Content area
Abstract
Among the first signs of scoliosis, the rib hump is a cosmetic deformity that cannot be assessed on radiographs, nor on a 3D reconstruction of the spine. It is mainly associated with rib cage deformity. It is therefore intuitive to suppose that the axial rotations of the ribs and of the back surface are highly correlated. Nevertheless, previous works have failed to demonstrate a strong relationship between these measurements. This might be explained by the limited accuracy of the technique used for the 3D reconstruction of the ribs. Consequently, in this work, a novel method for the 3D reconstruction of scoliotic ribs is proposed. It requires a pair of standard frontal and sagittal radiographs, which are usually acquired for the 3D reconstruction of the spine. The proposed method is evaluated on 13 patients with scoliosis and compared to the existing reconstruction technique. Results obtained using the new method show a strong correlation between the ribs' and the back surface's axial rotations, which leads to the conclusion that the back surface rotation allows clinicians to quantify the rib hump without resorting to X-rays.
Besides the rib hump, the trunk of a scoliotic patient reveals other deformations. In order to provide a complete scoliosis assessment by taking patients' concerns more into account, it is necessary to consider all of the cosmetic deformities of the trunk. Thus, a new set of trunk shape features is introduced. Thanks to its original functional representation, this new index allows a local description of the deformation at all trunk levels and not only at its apex. The proposed index includes complementary measurements taken in the three planes: the back surface axial rotation, the trunk deviation in the frontal plane and the trunk deviation in the sagittal plane. The reliability of this index has been demonstrated on a cohort of 32 scoliotic patients. For each patient, the surface of the trunk was acquired twice with patient repositioning. The reliability study also helped define, for each trunk level, a minimum interval beyond which a difference between two trunk acquisitions is significant. This notably allows for radiography-free monitoring of trunk deformities and assessment of the cosmetic outcome of the scoliosis treatment.
The new index was then used to prove the existence of natural clusters of trunk surfaces in a cohort of 236 individuals, comprised of healthy subjects and scoliosis patients. The resulting clusters constitute an original classification of scoliotic trunk deformities, varying according to the type and the severity of the deformations and obtained using a non-supervised approach, unlike existing scoliosis trunk classifications. A classification of the different types of scoliotic trunks helps to define a common vocabulary for clinicians and to identify for each class a distinct pattern of progression, and eventually, a suitable management. The 11 clusters obtained are consistent in 68% of cases with the spinal curve type as assessed on a frontal radiograph. In addition, since the number of resulting clusters is greater than the number of spinal curve types considered, the proposed new classification provides complementary information to the current classification of spinal curves.
A correlation analysis between the new trunk measurements and those of the spine in the three planes shows a strong relation between the trunk and the spinal lateral deviation in the frontal plane, and to a lesser extent in the sagittal plane. However, in the axial plane, no significant correlation was demonstrated between the rotations of the vertebrae and of the back surface. Thus, it can be concluded that the new trunk deformity index could be used in the context of diagnosis and monitoring to document the rib hump and the spinal deviations in the coronal and sagittal planes, while reducing patients' exposure to X-rays.
In conclusion, this thesis proposes a new and reliable multi-level 3D index for quantitative and radiation-free assessment of cosmetic trunk deformities, which constitute the main concern for young scoliotic patients. It can replace the scoliometer and plumb line used for the clinical assessment of patients with scoliosis. It can also help to reduce patient exposure to X-rays, especially during clinical monitoring. The use of this index in clinical practice will be a first meaningful step towards a more complete management of scoliotic deformities, thereby reducing the gap between what patients perceive and what clinicians are able to assess. (Abstract shortened by UMI.)
