In Patterns (New York, N.Y.)
Treatment decisions for brain metastatic disease rely on knowledge of the primary organ site and are currently made with biopsy and histology. Here, we develop a deep-learning approach for accurate non-invasive digital histology with whole-brain magnetic resonance imaging (MRI) data. Contrast-enhanced T1-weighted and fast spoiled gradient echo brain MRI exams (n = 1,582) were preprocessed and input to the proposed deep-learning workflow for tumor segmentation, modality transfer, and primary site classification into one of five classes. Tenfold cross-validation generated an overall area under the receiver operating characteristic curve (AUC) of 0.878 (95% confidence interval [CI]: 0.873,0.883). These data establish that whole-brain imaging features are discriminative enough to allow accurate diagnosis of the primary organ site of malignancy. Our end-to-end deep radiomic approach has great potential for classifying metastatic tumor types from whole-brain MRI images. Further refinement may offer an invaluable clinical tool to expedite primary cancer site identification for precision treatment and improved outcomes.
Lyu Qing, Namjoshi Sanjeev V, McTyre Emory, Topaloglu Umit, Barcus Richard, Chan Michael D, Cramer Christina K, Debinski Waldemar, Gurcan Metin N, Lesser Glenn J, Lin Hui-Kuan, Munden Reginald F, Pasche Boris C, Sai Kiran K S, Strowd Roy E, Tatter Stephen B, Watabe Kounosuke, Zhang Wei, Wang Ge, Whitlow Christopher T
2022-Nov-11
MRI, brain metastasis, classification, deep learning, primary organ site, vision transformer
