Longitudinal CT Scans of 54 Patients with Liver Metastases

A retrospective longitudinal collection of contrast-enhanced abdominal CT scans from 54 patients with liver metastases who underwent systemic chemotherapy. The dataset includes preprocessed CT volumes and expert-validated segmentation masks for the liver and metastases, assembled by author valipour, zahra and hosted on Harvard Dataverse. Imaging acquisition dates and scan intervals are provided in metadata to enable precise longitudinal timing analysis.

Use Cases

• Developing liver segmentation algorithms based on expert-validated liver masks
• Evaluating deformable image registration methods using longitudinal CT volumes
• Tracking tumor progression over time based on metastasis segmentation masks and scan intervals
• Training models for automated detection of liver metastases in contrast-enhanced CT scans

Strengths

• Includes 54 patient cases, each with a full CT volume and two expert-validated segmentation masks
• Provides longitudinal timing data with acquisition dates and scan intervals for precise analysis
• Data is preprocessed and stored in NIfTI format, a standard for medical imaging

Limitations

• Column-level documentation is absent; field semantics must be inferred after download
• Row count is unknown, which may limit suitability assessment
• Data may reflect temporal or institutional bias inherent to the retrospective collection from a single study

Provenance

Source

Harvard Dataverse

Collection Method

Retrospectively assembled from patient records

Freshness

Last updated 2026-06-18 03:52:02; freshness should be verified