SCOLIO-SEE

3D-IMAGE PROCESSING SYSTEM FOR HELPING PHYSICIANS IN THE DIAGNOSIS AND MONITORING OF SCOLIOSIS

 Partecipanti

Mappa

 Word cloud

Esplora la "nuvola delle parole (Word Cloud) per avere un'idea di massima del progetto.

 Obiettivo del progetto (Objective)

'Scoliosis is a three-dimensional (3D) deformity of the human spinal column and ribcage, caused by lateral curvature deviating from the midline and provoking acute pain, stomach and respiratory problems. This internal deformity is reflected in the external shape of the back; treatment aims at improving the health and the aesthetics of the patients. From the health point of view, the golden standard for diagnosing and monitoring scoliosis is the Cobb angle which refers to the internal curvature of the spine (by means of radiographs); from the external or aesthetic point of view there are a well-known variety of measurements and surface indices that quantify the severity of scoliosis. Many studies show that it is not possible to predict the Cobb angle by means of previous X-rays, nor to infer a concrete value of the Cobb angle by means only of surface indices, because radiograph and surface topography measurements are not redundant, but complementary. Nevertheless, although it is not possible to eliminate X-ray radiographs, their number can be reduced to a large extent by an adequate combination of surface and radiographic measurements; different studies show that the reduction in the number of radiographs is very important because patients with scoliosis have higher risk of cancer due to repeated radiation exposure. SCOLIO-SEE is a software and hardware system that, for the first time, will combine the data obtained by both radiographic and surface topography techniques; this matching between internal and external deformation will permit monitoring spine curvature evolution after treatment by analyzing only surface changes, reducing the number of radiographs needed to a level that it is not harmful for the patients. Additionally, a software module will be developed to compare the 3D images taken for each individual patient in two successive visits, to show the 3D-changes occurred in the whole back surface occurred in this time interval.'

Introduzione (Teaser)

Scoliosis affects millions of people throughout the EU, significantly impacting their health and self-esteem. EU-funded researchers are developing a 3D image processing system for diagnosing and monitoring scoliosis.

Descrizione progetto (Article)

An abnormal spinal curvature, scoliosis is most often treated using a combination of bracing, exercise and surgery based on severity. Current diagnostic and monitoring protocols involve the use of an inclinometer followed by X-rays to measure and quantify the severity of spinal curvature. Both the complexity of quantifying scoliosis grade and the need for frequent X-rays during treatment have highlighted the need for a better technique.

Research organisations and small and medium-sized enterprises (SMEs) joined forces under the '3D-image processing system for helping physicians in the diagnosis and monitoring of scoliosis' (http://www.scolio-see.eu/ (SCOLIO-SEE)) project. They will develop an integrated software and hardware system for 3D image processing of the spine using machine learning to connect internal and external spinal deformation. Their objective is better quantification of scoliosis grade and minimising the need for X-rays.

Project members worked on determining the SCOLIO-SEE system specifications and relevant parameters during the first part of the project. Artificial neural network (ANN) algorithms will be applied to this data for accurate estimation and prediction of extent of spinal curvature for diagnosis and treatment.

A major achievement is the development of the Formetric measurement technology prototype that is undergoing evaluation during the project's lifetime. Already, over 100 patients have been recruited for the study.

Significant progress has been achieved with regard to database development, as well as collecting and combining surface topographic and radiographic spinal measurements using ANN methodology. Researchers are currently working on mapping the spinal X-ray images with the 3D biomechanical model from the Formetric imager. Data of 10 human back surfaces is being used to evaluate and compare 3D images acquired at different time-points.

Successful project outcomes would take us a step closer to commercialising a non-invasive, safe and affordable diagnostic and monitoring system for scoliosis. Furthermore, this system could also be adapted for other parts of the body. Besides improving the quality of life of patients, the competitiveness of consortium SMEs would also be enhanced.