Automated Whole-Body Bone Lesion Detection for Multiple Myeloma on 68Ga-Pentixafor PET/CT Imaging Using Deep Learning Methods

• Published in: Contrast media and molecular imaging Vol. 2018; no. 2018; pp. 1 - 11
• Main Authors: Menze, Bjoern H., Shi, Kuangyu, Piraud, Marie, Christ, Patrick, Li, Hongwei, Zhao, Yu, Lipkova, Jana, Tetteh, Giles, Xu, Lina, Buck, Andreas
• Format: Journal Article
• Language: English
• Published: Cairo, Egypt Hindawi Publishing Corporation 01.01.2018; Hindawi; Hindawi Limited
• Subjects: Artificial intelligence; Artificial neural networks; Automation; Bone imaging; Bone lesions; Bone marrow; Change detection; Computed tomography; Computer simulation; Deep learning; Diagnostic systems; Feasibility studies; Identification methods; Image processing; Image segmentation; Learning algorithms; Lesions; Localization; Machine learning; Medical imaging; Medical prognosis; Metabolism; Multiple myeloma; Neural networks; NMR; Nuclear magnetic resonance; Population studies; Positron emission; Semantics; Support vector machines; Tomography
• ISSN: 1555-4309; 1555-4317; 1555-4317
• DOI: 10.1155/2018/2391925

The identification of bone lesions is crucial in the diagnostic assessment of multiple myeloma (MM). 68Ga-Pentixafor PET/CT can capture the abnormal molecular expression of CXCR-4 in addition to anatomical changes. However, whole-body detection of dozens of lesions on hybrid imaging is tedious and error prone. It is even more difficult to identify lesions with a large heterogeneity. This study employed deep learning methods to automatically combine characteristics of PET and CT for whole-body MM bone lesion detection in a 3D manner. Two convolutional neural networks (CNNs), V-Net and W-Net, were adopted to segment and detect the lesions. The feasibility of deep learning for lesion detection on 68Ga-Pentixafor PET/CT was first verified on digital phantoms generated using realistic PET simulation methods. Then the proposed methods were evaluated on real 68Ga-Pentixafor PET/CT scans of MM patients. The preliminary results showed that deep learning method can leverage multimodal information for spatial feature representation, and W-Net obtained the best result for segmentation and lesion detection. It also outperformed traditional machine learning methods such as random forest classifier (RF), k-Nearest Neighbors (k-NN), and support vector machine (SVM). The proof-of-concept study encourages further development of deep learning approach for MM lesion detection in population study.