Scientific Data / Article in Press
AIR-LEISH: A Dataset of Giemsa-Stained Microscopy Images for AI-based Leishmania amastigotes Detection
This paper introduces AIR-LEISH, a new dataset of 180 Giemsa-stained microscopy images of intracellular Leishmania amastigotes within experimentally infected macrophages. It includes expert annotations for 8,140 Leishmania amastigotes and 1,511 macrophages, designed to advance AI-based object detection and image segmentation for leishmaniases research and drug discovery. The dataset is publicly available on Zenodo, fostering collaborative initiatives.
Executive Impact: At a Glance
The AIR-LEISH dataset addresses a critical gap in leishmaniases research by providing a comprehensive, expert-annotated collection of microscopy images of intracellular Leishmania amastigotes. This resource enables the development of AI tools for rapid and accurate parasite detection, overcoming limitations of manual examination which is time-consuming and expertise-dependent. By facilitating automated analysis, AIR-LEISH accelerates drug discovery, improves diagnostic capabilities, and supports public health initiatives, particularly in low-resource settings.
Deep Analysis & Enterprise Applications
Select a topic to dive deeper, then explore the specific findings from the research, rebuilt as interactive, enterprise-focused modules.
Dataset Scale and Impact
8,140 Total Leishmania Amastigotes AnnotatedThis significant number of expert-annotated amastigotes, across 180 images, provides a robust foundation for training highly accurate AI models, addressing the critical lack of suitable public datasets for leishmaniases.
Robust Annotation and Validation Workflow
| Feature | U-Net (Segmentation) | YOLOv8 (Detection) |
|---|---|---|
| Amastigote Precision (Test Set) |
|
|
| Amastigote Recall (Test Set) |
|
|
| Overall Segmentation/Detection Capabilities |
|
|
| Key Strengths |
|
|
Accelerating Leishmaniases Research & Diagnostics
The AIR-LEISH dataset supports a wide range of applications, from advancing drug discovery and pathogenesis research to developing diagnostic-supporting tools. Its open availability on Zenodo facilitates computational tool development, accelerating image-based research and minimizing manual effort in low-resource settings.
~1.25x Increase in infection rates observed in Set2 compared to Set1, highlighting biological diversity studies possible with the dataset. The dataset's utility extends to other intracellular pathogens and transfer learning applications.
Calculate Your Potential AI-Driven ROI
Estimate the efficiency gains and cost savings your enterprise could achieve by automating key processes with AI, based on industry benchmarks and operational data.
Your AI Implementation Roadmap
A typical enterprise AI journey with Own Your AI follows a structured, efficient timeline designed for rapid value delivery and seamless integration.
Phase 1: Discovery & Strategy (2-4 Weeks)
In-depth analysis of current workflows, identification of high-impact AI opportunities, data readiness assessment, and tailored strategy development with clear KPIs.
Phase 2: Pilot & Proof of Concept (6-10 Weeks)
Rapid prototyping and development of a targeted AI solution for a specific use case, demonstrating tangible results and validating the chosen approach.
Phase 3: Scaled Development & Integration (10-16 Weeks)
Full-scale development of the AI solution, robust testing, seamless integration into existing enterprise systems, and comprehensive team training.
Phase 4: Optimization & Ongoing Support (Ongoing)
Continuous monitoring, performance optimization, model retraining, and dedicated support to ensure sustained value and adaptability to evolving business needs.
Ready to Transform Your Enterprise with AI?
Schedule a complimentary strategy session with our AI experts to explore how these insights can be tailored to your organization's unique challenges and opportunities.