Tryp V2/V3: Microscopy Images of Thick Blood Smears for Trypanosome Detection

An extended dataset of microscopy images of unstained thick blood smears, with bounding-box annotations for detecting Trypanosoma brucei brucei and Trypanosoma cruzi. It builds on the original 2023 Tryp release with two new versions and three out-of-distribution evaluation pools. The dataset is 8.9 GB in size and was last updated in May 2026.

Use Cases

• Training object detection models for Trypanosoma brucei brucei based on bounding-box annotations in thick blood smear images.
• Evaluating model robustness using out-of-distribution evaluation pools containing Plasmodium and tuberculosis images.
• Benchmarking detection algorithms for Trypanosoma cruzi based on incorporated images from a separate study.
• Comparing model performance across different dataset versions (V1, V2, V3) to assess incremental improvements.

Strengths

• Includes three distinct dataset versions (V1, V2, V3) for flexible training setups.
• Provides three out-of-distribution evaluation pools (Plasmodium, tuberculosis) for robustness testing.
• Original V1 data is preserved unchanged to maintain compatibility with prior citations and code.
• Dataset size is 8.9 GB, indicating a substantial collection of image data.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last updated 2026-05-25 13:40:57; freshness should be verified.

Provenance

Source

figshare

Collection Method

Curated from multiple research publications, including Anzaku et al. (2023), Morais et al. (2022), and Karlen et al. (2018).

Freshness

Last updated 2026-05-25 13:40:57.