Combining Deep Learning and Radiomics for Automated, Objective, Comprehensive Bone Marrow Characterization From Whole-Body MRI: A Multicentric Feasibility Study.
Markus WennmannAndré KleinFabian BauerJiri ChmelikMartin GrözingerCharlotte UhlenbrockJakob LochnerTobias NonnenmacherLukas Thomas RotkopfSandra SauerThomas HielscherMichael GötzRalf Omar FlocaPeter NeherDavid BonekampJens HillengassJens KleesiekNiels WeinholdTim Frederik WeberHartmut GoldschmidtStefan DelormeKlaus Maier-HeinHeinz-Peter SchlemmerPublished in: Investigative radiology (2022)
This pilot study demonstrates the feasibility of automatic, objective, comprehensive BM characterization from wb-MRI in multicentric data sets. This concept allows the extraction of high-dimensional phenotypes to capture the complexity of disseminated BM disorders from imaging. Further studies need to assess the clinical potential of this method for automatic staging, therapy response assessment, or prediction of biopsy results.
Keyphrases
- deep learning
- contrast enhanced
- bone marrow
- machine learning
- magnetic resonance imaging
- artificial intelligence
- convolutional neural network
- high resolution
- big data
- diffusion weighted imaging
- mesenchymal stem cells
- lymph node
- electronic health record
- computed tomography
- magnetic resonance
- stem cells
- human health
- risk assessment
- fluorescence imaging
- cell therapy
- data analysis
- case control
- neural network