Germ Cell Tumor, Lymphoma, Neuroblastoma, Sarcoma, Teratoma, Wilms' Tumor
Conditions
Brief summary
The aim of this study was to evaluate the diagnostic efficacy of computer aided diagnostic tool for retroperitoneal tumor using machine learning and deep learning techniques on computed tomography images in children.
Detailed description
The retroperitoneal space extends from the lumbar region to the pelvic region and houses vital structures such as the kidney, the ureter, the adrenal glands, the pancreas, the aorta and its branches, the inferior vena cava and its tributaries, lymph nodes, and loose connective tissue meshwork along with fat. This space thus allows the silent growth of primary and metastatic tumors, such that clinical features appear often too late. The therapeutic regimen differs on various types of retroperitoneal tumor in children. It is damaging for pediatric patients to acquire histological specimens through invasive procedures. Hence, an urgent evaluation is absolutely necessary for preoperative diagnosis in such cases via noninvasive approaches. This study is a retrospective-prospective design by West China Hospital, Sichuan University, including clinical data and radiological images. A retrospective database was enrolled for patients with definite histological diagnosis and available computed tomography images from June 2010 and December 2020. The investigators have constructed deep learning and machine learning radiomics diagnostic models on this retrospective cohort and validated it internally. A prospective cohort would recruit infantile patients diagnosed as retroperitoneal tumor since January 2021. The proposed deep learning model would also be validated in this prospective cohort externally. The aim of this study was to evaluate the diagnostic efficacy of computer aided diagnostic tool for retroperitoneal tumor using machine learning and deep learning techniques on computed tomography images in children.
Interventions
DIAGNOSTIC_TESTRadiomic Algorithm
Different radiomic, machine learning, and deep learning strategies for radiomic features extraction, sorting features and model constriction.
Sponsors
West China Hospital
Study design
Observational model
COHORT
Time perspective
OTHER
Eligibility
Sex/Gender
ALL
Age
0 Years to 18 Years
Healthy volunteers
No
Inclusion criteria
* Age up to 18 years old * Receiving no treatment before diagnosis * With written informed consent
Exclusion criteria
* Clinical data missing * Unavailable computed tomography images * Without written informed consent
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Pathological tumor diagnosis | Baseline | The diagnosis is defined by histopathological specimens from surgery and/or biopsy. |
Countries
China
Contacts
Primary ContactYuhan Yang, MD
Outcome results
None listed