MODUS: Large-Scale Multi-Magnification Microscopic Dataset for Urine Sediment Detection

MODUS is a large-scale microscopic image dataset for clinical urine sediment object detection, containing 3,181 high-resolution images with 63,047 manually verified annotations. It covers a fine-grained taxonomy of 22 clinical categories, such as cells, casts, and crystals, and incorporates two standard optical magnifications: 10x and 40x. The dataset was authored by Yunqi Zhu and last updated on May 9, 2026.

Use Cases

• Train object detection models for urine sediment components based on the 22 clinical categories.
• Benchmark multi-scale image analysis algorithms using the 10x and 40x magnification images.
• Develop automated clinical decision support tools for urinalysis based on the annotated microscopic images.

Strengths

• Large scale with 3,181 high-resolution images.
• Extensive manual verification with 63,047 annotations.
• Fine-grained taxonomy covering 22 distinct clinical categories.
• Includes multi-magnification data with 10x and 40x optical views.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Freshness should be verified as the last update was in 2026.

Provenance

Source

figshare

Collection Method

Manually annotated microscopic images.

Freshness

Last updated 2026-05-09 12:39:53