A Support Vectors Classifier Approach to Predicting the Risk of Progression of Adolescent Idiopathic Scoliosis

Dept. of Electr. & Comput. Eng., Univ. of Alberta, Edmonton, Alta., Canada
IEEE Transactions on Information Technology in Biomedicine (Impact Factor: 2.49). 07/2005; 9(2):276 - 282. DOI: 10.1109/TITB.2005.847169
Source: IEEE Xplore


A support vector classifier (SVC) approach was employed in predicting the risk of progression of adolescent idiopathic scoliosis (AIS), a condition that causes visible trunk asymmetries. As the aetiology of AIS is unknown, its risk of progression can only be predicted from measured indicators. Previous studies suggest that individual indicators of AIS do not reliably predict its risk of progression. Complex indicators with better predictive values have been developed but are unsuitable for clinical use as obtaining their values is often onerous, involving much skill and repeated measurements taken over time. Based on the hypothesis that combining common indicators of AIS using an SVC approach would produce better prediction results more quickly, we conducted a study using three datasets comprising a total of 44 moderate AIS patients (30 observed, 14 treated with brace). Of the 44 patients, 13 progressed less than 5° and 31 progressed more than 5°. One dataset comprised all the patients. A second dataset comprised all the observed patients and a third comprised all the brace-treated patients. Twenty-one radiographic and clinical indicators were obtained for each patient. The result of testing on the three datasets showed that the system achieved 100% accuracy in training and 65%-80% accuracy in testing. It outperformed a "statistically equivalent" logistic regression model and a stepwise linear regression model on the said datasets. It took less than 20 min per patient to measure the indicators, input their values into the system, and produce the needed results, making the system viable for use in a clinical environment.

1 Follower
9 Reads
  • Source
    • "In the last decade, SVM learning has found a wide range of applications [19], including image segmentation [20] and classification [21], object recognition [22], image fusion [23], and stereo correspondence [24]. More recently, SVMs have been employed in several applications in biomedicine: gait degeneration due to age [25], EEG signal classification [26], brain computer interfacing (BCI) [27], [28], analysis and prediction of scoliosis [29], [30], electrogastrogram analysis [31], and color Doppler echocardiography [32]. Relevant studies involving SVM and heart rate time series are the Hermite characterization of QRS complex [33], where a heartbeat is characterized as normal or abnormal and the detection of risky situations for fetal assessment [34]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: In this study, heartbeat time series are classified using support vector machines (SVMs). Statistical methods and signal analysis techniques are used to extract features from the signals. The SVM classifier is favorably compared to other neural network-based classification approaches by performing leave-one-out cross validation. The performance of the SVM with respect to other state-of-the-art classifiers is also confirmed by the classification of signals presenting very low signal-to-noise ratio. Finally, the influence of the number of features to the classification rate was also investigated for two real datasets. The first dataset consists of long-term ECG recordings of young and elderly healthy subjects. The second dataset consists of long-term ECG recordings of normal subjects and subjects suffering from coronary artery disease.
    IEEE transactions on information technology in biomedicine: a publication of the IEEE Engineering in Medicine and Biology Society 09/2008; 13(4):512-8. DOI:10.1109/TITB.2008.2003323 · 2.49 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Clinical and health-care knowledge management (KM) as a discipline has attracted increasing worldwide attention in recent years. The approach encompasses a plethora of interrelated themes including aspects of clinical informatics, clinical governance, artificial intelligence, privacy and security, data mining, genomic mining, information management, and organizational behavior. This paper introduces key manuscripts which detail health-care and clinical KM cases and applications.
    IEEE Transactions on Information Technology in Biomedicine 06/2005; 9(2):157-161. DOI:10.1109/TITB.2005.849395 · 2.49 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Spinal deformities are diagnosed using posteroanterior (PA) radiographs. Automatic detection of the spine on conventional radiographs would be of interest to quantify curve severity, would help reduce observer variability and would allow large-scale retrospective studies on radiographic databases. The goal of this paper is to present a new method for automatic detection of spinal curves from a PA radiograph. A region of interest (ROI) is first extracted according to the 2-D shape variability of the spine obtained from a set of PA radiographs of scoliotic patients. This region includes 17 bounding boxes delimiting each vertebral level from T1 to L5. An adaptive filter combining shock with complex diffusion is used to individually restore the image of each vertebral level. Then, texture descriptors of small block elements are computed and submitted for training to support vector machines (SVM). Vertebral body's locations are thereby inferred for a particular vertebral level. The classifications of block elements for all 17 SVMs are identified in the image and a voting system is introduced to cumulate correctly predicted blocks. A spline curve is then fitted through the centers of the predicted vertebral regions and compared to a manual identification using a Student t-test. A clinical validation is performed using 100 radiographs of scoliotic patients (not used for training) and the detected spinal curve is found to be statistically similar (p < 0.05) in 93% of cases to the manually identified curve.
    IEEE transactions on bio-medical engineering 02/2010; 57(5):1143-51. DOI:10.1109/TBME.2009.2037214 · 2.35 Impact Factor
Show more