---
title: "CT Scans of Brain"
description: "This brain CT dataset comprises over 70,000 DICOM studies with labeled pathologies such as intracerebral hemorrhage, ischemic stroke, and vessel abnormalities, providing segmentation data and…"
url: "https://unidata.pro/datasets/ct-scans-of-brain/"
date_modified: "2025-12-09T11:01:50+03:00"
language: "en-US"
---
This brain CT dataset comprises over 70,000 DICOM studies with labeled pathologies such as intracerebral hemorrhage, ischemic stroke, and vessel abnormalities, providing segmentation data and annotations to support medical research, computer vision, and deep learning–based diagnostic algorithms.

## Dataset Structure

### The Numbers Section

**Numbered list:**

- **Number:** 20,000+ — **Text:** studies with protocol
- **Number:** 50,000+ — **Text:** studies without protocol
- **Number:** 5 — **Text:** pathologies

### Tooltips Section

**Tooltip items:**

- **Name:** Medicine
- **Name:** Computer vision
- **Name:** Machine Learning
- **Name:** Segmentation
- **Name:** Classification

### Dataset Information

**Table with data:**

| Characteristic | Data |
| --- | --- |
| Description | Brain CT scans with or without protocols |
| Data types | DiCOM |
| Markup | Segmentation of pathologies |
| Tasks | Pathology recognition, computer vision. |
| Number of studies | 70,000+ |
| Labeling | Information about each study, including target pathology (1 for presence, 0 for absence) |
| Pathologies | Intracerebral hemorrhage + ischemic stroke, angiography of intracranial vessels to identify brain vessel pathology, intracerebral hemorrhage, strokes (ischemic), automation of routine measurements of brain structures. |

**Media Slider:**

- **Image in the slider:** ![](https://unidata.pro/wp-content/uploads/2025/03/image2.webp)
- **Image in the slider:** ![](https://unidata.pro/wp-content/uploads/2025/03/image1-3.webp)

**Link to the sample:** [Download sample](https://drive.google.com/drive/folders/1okprzaCrYSdtWxBO5wSopxxPRPuwiMVj)

### Technical Specifications

**Table with data:**

| Characteristic | Data |
| --- | --- |
| File extension | DiCOM |
| Extension of labeling file | csv |

**Source and data collection methodology:** Source and collection methodology. Data was collected by a partner of Unidata

### Dataset Use Cases - Slider

**Industry Cards:**

- **Industry:** Medical Imaging and Diagnostics — **Title:** Detecting Brain Tumors and Hemorrhages — **Text:** CT Scans of Brain Dataset provides high-quality CT images for identifying brain tumors, intracranial hemorrhages, and traumatic brain injuries. Radiologists and researchers use this training data to build and validate segmentation algorithms that enhance early detection accuracy and improve diagnostic workflows in both clinical and research environments.
- **Industry:** Medical AI Development — **Title:** Training Deep Learning Models for Brain Analysis — **Text:** This dataset serves as reliable training data for developing machine learning and deep learning models in medical image analysis. The dataset contains segmented brain structures and corresponding annotations that help algorithms learn to detect and classify abnormalities in complex CT scans and brain tissues.
- **Industry:** Healthcare Research and Radiology Studies — **Title:** Healthcare Research and Radiology Studies — **Text:** Researchers use these Brain CT scans to evaluate new imaging techniques, refine segmentation performances, and compare results across different learning algorithms. The dataset comprises DiCOM files and consistent image quality, supporting studies that improve understanding of brain cancers, lesions, and post-injury tissue behavior.
- **Industry:** Medical Software and Tool Testing — **Title:** Benchmarking Segmentation and Detection Tools — **Text:** This dataset provides standardized medical images for testing segmentation tools, image reconstruction software, and automated diagnostic systems. With detailed brain CT exams and segmentation masks, developers can assess system accuracy and optimize medical imaging pipelines for clinical use.

### Fact

**FAQs Heading:** FAQs

**List of Questions:**

- **Question:** What is included in this dataset? — **Answer:** The dataset contains over 70,000 DiCOM brain CT studies, with detailed metadata and pathology annotations. Each study includes segmentation masks and binary labels indicating the presence or absence of specific conditions like intracerebral hemorrhage, ischemic stroke, and vascular abnormalities.
- **Question:** What types of annotations are provided? — **Answer:** Annotations are provided in CSV format, detailing the target pathology for each study (1 for presence, 0 for absence). Additionally, segmentation masks are included to support the precise localization of brain abnormalities for training and validating AI-based diagnostic models.
- **Question:** Can I request a sample of the dataset before purchasing or downloading it? — **Answer:** Yes. Unidata provides free sample data for initial testing and evaluation. This allows researchers and developers to verify the dataset’s DiCOM format, labeling consistency, and image quality before proceeding with a full purchase.
- **Question:** What should I consider before purchasing this dataset? — **Answer:** Before buying the dataset, review its structure, file format, and labeling details to ensure it aligns with your research or clinical objectives. Confirm that it supports your intended use cases.
- **Question:** Is it possible to request a custom dataset? — **Answer:** Absolutely. Unidata offers custom dataset creation based on user requirements, such as specific pathologies, scan types, or imaging protocols. You can also request tailored segmentation masks or annotations to better fit your diagnostic, research, or training objectives.
- **Question:** How are Unidata datasets licensed? — **Answer:** Unidata datasets follow a dual-licensing model: free samples are provided for testing, while full datasets are available exclusively through purchase. This ensures that research and commercial users receive high-quality, verified medical imaging data for their projects.
- **Question:** How are Unidata datasets stored? — **Answer:** All datasets are securely hosted on AWS cloud infrastructure, ensuring robust performance and accessibility. Unidata adheres to ISO 27001 and ISO 27701 standards, maintaining high levels of data protection and ensuring safe handling of sensitive medical images.
- **Question:** Is this a real-world dataset or synthetic data? — **Answer:** This dataset contains real-world brain CT scans from clinical environments. Each study reflects authentic imaging conditions.

[Full list of this site's AI-readable pages](https://unidata.pro/llms.txt)