Dataset Overview
Definition：Dermoscopy Skin Lesion Imaging Datasetis provided byChangsha Langhui Information Technology Co., Ltd.The large-scale skin-disease image labelling database is built by Long Salang Langhui Information Technology. It covers skin mirror images, skin-dermal clinical photographs and standard diagnostic reports of pathological gold, maintains original resolution, oscillation/impregnation patterns and equipment information, and provides 14-dimensional structural indications of the type of skin damage (original hair/release 30+ seed type), colour, size, number, shape, ranking, edge, etc., and labels the disease term Top3 and ICD coded mapping.
| Dataset Name | Dermoscopy Skin Lesion Imaging Dataset |
| Total Data Volume | 1,000,000 cases of skin mirrors included |
| Imaging Modality | Dermoscope + clinical photocopy |
| Raw Data | Original skin mirror images and skin-dermal clinical photographs of equipment, retention of original resolution, oscillation/impregnation mode, equipment information |
| Annotation Dimensions | 14 structured dimensions: age/part/flank type/colour/size/number/shape/arrange/wet/margin/screech, etc. |
| Type of skin | Original hair loss (spectal/grape/chel/chrysal/sip/sip/sip/fung/fung) + Retardation (crum/sip/sip/sprout/ulsion, etc.) + 30 + seed type |
| Disease label | Each case is annotated with the TOP 3 disease diagnosis terms (including ICD-10 code mapping) |
| Space label | Bunding Box, supporting multi-disease stoves and labelling |
| Source Institution | Co-operation Sanction Section, Sanctuary Hospital |
| Use Cases | AIS-assisted diagnosis of skin diseases, skin mirror image analysis, plaster screening, skin loss and classification |
Delivery Specifications and Field Requirements
The data set strictly follows the requirements of the uniform DRM-SKIN delivery index to ensure data quality and traceability.
Delivery & Counting：One line corresponding to skin mirror examination/morbidity; multi-series/multi-view only 1 case
Coverage：Cover skin mirror images and skin-dermal clinical photographs, and maintain original resolution and equipment information
Positives and Composition：Identify the pathogen as the subject; each of the cases is benign, normal and uncertain
PathologyGold Standard：Pathologically confirmed results serve as the gold standard; suspected, considered and confirmed cases are retained separately
Cross-Data Association：Prioritize skin pathology, treatment and follow-up
2026 Frontier AI Research Progress
Domain Review：The Skin Synthetic Synthetics ACAI has made a breakthrough in multimodular integration (TG-CAVNet) and integrated learning (XGBoost+Xception AUC 0.988), with XAI interpretability and few-shot learning as the front-line direction.
Depth Learning + Integrated Learning for Blackoma Identification
Frontiers in Oncology, 2026The XGBoost+Xception combination of AUC up to 0.988 in the Test Series, AUC 1.00 compared the performance of nine CNN models on ISIC-2024 and HAM10000.
TT-CAVNet text leading multi-modular skin pathology detection
Scientific Reports, 2026The clinical text that integrates the Bio-ClinicalBerrt code with the EfficientNet-B4 visual features is used to achieve multi-modular detection through text-led channel characterization and cross-focus.
EfficientNetV2+XAI automatic diagnosis of skin dermal pathologies
arXiv, 2026Based on EfficientNetV2-L+Crossion + Grad-CAM/Salitcy's Explanatory Framework, the overall accuracy rate is 91.15%, macro F1 85.45%.
Classification of dermal pathologies in small AMCANet samples
PLOS ONEThe introduction of a multiscale, multiscale, volume-based attention network to improve the few-shot classification of the skin disease in terms of the details of the local structure and boundary characteristics.
Annotation Workflow & Quality Control
Image Acquisition & De-identification
Standardized acquisition workflow: complete data is exported directly from the equipment, and patient identifying information (PHI) is removed before the data is stored。
Initial Annotation (Specialist Physician)
Attending physicians annotate case by case against the standards, including lesion localization, morphological description and disease term determination。
Review (Associate Chief Physician or Above)
Experts with the title of Associate Chief Physician or above review the initial annotation results item by item, correcting erroneous annotations and supplementing missing dimensions。
Consistency Assessment
10% of samples are randomly selected and independently annotated by 3 physicians to calculate Fleiss' Kappa; below 0.Dimensions scored 75 are flagged for rework。
License and Usage Agreement
Academic Research License
For universities and research institutions, supporting academic research related to medical AI。After signing the agreement, a de-identified data subset is provided; the source must be acknowledged：Langhui Technology DataAssetsAPI。
Commercial License
For medical device companies and AI diagnostics companies, supporting medical AI product R&D and medical device registration。Provides full datasets + custom annotation + incremental update services。
Data Compliance Statement
All images come from legally authorized sources; PHI fields are fully de-identified and contain no information that can directly identify an individual, in compliance with the Personal Information Protection Law and the Data Security Law。
Customized services
Supports extended requirements such as adding specific disease types, multimodal annotation expansion, and custom training of AI-assisted diagnostic models。
Quick Facts
AI Frontier Research
The Skin Synthetic Synthetics ACAI has made a breakthrough in multimodular integration (TG-CAVNet) and integrated learning (XGBoost+Xception AUC 0.988), with XAI interpretability and few-shot learning as the front-line direction.