Samuel Kadoury1, Hubert Labelle. 1. Philips Research North America, 345 Scarborough Rd, Briarcliff Manor, NY 10510, USA. samuel.kadoury@philips.com
Abstract
PURPOSE: Understanding how to classify and quantify three-dimensional (3D) spinal deformities remains an open question in adolescent idiopathic scoliosis. The objective of this study was to perform a 3D manifold characterization of scoliotic spines demonstrating thoracic deformations using a novel geometric and intuitive statistical tool to determine patterns in pathological cases. METHODS: Personalized 3D reconstructions of thoracic (T)/lumbar (L) spines from a cohort of 170 Lenke Type-1 patients were analyzed with a non-linear manifold embedding algorithm in order to reduce the high-dimensionality of the data, using statistical properties of neighbouring spine models. We extracted sub-groups of the data from the underlying manifold structure using an unsupervised clustering algorithm to understand the inherent distribution and determine classes of pathologies which appear from the low-dimensional space. RESULTS: For Lenke Type-1 patients, four clusters were detected from the low-dimensional manifold of 3D models: (1) normal kyphosis (T) with hyper-lordosis (L) and high Cobb angles (37 cases), (2) low kyphosis (T) and normal lordosis (L), with high rotation of plane of maximum curvature (55 cases), (3) hypo-kyphotic (T) and hyper-lordosis (L) (21 cases) and (4) hyper-kyphotic (T) with strong vertebral rotation (57 cases). Results show the manifold representation can potentially be useful for classification of 3D spinal pathologies such as idiopathic scoliosis and serve as a tool for understanding the progression of deformities in longitudinal studies. CONCLUSIONS: Quantitative evaluation illustrates that the complex space of spine variability can be modeled by a low-dimensional manifold and shows the existence of an additional hyper-kyphotic subgroup from the cohort of 3D spine reconstructions of Lenke Type-1 patients when compared with previous findings on the 3D classification of spinal deformities.
PURPOSE: Understanding how to classify and quantify three-dimensional (3D) spinal deformities remains an open question in adolescent idiopathic scoliosis. The objective of this study was to perform a 3D manifold characterization of scoliotic spines demonstrating thoracic deformations using a novel geometric and intuitive statistical tool to determine patterns in pathological cases. METHODS: Personalized 3D reconstructions of thoracic (T)/lumbar (L) spines from a cohort of 170 Lenke Type-1patients were analyzed with a non-linear manifold embedding algorithm in order to reduce the high-dimensionality of the data, using statistical properties of neighbouring spine models. We extracted sub-groups of the data from the underlying manifold structure using an unsupervised clustering algorithm to understand the inherent distribution and determine classes of pathologies which appear from the low-dimensional space. RESULTS: For Lenke Type-1patients, four clusters were detected from the low-dimensional manifold of 3D models: (1) normal kyphosis (T) with hyper-lordosis (L) and high Cobb angles (37 cases), (2) low kyphosis (T) and normal lordosis (L), with high rotation of plane of maximum curvature (55 cases), (3) hypo-kyphotic (T) and hyper-lordosis (L) (21 cases) and (4) hyper-kyphotic (T) with strong vertebral rotation (57 cases). Results show the manifold representation can potentially be useful for classification of 3D spinal pathologies such as idiopathic scoliosis and serve as a tool for understanding the progression of deformities in longitudinal studies. CONCLUSIONS: Quantitative evaluation illustrates that the complex space of spine variability can be modeled by a low-dimensional manifold and shows the existence of an additional hyper-kyphotic subgroup from the cohort of 3D spine reconstructions of Lenke Type-1patients when compared with previous findings on the 3D classification of spinal deformities.
Authors: L P D'Andrea; R R Betz; L G Lenke; D H Clements; T G Lowe; A Merola; T Haher; J Harms; G K Huss; K Blanke; S McGlothlen Journal: Spine (Phila Pa 1976) Date: 2000-07-15 Impact factor: 3.468
Authors: Archana P Sangole; Carl-Eric Aubin; Hubert Labelle; Ian A F Stokes; Lawrence G Lenke; Roger Jackson; Peter Newton Journal: Spine (Phila Pa 1976) Date: 2009-01-01 Impact factor: 3.468
Authors: Samuel Kadoury; Farida Cheriet; Marie Beauséjour; Ian A Stokes; Stefan Parent; Hubert Labelle Journal: Eur Spine J Date: 2008-11-13 Impact factor: 3.134
Authors: Edgar García-Cano; Fernando Arámbula Cosío; Luc Duong; Christian Bellefleur; Marjolaine Roy-Beaudry; Julie Joncas; Stefan Parent; Hubert Labelle Journal: Med Biol Eng Comput Date: 2018-06-09 Impact factor: 2.602
Authors: Wei Wei Jiang; Connie Lok Kan Cheng; Jason Pui Yin Cheung; Dino Samartzis; Kelly Ka Lee Lai; Michael Kai Tsun To; Yong Ping Zheng Journal: Eur Spine J Date: 2018-06-25 Impact factor: 3.134
Authors: Zongshan Hu; Claudio Vergari; Laurent Gajny; Zhen Liu; Tsz-Ping Lam; Zezhang Zhu; Yong Qiu; Gene C W Man; Kwong-Hang Yeung; Winnie C W Chu; Jack C Y Cheng; Wafa Skalli Journal: Quant Imaging Med Surg Date: 2021-07
Authors: Mingjie Yang; Cheng Zeng; Song Guo; Jie Pan; Yingchao Han; Zeqing Li; Lijun Li; Jun Tan Journal: PLoS One Date: 2014-08-26 Impact factor: 3.240