Abstract
PurposeThis single-center study aimed to develop a convolutional neural network to segment multiple consecutive axial magnetic resonance imaging (MRI) slices of the lumbar spinal muscles of patients with lower back pain and automatically classify fatty muscle degeneration.MethodsWe developed a fully connected deep convolutional neural network (CNN) with a pre-trained U-Net model trained on a dataset of 3,650 axial T2-weighted MRI images from 100 patients with lower back pain. We included all qualities of MRI; the exclusion criteria were fractures, tumors, infection, or spine implants. The training was performed using k-fold cross-validation (k = 10), and performance was evaluated using the dice similarity coefficient (DSC) and cross-sectional area error (CSA error). For clinical correlation, we used a simplified Goutallier classification (SGC) system with three classes.ResultsThe mean DSC was high for overall muscle (0.91) and muscle tissue segmentation (0.83) but showed deficiencies in fatty tissue segmentation (0.51). The CSA error was small for the overall muscle area of 8.42%, and fatty tissue segmentation showed a high mean CSA error of 40.74%. The SGC classification was correctly predicted in 75% of the patients.ConclusionOur fully connected CNN segmented overall muscle and muscle tissue with high precision and recall, as well as good DSC values. The mean predicted SGC values of all available patient axial slices showed promising results. With an overall Error of 25%, further development is needed for clinical implementation. Larger datasets and training of other model architectures are required to segment fatty tissue more accurately.
Highlights
Low back pain (LBP) generally correlates with the grade of fatty infiltration of the lumbar multifidus muscle [1], independent of the weight and activity of the patient
In the field of medical image processing, computational models based on convolutional neural networks (CNNs) show promising results in pattern recognition and image segmentation
This study aims to develop a CNN to segment multiple consecutive axial magnetic resonance imaging (MRI) slices of the lumbar spinal muscles of patients with lower back pain and automatically classify fatty muscle degeneration
Summary
Low back pain (LBP) generally correlates with the grade of fatty infiltration of the lumbar multifidus muscle [1], independent of the weight and activity of the patient. Degeneration of the lumbar paravertebral musculature seems to influence the outcomes of patients with degenerative spinal degeneration [3,4,5]. As described in more detail below, we questioned whether an automated method could reliably detect fatty infiltration in the paravertebral musculature. We chose a model that uses a U-NET architecture, established as a fast and secure method for the automated semantic segmentation of biomedical images. This model uses a fully convolutional network and replaces upsampling operators with pooling operations and is designed for image segmentation [6]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have