Callisto Offers Free Datasets of X-rays for Suspected Lung Cancer and MRIs for Prostate Cancer with Ground Truth for AI and Clinical Research (Commercial Use Permitted)
To commemorate the 2nd anniversary of its "Callisto DataHub" medical image data platform, Callisto Co., Ltd. is distributing free datasets of 50 cases each of suspected lung cancer X-rays and prostate cancer MRIs with ground truth annotations by specialists. These datasets, sourced from Japanese medical institutions, include written findings and are available for commercial use. Callisto DataHub provides anonymized medical data to companies and academic institutions in the US, Europe, South Korea, and Japan.
📋 Article Processing Timeline
- 📰 Published: March 26, 2026 at 22:33
- 🔍 Collected: March 28, 2026 at 21:59 (47h 26m after Published)
- 🤖 AI Analyzed: April 14, 2026 at 22:29 (408h 29m after Collected)
Callisto Co., Ltd. (Headquarters: Shinjuku-ku, Tokyo, CEO: Charis Han), is commemorating the 2nd anniversary of its medical image data platform, Callisto DataHub, by distributing free datasets of 50 cases each of suspected lung cancer X-rays and prostate cancer MRIs with ground truth annotations from specialists. Both datasets originate from Japanese medical facilities, come with written findings, and are available for commercial use. Callisto DataHub sells anonymized radiological/pathological images, clinical information, and molecular data collected from Japanese medical facilities. It is already being used by more than 15 medical AI companies, medical device manufacturers, pharmaceutical companies, and academic institutions in the United States, Europe, South Korea, and Japan (data provided includes: X-ray, mammography, CT, MRI, PET, WSI, etc.).
### Features of Datasets Provided by Callisto DataHub
1. **Japanese Data**
As the data originates from Japanese medical facilities, it is suitable for research and development within Japan (including for pharmaceutical approval applications and marketing) and for R&D targeting Asian populations.
2. **Anonymized**
You can use properly anonymized datasets that have undergone procedures considerate of ethical and legal aspects.
3. **Pre-selected**
Cases that do not meet requirements (e.g., cases with severe noise or complications) have been excluded through image verification, allowing for purpose-driven research and development.
4. **Standardized**
Image data, clinical information including findings, and molecular data are all standardized, reducing the burden of preprocessing and verification work.
5. **Annotated**
If necessary, annotations by specialists can also be performed, eliminating the need for separate annotation work. For major modalities and diseases, datasets with ground truth may be prepared in advance.
6. **Rapid Delivery**
Datasets can typically be provided in about one month, significantly reducing the time required for data preparation. Depending on the data volume and annotation content, it may take about two months.
### Types of Datasets Provided by Callisto DataHub
1. **Radiological Images**
X-ray (e.g., lung nodules, interstitial shadows), mammography (e.g., breast cancer, normal breast), CT (e.g., lung nodules, cerebral infarction), MRI (e.g., brain tumor, breast cancer), PET, DXA, fluoroscopy/angiography, nuclear medicine, etc.
2. **Pathology Images**
WSI (e.g., lung cancer, interstitial lung...)
### Features of Datasets Provided by Callisto DataHub
1. **Japanese Data**
As the data originates from Japanese medical facilities, it is suitable for research and development within Japan (including for pharmaceutical approval applications and marketing) and for R&D targeting Asian populations.
2. **Anonymized**
You can use properly anonymized datasets that have undergone procedures considerate of ethical and legal aspects.
3. **Pre-selected**
Cases that do not meet requirements (e.g., cases with severe noise or complications) have been excluded through image verification, allowing for purpose-driven research and development.
4. **Standardized**
Image data, clinical information including findings, and molecular data are all standardized, reducing the burden of preprocessing and verification work.
5. **Annotated**
If necessary, annotations by specialists can also be performed, eliminating the need for separate annotation work. For major modalities and diseases, datasets with ground truth may be prepared in advance.
6. **Rapid Delivery**
Datasets can typically be provided in about one month, significantly reducing the time required for data preparation. Depending on the data volume and annotation content, it may take about two months.
### Types of Datasets Provided by Callisto DataHub
1. **Radiological Images**
X-ray (e.g., lung nodules, interstitial shadows), mammography (e.g., breast cancer, normal breast), CT (e.g., lung nodules, cerebral infarction), MRI (e.g., brain tumor, breast cancer), PET, DXA, fluoroscopy/angiography, nuclear medicine, etc.
2. **Pathology Images**
WSI (e.g., lung cancer, interstitial lung...)