Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Learning-based Bone Quality Classification Method for Spinal Metastasis

Shiqi Peng, Bolin Lai, Guangyu Yao, Xiaoyun Zhang, Ya Zhang, Yan-Feng Wang, Hui Zhao

TL;DR

This work addresses automatic staging of spinal metastasis by classifying bone quality and posterolateral involvement from CT slices. It introduces a multi-task learning framework that decomposes bone quality into two binary subtasks (blastic and lytic) and jointly learns posterolateral involvement, using hard-sharing DenseNet backbones and an MLP fusion. Self-paced learning and label smoothing are incorporated to improve generalization on a limited, fine-grained dataset. The proposed Final-model outperforms a DenseNet baseline across slice- and vertebrae-level metrics, suggesting practical utility for radiologists in CT-based assessment of spinal metastasis.

Abstract

Spinal metastasis is the most common disease in bone metastasis and may cause pain, instability and neurological injuries. Early detection of spinal metastasis is critical for accurate staging and optimal treatment. The diagnosis is usually facilitated with Computed Tomography (CT) scans, which requires considerable efforts from well-trained radiologists. In this paper, we explore a learning-based automatic bone quality classification method for spinal metastasis based on CT images. We simultaneously take the posterolateral spine involvement classification task into account, and employ multi-task learning (MTL) technique to improve the performance. MTL acts as a form of inductive bias which helps the model generalize better on each task by sharing representations between related tasks. Based on the prior knowledge that the mixed type can be viewed as both blastic and lytic, we model the task of bone quality classification as two binary classification sub-tasks, i.e., whether blastic and whether lytic, and leverage a multiple layer perceptron to combine their predictions. In order to make the model more robust and generalize better, self-paced learning is adopted to gradually involve from easy to more complex samples into the training process. The proposed learning-based method is evaluated on a proprietary spinal metastasis CT dataset. At slice level, our method significantly outperforms an 121-layer DenseNet classifier in sensitivities by $+12.54\%$, $+7.23\%$ and $+29.06\%$ for blastic, mixed and lytic lesions, respectively, meanwhile $+12.33\%$, $+23.21\%$ and $+34.25\%$ at vertebrae level.

Learning-based Bone Quality Classification Method for Spinal Metastasis

TL;DR

This work addresses automatic staging of spinal metastasis by classifying bone quality and posterolateral involvement from CT slices. It introduces a multi-task learning framework that decomposes bone quality into two binary subtasks (blastic and lytic) and jointly learns posterolateral involvement, using hard-sharing DenseNet backbones and an MLP fusion. Self-paced learning and label smoothing are incorporated to improve generalization on a limited, fine-grained dataset. The proposed Final-model outperforms a DenseNet baseline across slice- and vertebrae-level metrics, suggesting practical utility for radiologists in CT-based assessment of spinal metastasis.

Abstract

Spinal metastasis is the most common disease in bone metastasis and may cause pain, instability and neurological injuries. Early detection of spinal metastasis is critical for accurate staging and optimal treatment. The diagnosis is usually facilitated with Computed Tomography (CT) scans, which requires considerable efforts from well-trained radiologists. In this paper, we explore a learning-based automatic bone quality classification method for spinal metastasis based on CT images. We simultaneously take the posterolateral spine involvement classification task into account, and employ multi-task learning (MTL) technique to improve the performance. MTL acts as a form of inductive bias which helps the model generalize better on each task by sharing representations between related tasks. Based on the prior knowledge that the mixed type can be viewed as both blastic and lytic, we model the task of bone quality classification as two binary classification sub-tasks, i.e., whether blastic and whether lytic, and leverage a multiple layer perceptron to combine their predictions. In order to make the model more robust and generalize better, self-paced learning is adopted to gradually involve from easy to more complex samples into the training process. The proposed learning-based method is evaluated on a proprietary spinal metastasis CT dataset. At slice level, our method significantly outperforms an 121-layer DenseNet classifier in sensitivities by , and for blastic, mixed and lytic lesions, respectively, meanwhile , and at vertebrae level.
Paper Structure (18 sections, 4 equations, 3 figures, 4 tables)

This paper contains 18 sections, 4 equations, 3 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Proposed Method
  3. Framework
  4. Feature Sharing:
  5. Loss Function:
  6. Self-paced Learning
  7. Implementation Details
  8. Experimental Results
  9. Dataset
  10. Metrics
  11. Experiments
  12. Ablation Studies
  13. Feature sharing method:
  14. Posterolateral Involvement:
  15. Bone Quality:
  16. ...and 3 more sections

Figures (3)

  • Figure 1: Typical examples for bone quality of spinal metastasis.
  • Figure 2: The proposed multi-task learning framework.
  • Figure 3: Confusion matrices for bone quality at slice level and vertebrae level.