is unknown; freshness unverified.

Provenance

Source

University of Lausanne / Lausanne University Hospital

Collection Method

Prospective observational study; LUS images recorded by a trained physician using a standardized protocol and reviewed by two experts.

Time Range

March 6, 2020 to April 3, 2020

Geography

Lausanne, Switzerland (Lausanne University Hospital Emergency Department)