CT-Based Automatic Spine Segmentation Using Patch-Based Deep Learning

Abstract

CT vertebral segmentation plays an essential role in various clinical applications, such as computer-assisted surgical interventions, assessment of spinal abnormalities, and vertebral compression fractures. Automatic CT vertebral segmentation is challenging due to the overlapping shadows of thoracoabdominal structures such as the lungs, bony structures such as the ribs, and other issues such as ambiguous object borders, complicated spine architecture, patient variability, and fluctuations in image contrast. Deep learning is an emerging technique for disease diagnosis in the medical field. This study proposes a patch-based deep learning approach to extract the discriminative features from unlabeled data using a stacked sparse autoencoder (SSAE). 2D slices from a CT volume are divided into overlapping patches fed into the model for training. A random under sampling (RUS)-module is applied to balance the training data by selecting a subset of the majority class. SSAE uses pixel intensities alone to learn high-level features to recognize distinctive features from image patches. Each image is subjected to a sliding window operation to express image patches using autoencoder high-level features, which are then fed into a sigmoid layer to classify whether each patch is a vertebra or not. We validate our approach on three diverse publicly available datasets: VerSe, CSI-Seg, and the Lumbar CT dataset. Our proposed method outperformed other models after configuration optimization by achieving 89.9% in precision, 90.2% in recall, 98.9% in accuracy, 90.4% in F-score, 82.6% in intersection over union (IoU), and 90.2% in Dice coefficient (DC). The results of this study demonstrate that our model’s performance consistency using a variety of validation strategies is flexible, fast, and generalizable, making it suited for clinical application.

Yoga Meets Intelligent Internet of Things: Recent Challenges and Future Directions

“…SC is computed based on AoD and SoE. The other indicators such as recall, precision, specificity, F1-score, and accuracy [ 129 , 130 , 131 , 132 , 133 , 134 , 135 , 136 , 137 ] are computed in Equations (1)–(5) below: where TP, TN, FN, and FP represent true positive, true negative, false negative, and false positive.…”

An efficient hybrid extreme learning machine and evolutionary framework with applications for medical diagnosis

“…Besides, recent advances in medical diagnosis have witnessed a surge in the adoption of deep learning methods. Studies such as (Ahmad, Ai, et al, 2019; Ahmad, Ding, et al, 2019; Alzghoul et al, 2023; Bataineh, 2019; Doppala et al, 2023; Furqan Qadri et al, 2019; Habib & Qureshi, 2023; Hirra et al, 2021; Qadri et al, 2019; Qadri et al, 2021; Qadri et al, 2022; Qadri et al, 2023; Vamsi et al, 2022) have demonstrated deep learning ability in dealing with large datasets and complex patterns. While deep learning methods provide robust performance (Ahmadian et al, 2021; Arora et al, 2022; Jalali, Ahmadian, Ahmadian, et al, 2022; Jalali, Arora, Panigrahi, et al, 2022; Jalali, Osorio, Ahmadian, et al, 2022; Mehnatkesh et al, 2023; Saffari et al, 2023), their computational density and the need for extensive data and training can be limited in certain scenarios.…”

A classification algorithm based on improved meta learning and transfer learning for few‐shot medical images

“…The success of deep learning is always attributed to the growing number of large data sets and powerful computing hardware, especially in the field of medical images, which rely heavily on accurate labels and gold standards [8,9]. However, the scarcity of medical image data annotation, small sample size, uneven distribution, and poor domain generalization still seriously affect the development of related fields.…”