Deep convolutional neural network-based segmentation and classification of difficult to define metastatic spinal lesions in 3D CT data

• Published in: Medical image analysis Vol. 49; pp. 76 - 88
• Main Authors: Chmelik, Jiri, Jakubicek, Roman, Walek, Petr, Jan, Jiri, Ourednicek, Petr, Lambert, Lukas, Amadori, Elena, Gavelli, Giampaolo
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01-10-2018; Elsevier BV
• Subjects: Artificial neural networks; Classification; Computed tomography; Computer aided detection; Convolutional neural network; Correlation analysis; CT analysis; Feature extraction; Image classification; Image detection; Image processing; Image segmentation; Lesions; Longitudinal studies; Medical imaging; Metastases; Metastasis; Neural networks; Post-production processing; Spinal cord; Spinal metastasis; Spine; Spine (lumbar); Tomography; Computer aided detection; 92C55; Spinal metastasis; 68U10; 68T10; Convolutional neural network; CT analysis
• ISSN: 1361-8415; 1361-8423
• DOI: 10.1016/j.media.2018.07.008

•Individual architecture of convolutional neural network.•Patient and scan protocol dependent pre-processing.•Medial axis transform post-processing for shape simplification of lesion candidates.•Usability of the proposed method on whole-spine scans (cervical, thoracic, lumbar).
[Display omitted]
This paper aims to address the segmentation and classification of lytic and sclerotic metastatic lesions that are difficult to define by using spinal 3D Computed Tomography (CT) images obtained from highly pathologically affected cases. As the lesions are ill-defined and consequently it is difficult to find relevant image features that would enable detection and classification of lesions by classical methods of texture and shape analysis, the problem is solved by automatic feature extraction provided by a deep Convolutional Neural Network (CNN). Our main contributions are: (i) individual CNN architecture, and pre-processing steps that are dependent on a patient data and a scan protocol – it enables work with different types of CT scans; (ii) medial axis transform (MAT) post-processing for shape simplification of segmented lesion candidates with Random Forest (RF) based meta-analysis; and (iii) usability of the proposed method on whole-spine CTs (cervical, thoracic, lumbar), which is not treated in other published methods (they work with thoracolumbar segments of spine only).
Our proposed method has been tested on our own dataset annotated by two mutually independent radiologists and has been compared to other published methods. This work is part of the ongoing complex project dealing with spine analysis and spine lesion longitudinal studies.