Supervised and unsupervised learning to classify scoliosis and healthy subjects based on non-invasive rasterstereography analysis

Tommaso Colombo (), Massimiliano Mangone (), Andrea Bernetti (), Marco Paoloni (), Valter Santilli () and Laura Palagi
Additional contact information
Tommaso Colombo: Department of Computer, Control and Management Engineering Antonio Ruberti (DIAG), University of Rome La Sapienza, Rome, Italy
Massimiliano Mangone: Department of Physical Medicine and Rehabilitation, University of Rome La Sapienza, Rome, Italy
Andrea Bernetti: Department of Physical Medicine and Rehabilitation, University of Rome La Sapienza, Rome, Italy
Marco Paoloni: Department of Physical Medicine and Rehabilitation, University of Rome La Sapienza, Rome, Italy
Valter Santilli: Department of Physical Medicine and Rehabilitation, University of Rome La Sapienza, Rome, Italy

No 2019-08, DIAG Technical Reports from Department of Computer, Control and Management Engineering, Universita' degli Studi di Roma "La Sapienza"

Abstract: Objective. Classify scoliosis versus healthy patients using rasterstereography non invasive surface acquisition, without prior knowledge from X-ray data.Methods. Data acquisition via rasterstereography; unsupervised learning for clustering and supervised learning for predicting models. Comparison among Support Vector Machine and Deep Network architectures. K-fold cross validation procedure for assessing the results.Results. The accuracy and the balanced accuracy of the best supervised model was close to 85%. Classification rates per class were measured using confusion matrix giving low percentage of misclassified patients.Conclusion. Rasterstereography turns out to be a good tool to identify scoliosis vs healthy patients with the advantage of not exposing patient to unhealthy X-Ray. Furthermore, thanks to the portability and the low cost of the rasterstereography, it is possible to use it to promote screening campaign.

Keywords: Data Mining; Rasterstereography; Non invasive support system; Scoliosis diagnosis; Support Vector Machine; Deep Learning (search for similar items in EconPapers)
Date: 2019
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://users.diag.uniroma1.it/~biblioteca/sites/de ... ocuments/2019-08.pdf First version, 2019 (application/pdf)
Our link check indicates that this URL is bad, the error code is: 403 Forbidden

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:aeg:report:2019-08

Access Statistics for this paper

More papers in DIAG Technical Reports from Department of Computer, Control and Management Engineering, Universita' degli Studi di Roma "La Sapienza" Contact information at EDIRC.
Bibliographic data for series maintained by Antonietta Angelica Zucconi ( this e-mail address is bad, please contact ).