Pathology Foundation Model
Conditions
Keywords
Randomized Controlled Trial
Brief summary
The investigators plan to conduct a multicenter, prospective, randomized controlled trial to systematically evaluate the incremental value of pathology-based artificial intelligence (AI) models in a pan-disease diagnostic workflow. The study will primarily compare interpretation using an AI-assisted platform with conventional independent slide reading in terms of diagnostic accuracy (e.g., AUC), reading efficiency (e.g., diagnostic time), diagnostic report quality, diagnostic confidence (Likert scale), and pathologists' satisfaction with the AI model. Investigators will also assess superiority among less experienced (junior) pathologists and non-inferiority among more experienced (senior) pathologists. Successful completion of this project will provide high-level prospective evidence to support standardized deployment, quality control, and broader implementation of pathology AI in clinical practice. This trial may also evaluate the potential benefits and risks of using AI tools in medical research.
Detailed description
In this study, investigators plan to enroll 60 pathologists with varying levels of experience and 2,000 patients requiring pathological diagnosis, with whole-slide images (WSIs) collected.
Interventions
OTHERAI model
Doctors in this group are required to use the AI model to assist their diagnoses. The AI pathology model will provide a predicted result for each case.
OTHERControl
Pathologists will independently diagnose each case based on their own clinical experience, and will record both their time to diagnosis and their diagnostic confidence.
Sponsors
Nanfang Hospital, Southern Medical University
Study design
Allocation
RANDOMIZED
Intervention model
CROSSOVER
Primary purpose
OTHER
Masking
SINGLE (Outcomes Assessor)
Eligibility
Sex/Gender
ALL
Age
18 Years to 100 Years
Healthy volunteers
Yes
Inclusion criteria
Pathologists: Inclusion Criteria: 1. Voluntarily provide written informed consent. 2. Age ≥ 20 years. 3. Have completed at least 1 year of training in pathological diagnosis.
Exclusion criteria
1. Individuals with reading difficulties or a reading disorder. 2. Unwilling to participate in this study. Patients: Inclusion Criteria: 1. Voluntarily provide written informed consent. 2. Age ≥ 18 years. 3. Have available digital pathology images and relevant clinical information.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Area under ROC curve (AUC) | Assessments will be conducted within one week after the pathologists' diagnoses | Area under the curve |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Diagnostic time per case | Measured immediately after the pathologists' diagnosis | Time required for the pathologist to complete the diagnosis of each case in the AI-assisted diagnosis group compared with the independent diagnosis group. Diagnostic time is defined as the duration (in minutes/seconds) from initiating case review to finalizing and submitting the diagnostic report in the study system. |
| Pathologists' diagnostic confidence | At the time of diagnosis for each case. | Self-reported diagnostic confidence of pathologists for each case in the AI-assisted diagnosis group compared with the independent diagnosis group. Diagnostic confidence will be rated by the reporting pathologist on a \[10\]-point Likert scale (e.g., 1 = very uncertain to 10 = very confident) immediately after completing the diagnosis. Higher scores indicate greater diagnostic confidence. |
Countries
China
Contacts
CONTACTZhengyu Zhang
STUDY_CHAIRLi Liang
Nanfang Hospital, Southern Medical University
Outcome results
None listed