PET/CT Feature Clusters for 289 Fever of Unknown Origin Patients

Sule Ceylan's dataset contains results from a retrospective cohort study of 289 adult patients with fever of unknown origin (FUO). The data includes consensus clustering results based on PET/CT imaging features, such as organ involvement and SUVmax, derived using four unsupervised algorithms. The dataset was last updated on 2026-05-29.

Use Cases

• Benchmark unsupervised clustering algorithms based on PET/CT feature patterns described in the study.
• Analyze relationships between imaging phenotypes (e.g., reticuloendothelial involvement) and clinical etiologies like infections or rheumatologic diseases.
• Study the concordance between data-driven imaging clusters and final clinical diagnoses using metrics like the Adjusted Rand Index.

Strengths

• Includes data from 289 patients with a defined mean age of 59 ± 17 years.
• Clustering results are based on a consensus of four different unsupervised algorithms.
• Specific imaging features are described, including involvement of reticuloendothelial organs, parenchymal organs, and large vessels.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count for the underlying feature data is unknown, which may limit suitability assessment.
• The dataset is very small at 14.7 KB, indicating limited scope, likely containing summary results rather than raw patient data.

Provenance

Source

Sule Ceylan via figshare.

Collection Method

Retrospective cohort study; data extracted from 18F-FDG PET/CT scans of adult FUO patients.

Freshness

Last updated 2026-05-29 04:50:49; freshness should be verified.