FindTrial
Sign In

An intelligent imaging decision system for skin tumors based on a multi-task self-supervised adversarial learning approach

An intelligent imaging decision system for skin tumors based on a multi-task self-supervised adversarial learning approach

Status
Recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2300074765
Enrollment
Unknown
Registered
2023-08-16
Start date
2023-08-16
Completion date
Unknown
Last updated
2023-08-21

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

Conditions

Skin disease

Interventions

Gold Standard:Surgery or biopsy
Index test:High frequency ultrasound image and clinical close-up image evaluation

Sponsors

Shanghai Skin Disease Hospital
Lead Sponsor

Eligibility

Sex/Gender
All
Age
18 Years to 80 Years

Inclusion criteria

Inclusion criteria: patients who received a preliminary diagnosis from the clinician; patients who underwent preoperative high-frequency ultrasound; and patients with definite histopathological findings

Exclusion criteria

Exclusion criteria: (1) History of infectious diseases; (2) The lesion has an open wound; (3) Minors under 18 years old and elderly people over 80 years old; (4) The lesion underwent biopsy before examination.

Design outcomes

Primary

MeasureTime frame
diagnosis based on clinical images;diagnosis based on ltrasound images;diagnosis based on CNN;AUC;accuracy;sensitivity;specificity;—

Countries

China

Contacts

Public ContactYujing Zhao

Shanghai Skin Disease Hospital, Shanghai, China

Zhaoyj@shskin.com+86 180 1733 6551

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Feb 9, 2026