Development and Validation of a Deep Learning System for Nasopharyngeal Carcinoma Using Endoscopic Images
Study Details
Study Description
Brief Summary
Develop a deep learning algorithm via nasal endoscopic images from eight NPC treatment centerto detect and screen nasopharyngeal carcinoma(NPC).
Condition or Disease | Intervention/Treatment | Phase |
---|---|---|
|
Detailed Description
Nasopharyngeal carcinoma (NPC) is an epithelial cancer derived from nasopharyngeal mucosa. Nasal endoscopy is the conventional examination for NPC screening. It is a major challenge for inexperienced endoscopists to accurately distinguish NPC and other benign dieseases. In this study, we collcet multi-center endoscopic images and train a deep learning model to detect NPC and indicate tumor location. Then, the model perfomance will be compared with endoscopists and be tested prospectively with external dataset.
Study Design
Arms and Interventions
Arm | Intervention/Treatment |
---|---|
Training Cohort Nasopharyngeal endoscopic images collected from 8 hospitals all over China |
|
Validation Cohort Nasopharyngeal endoscopic images collected from 8 hospitals all over China |
Other: Diagnostic
Training dataset was used to train the deep learning model, which was validated and tested by external dataset.
|
Testing Cohort Nasopharyngeal endoscopic images prospectively collected from 8 hospitals all over China |
Other: Diagnostic
Training dataset was used to train the deep learning model, which was validated and tested by external dataset.
|
Outcome Measures
Primary Outcome Measures
- Area under the receiver operating characteristic curve of the deep learning algorithm [baseline]
The investigators will calculate the area under the receiver operating characteristic curve of deep learning algorithm and compare this index between deep learning system and human doctors.
Secondary Outcome Measures
- Sensitivity of the deep learning system [baseline]
The investigators will calculate the sensitivity of deep learning algorithm and compare this index between deep learning system and human doctors.
- Specificity of the deep learning system [baseline]
The investigators will calculate the specificity of deep learning algorithm and compare this index between deep learning system and human doctors.
Eligibility Criteria
Criteria
Inclusion Criteria:
-
The quality of endoscopic images should clinical acceptable.
-
Patients were diagnosed with biopsy(NPC, benign hyperplasia). Control corhort(normal nasopharynx) don't require bispsy result.
Exclusion Criteria:
-
images with spots from lens flares or stains, and overexposure were excluded from further analysis.
-
image can not expose most part of lesion clearly.
Contacts and Locations
Locations
Site | City | State | Country | Postal Code | |
---|---|---|---|---|---|
1 | Fujian Medical University Union Hospital | Fuzhou | Fujian | China | |
2 | Quan Zhou First Affiliated Hospital of Fujian Medical University | Quanzhou | Fujian | China | |
3 | The People' s Hospital of Jiangmen | Jiangmen | Guangdong | China | |
4 | First Affiliated Hospital of Guangxi Medical University | Nanning | Guangxi | China | |
5 | The People' s Hospital of Guangxi Zhuang Autonomous Region | Nanning | Guangxi | China | |
6 | Xiangya Hospital of Central South University | Changsha | Hunan | China | |
7 | The First Affiliated Hospital of Nanchang University | Nanchang | Jiangxi | China | |
8 | Eye&ENT Hospital of Fudan University | Shanghai | Shanghai | China |
Sponsors and Collaborators
- Eye & ENT Hospital of Fudan University
- Xiangya Hospital of Central South University
- The First Affiliated Hospital of Nanchang University
- Fujian Medical University Union Hospital
- Quan Zhou First Affiliated Hospital of Fujian Medical University
- First Affiliated Hospital of Guangxi Medical University
- The People' s Hospital of Guangxi Zhuang Autonomous Region
- The People' s Hospital of Jiangmen
Investigators
- Principal Investigator: Hongmeng Yu, MD PhD, Eye&ENT Hospital, Fudan University
Study Documents (Full-Text)
None provided.More Information
Publications
None provided.- AIAD202204