Deep Learning-Based Identification of Spinal Metastasis in Lung Cancer Using Spectral CT Images

Abstract

In this study, deep learning algorithm-based energy/spectral computed tomography (CT) for the spinal metastasis from lung cancer was used. A dilated convolutional U-Net model (DC-U-Net model) was first proposed, which was used to segment the energy/spectral CT image of patients with the spinal metastasis from lung cancer. Subsequently, energy/spectral CT images under different energy levels were collected for the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) comparison. It was found the learning rate of the model decreased exponentially as the number of training increased, with the lung contour segmented out of the image. Under 40–65 keV, the CT value of bone metastasis from lung cancer decreased with increasing energy, as with the average rank sum test result. The SNR and CNR values were the highest under 60 keV. The detection rate of the deep learning algorithm below 60 keV was 81.41%, and that of professional doctors was 77.56%. The detection rate of the deep learning algorithm below 140 keV was 66.03%, and that of professional doctors was 64.74%. In conclusion, the DC-U-Net model demonstrates better segmentation effects versus the convolutional neutral networ k (CNN), with the lung contour segmented. Further, a higher energy level leads to worse segmentation effects on the energy/spectral CT image.

Application of Artificial Intelligence Methods for Imaging of Spinal Metastasis

“…Subsequently, with the recent advances in artificial intelligence in medical imaging [ 110 , 111 , 112 ], there were substantial improvements in the detection of spinal metastases with the aid of deep learning and convolutional neural networks. This has resulted in improvement in the accuracy of computer-assisted automated detection of spinal metastases across various imaging modalities with significant reduction in false positive and negative rates [ 75 , 85 , 88 , 89 ].…”

“…Their methods were able to detect all spinal metastatic lesions from their datasets, with a relatively low false positive rate of 0.40 per case. Fan XJ et al [ 85 ] developed a DC-U-Net model using energy/spectral CT imaging to improve detection of spinal metastases in patients with lung cancer. Their work utilised the fact that characteristic X-ray absorption by substances differs under various energy levels, which provides better detection, segmentation and differentiation of bone lesions.…”

Improved UNet Deep Learning Model for Automatic Detection of Lung Cancer Nodules

“…The study's goal was to develop a clinical standard for lung cancer bone metastases. Figure 1 shows the segmentation approach used by the Improved UNet model [ 17 ].…”

Emerging Applications of Deep Learning in Bone Tumors: Current Advances and Challenges

“…Trained on voxel-wise labeled images by two independent radiologists, the machine had 92% sensitivity in classifying and localizing small metastatic lesions greater than 1.4 mm 3 under object-wise evaluation. Similarly, a deep learning-based DC-U-Net model ( 116 ) was also built to identify and segment spinal metastasis in lung cancer on spectral CT (dual-energy) images and have expert-level performance. Fan et al ( 117 ) proposed a deep approach that employed the AdaBoost algorithm to classify images and the Chan-Vese (CV) algorithm to segment the lesion for the diagnosis of spinal bone metastasis in lung cancer patients on MRI images, achieving a high classification accuracy of 96.55%.…”