peripheral lymphadenopathy
Conditions
Interventions
Gold Standard:Normal lymph nodes:follow-up
lymphadenopathy: histopathological diagnosis by core needle biopsy or surgical resection.
diagnostic 
of 
deep 
learning 
model 
in 
differentiating 
and 
lymphadenopathies(sensitivity, 
PPV, 
NPV, 
AUCs)
2.The 
Sponsors
The First Affiliated Hospital of Sun Yat-Sen University
Eligibility
Sex/Gender
All
Age
No minimum to 100 Years
Inclusion criteria
Inclusion criteria: 1.Healthy population: without any known history of cancer,or severe disease need treatment; normal lymph nodes on US; no enlargement in 6-month follow-up; 2.Patientsneed to undergo US examination to exclude peripheral lymphadenopathy because of palpable mass in neck, groin or axilla; history of cancer, need to undergo US examination to exclude peripheral lymph node metastasis; final histopathologic diagnosis by FNA or surgical resection. 3.Sonographic images*: (1) static sonographic images: each lymph node should obtain at least one grey-scale image and one color doppler image. 1) Grey-scale images: with a linear probe(Frequency >=7 MHz); Depth less then 5cm; Visualized at least one lymph node with its largest section; clear and clean without labels; 2) Color doppler images: ROI should include the entire target lymph nodes, the scale < 6cm/s and adjust the color gain as large as possible until the color signal just has no overflow. (2)sonographic videos(prosectively enrolled ): grey-scale videos and color doppler videos including at least one entire lymph node, other requirements are the same as static images;
Exclusion criteria
Exclusion criteria: Nil
Design outcomes
Primary
| Measure | Time frame |
|---|---|
| The diagnostic performance of deep learning model in differentiating benign and malignant lymphadenopathies;The diagnostic performance of deep learning model in classifying different lymphadenopathies;The comparison of diagnostic performances between deep learning model and human experts(Delong test).;With the help of deep learning model, how much of the accuracy can be improved, and how much of the biospy rate can be reduced in junior doctors and senior doctors, respectively; | — |
Secondary
| Measure | Time frame |
|---|---|
| The diagnostic performance of human experts in differentiating benign and malignant lymphadenopathies;The diagnostic performance of human experts in classifying different lymphadenopathies;The agreement among human experts(weighted ?statistics.); | — |
Countries
China
Contacts
Public ContactLuyao Zhou
The First Affiliated Hospital of Sun Yat-Sen University
Outcome results
None listed