Skip to content

Research on Intelligent Recognition and Diagnosis of Burn Wound Images Based on Deep Learning

Research and Development of a Deep Learning-Based Intelligent Diagnosis System for Burn Grading

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2600129509
Enrollment
Unknown
Registered
2026-08-05
Start date
2025-03-31
Completion date
Unknown
Last updated
2026-08-10

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

burn

Interventions

Observation group:None

Sponsors

Affiliated Hospital of Jiangnan University
Lead Sponsor

Eligibility

Sex/Gender
All

Inclusion criteria

Inclusion criteria: 1. Patients with a confirmed diagnosis of skin burns; 2. Burns predominantly caused by thermal energy, with wounds in the acute phase (injury<= 72 hours); 3. clear, unobstructed clinical images of burn wounds suitable for deep learning analysis; 4. Photographs of the burn wound area, with pixelation applied to private areas where applicable; 5. Burn depth determined by two or more specialist burn surgeons as the gold standard.

Exclusion criteria

Exclusion criteria: 1. Image blurring, abnormal exposure, severe occlusion, or insufficient resolution; 2. Wounds complicated by infection, necrosis, eschar coverage, or application of coloured medications; 3. Concurrent presence of other skin lesions such as dermatoses, chronic ulcers, or tumours; 4. Old burns, scar-stage wounds, or healing-stage wounds.

Design outcomes

Primary

MeasureTime frame
Macro-average F1 score for burn depth classification;

Secondary

MeasureTime frame
Dice coefficient for burn area segmentation;

Countries

China

Contacts

Public ContactJia Zhigang

Affiliated Hospital of Jiangnan University

jiazhigang.5@163.com+86 151 9021 8027

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Aug 25, 2026