Terahertz Time-Domain Imaging of Diabetic and Non-Diabetic Patient Feet

A 11.6 GB dataset contains Terahertz Time-Domain imaging of the human plantar surface from type-2 diabetic and non-diabetic volunteers. It comprises raw terahertz signals and participant metadata, including demographic and clinical information, acquired in reflection geometry via raster scanning. The data were collected by Goretti G. Hernandez-Cardoso as part of a previously published clinical study.

Use Cases

• Develop classification models for diabetic foot risk based on terahertz waveform features mentioned in the description
• Analyze correlations between terahertz signal properties and clinical metadata such as demographic information
• Train anomaly detection algorithms for early signs of foot pathology using spatially resolved raster scan data
• Validate non-invasive imaging techniques for plantar tissue assessment against clinical study outcomes

Strengths

• Dataset size is 11.6 GB, indicating a substantial volume of imaging data.
• Includes participant metadata with demographic and clinical information, enabling multi-modal analysis.
• Data were acquired as part of a published clinical study, suggesting a structured collection process.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Goretti G. Hernandez-Cardoso

Collection Method

Measurements were collected in reflection geometry using a fiber-coupled terahertz system and include spatially resolved terahertz waveforms obtained through raster scanning of both feet soles.

Freshness

Last updated 2026-06-01 23:26:01; freshness should be verified.