What is the MultiOrg Dataset?

It is a high-resolution lung organoid microscopy image dataset with over 400 images and 60,000 annotated organoids, designed for AI and medical research applications.

How can this dataset be used in AI and machine learning?

The dataset can train deep learning models for organoid detection, segmentation, and classification tasks in biomedical research and precision medicine.

Does the dataset include diverse organoid samples?

Yes. The dataset includes a variety of lung organoid images to ensure model robustness across different conditions and experimental setups.

What annotation and delivery formats are provided?

Annotations are provided in JSON format along with the corresponding high-resolution JPEG images, validated through GTS's quality assurance process.

Is the dataset compliant with privacy and industry standards?

Yes. All dataset creation and delivery comply with GDPR and ISO 9001:2015 / ISO 27001:2013 standards, ensuring secure and reliable data.

MultiOrg Dataset

Home » Dataset Download » MultiOrg Dataset

MultiOrg Dataset

Datasets

MultiOrg Dataset

File

MultiOrg Dataset

Use Case

MultiOrg Dataset

Description

Discover the MultiOrg dataset, a comprehensive lung organoid microscopy image collection with over 400 high-resolution images and 60,000 expertly annotated organoids.

Description:

The MultiOrg dataset is a comprehensive lung organoid 2D microscopy image collection designed for benchmarking object detection methods and studying uncertainty estimation. It includes over 400 images of entire microscopy plate wells and more than 60,000 annotated organoids. These organoids originate from diverse biological experimental setups, featuring two different types of organoids grown under varying conditions. Annotations were made by two expert annotators using bounding boxes.

Download Dataset

Key Features:

Large Scale: Over 400 high-resolution microscopy images and 60,000 annotated organoids.
Expert Annotations: Two expert annotators provided bounding box annotations, ensuring high-quality labeling.
Biological Diversity: Organoids stem from different biological experiments, enhancing the dataset’s robustness.
Triple Label Sets: Three unique label sets for test images, derived from the two annotators at different time points, allow for the quantification of label noise.
Applications: Useful for multi-rater benchmarking, studying label noise, and improving object detection methods in microscopy images.

This dataset is sourced from Kaggle.

Contact Us

Let's Discuss your Data collection Requirement With Us

To get a detailed estimation of requirements please reach us.

MultiOrg Dataset