Dataset Overview
Definition：Head CT Plain and Contrast-Enhanced Images and Reports Datasetis provided byChangsha Langhui Information Technology Co., Ltd.Large-scale neurological image data collections are obtained. This covers 5 million head-sweeping CT+ enhanced DCOM sequences (including salinization/articular/venial period), with structured diagnostic reports covering major nervous system diseases such as brain infarction, brain haemorrhage, and intracranial tumours.
| Dataset Name | Head CT Plain and Contrast-Enhanced Images and Reports Dataset |
| Total Data Volume | Five million CT-sequence images. |
| Data Type | DICOM sequence image (sweeping/articular/venial) + text report |
| Data Source | Resale (partnerships with more than 50 hospitals) |
| License Type | Sublicensing Including direct authorization from three Grade-A Tertiary Hospitals |
| Covered Disease Categories | Brain infarction, brain bleeding, intracranial tumor, subcorrosive bleeding, brain trauma. |
| Sequence Information | Scanning + arterial + IV 3 complete scan, layer thickness, contrast information |
Core Data Fields
Each piece of data contains complete structured fields that cover everything from raw data to specialized diagnostics.
| Field Category | Field Count | Typical Fields |
|---|---|---|
| DICOM Sequence | 10 | 3rd phase of smoothing/articular/cissory period images, layers, matrix size, pixel spacing, window wide slots, scanning parameters |
| Imaging Findings | 8 | Location/size/density/boundary, position effects, mid-line deviation, brain oedema, haemorrhagic transformation |
| Diagnostic Conclusion | 6 | Main, secondary, normal/unusual markers, ICD-10 code, NIHSS rating correlation |
| Patient Information | 4 | Age, gender, examination date, clinical department (DICOM-compliant de-identification) |
| Enhanced information | 3 | Name of the contrast agent, dosage, injection rate, delay |
Data Sample Showcase
The following is a de-identified structured JSON sample illustrating the typical data format.
{
"study_id": "HCT-2023-00123456",
"meta": {"age_group": "65-75","gender": "male","department": "Cardiology"},
"sequences": {"non_contrast": true,"arterial": true,"venous": false},
"findings": {"lesion": {"location": "sample value","size": "sample value","density": "sample value"},"mass_effect": "sample value","midline_shift_mm": 2},
"diagnosis": {"primary": "sample value","abnormal": true,"icd": "I63.5","nihss": 8}
}AI Application Scenarios
I'm a brain-scrutinizing AI.
Five million cases of CT-DIV model for acute brain infarction and brain haemorrhage autodetection, severing of the cooks and volume calculation.
Cerebral tumour detection
Sweep + enhanced multiphase phase data are used to train intracranial tumor detection models and to use enhanced features to achieve benign recognition and classification of the tumours.
Emergency C.T. Smart consultation.
The development of an automatic CT-based system for the head of emergency care, which will rapidly identify acute diseases such as brain haemorrhage and widespread brain infarction, will increase the efficiency of the green passages in emergency cases.
ASPECTS automatic rating
The data on brain infarction are used to train the APPECTS automatic scoring model, which supports clinical doctors in rapid assessment of the extent of infarction and decision-making on the sluice.
Autodetection of middle-line offsets
Based on the CT-based training model for the automatic measurement of mid-line deviation, the progress of the brain emema and changes in the internal pressure of the skull are monitored in real time.
Auto Generate Neural Image Report
Image-report matching data are used to train automatic generation models for neuroimage reports and to improve the efficiency of radiology.
Frequently asked questions
Need for customised medical data programs?
Changsha Langhui Information Technology Co., Ltd. The DataAssetsAPI platform is dedicated to providing high-quality, compliant medical data assets for AI companies and scientific research institutions. Cooperate with 50 + hospitals, including Fudan 100 type hospitals.
Data Sample Preview · Customized Data Solutions · Professional Technical Support