A Region-Based Deep Level Set Formulation for Vertebral Bone Segmentation of Osteoporotic Fractures

Abstract

Accurate segmentation of the vertebrae from medical images plays an important role in computer-aided diagnoses (CADs). It provides an initial and early diagnosis of various vertebral abnormalities to doctors and radiologists. Vertebrae segmentation is very important but difficult task in medical imaging due to low-contrast imaging and noise. It becomes more challenging when dealing with fractured (osteoporotic) cases. This work is dedicated to address the challenging problem of vertebra segmentation. In the past, various segmentation techniques of vertebrae have been proposed. Recently, deep learning techniques have been introduced in biomedical image processing for segmentation and characterization of several abnormalities. These techniques are becoming popular for segmentation purposes due to their robustness and accuracy. In this paper, we present a novel combination of traditional region-based level set with deep learning framework in order to predict shape of vertebral bones accurately; thus, it would be able to handle the fractured cases efficiently. We termed this novel Framework as BFU-Net^which is a powerful and practical framework to handle fractured vertebrae segmentation efficiently. The proposed method was successfully evaluated on two different challenging datasets: (1) 20 CT scans, 15 healthy cases, and 5 fractured cases provided at spine segmentation challenge CSI 2014; (2) 25 CT image data (both healthy and fractured cases) provided at spine segmentation challenge CSI 2016 or xVertSeg.v1 challenge. We have achieved promising results on our proposed technique especially on fractured cases. Dice score was found to be 96.4 ± 0.8% without fractured cases and 92.8 ± 1.9% with fractured cases in CSI 2014 dataset (lumber and thoracic). Similarly, dice score was 95.2 ± 1.9% on 15 CT dataset (with given ground truths) and 95.4 ± 2.1% on total 25 CT dataset for CSI 2016 datasets (with 10 annotated CT datasets). The proposed technique outperformed other state-of-the-art techniques and handled the fractured cases for the first time efficiently.

Fast Segmentation of Vertebrae CT Image Based on the SNIC Algorithm

“…Jiang Qiulin et al [ 1 ] proposed a segmentation algorithm based on the model and model, which has a good segmentation effect for thyroid ultrasound images with blurring boundaries and uneven gray levels. Rehman et al [ 2 ] proposed a new framework that combines traditional region-based level set algorithms and deep learning, which can accurately predict the shape of segmented vertebrae, and conducted experiments on multiple data sets, which has better performance in handling fracture cases. Chondro et al [ 3 ] proposed a lung segmentation algorithm based on statistical area growth and adaptive graph cutting.…”

“…Scholars have proposed many theories and methods for medical image segmentation, including based on traditional algorithms such as thresholds, region growth, level sets, and active contours. There are many methods [ 1 , 2 , 3 , 4 , 5 ] of segmentation recently, which have achieved good results in the segmentation accuracy, but most of them still have potential for improvement in segmentation speed. At present, the superpixel segmentation algorithm has developed in the image field [ 6 ] with a smaller calculation amount, faster-running speed, stronger anti-noise, and more robust, which is widely used in image segmentation and classification in various fields [ 7 , 8 ].…”

Research on multi-path dense networks for MRI spinal segmentation

“…Dominik GaweB et al [ 35 ] combined multiple stages of deep learning to recognize and separate different tissues of the human spine. Faisal Rehman1 et al [ 36 ] presented a novel combination of the traditional region-based level set with deep learning framework in order to predict shape of vertebral bones accurately. Martin Kolarík et al [ 37 ] designed a 3D Dense-U-Net neural network architecture implementing densely connected layers for high-resolution 3D volumetric segmentation of medical image data.…”

Bone Age Assessment Empowered with Deep Learning: A Survey, Open Research Challenges and Future Directions

“…Furthermore, the segmentation of the vertebral osteoporosis bones is difficult due to the complexity of the bone shape. A robust FU-net based model was proposed in [ 68 ] for the segmentation of vertebral bone. The U-net model is a U-shaped deep learning model that has contraction on the left side and expansion on the right side.…”