/
/
Machine Learning Algoritmer til tidlig påvisning af Bone Metastases i en eksperimentel rotte model
A subscription to JoVE is required to view this content. Sign in or start your free trial.
1Department of Radiology, University Hospital Erlangen,Friedrich-Alexander Universität Erlangen-Nürnberg, 2Department of Medical Physics in Radiation Oncology,German Cancer Research Center, 3Department of Internal Medicine 3, University Hospital Erlangen,Friedrich-Alexander Universität Erlangen-Nürnberg
Chapters
- 00:04Introduction
- 00:46Magnetic Resonance Imaging (MRI) Analysis
- 02:11Positron Emission Tomography/Computed Tomography (PET/CT) Analysis
- 03:03Tumor-Take Rate Calculation
- 03:30Feature Selection
- 04:23Machine Learning (ML) Analysis, Model-Averaged Neural Network (avNNet) ML Algorithm Training, and ML Algorithm Data Analysis
- 05:12Results: Representative Machine Learning Algorithm Rat Bone Metastasis Detection
- 06:19Conclusion
Denne protokol var designet til at træne en machine learning algoritme til at bruge en kombination af billeddannelse parametre stammer fra magnetisk resonans imaging (MR) og positron emission tomografi / computertomografi (PET / CT) i en rotte model af brystkræft knoglemetastaser til at opdage tidlig metastatisk sygdom og forudsige efterfølgende progression til makrometastaser.
