ABSTRACT
High-intensity focused ultrasound (HIFU) is a minimally invasive therapy modality in which ultrasound beams are concentrated at a focal region, producing a rise of temperature and selective ablation within the focal volume and leaving surrounding tissues intact. HIFU has been proposed for the safe ablation of both malignant and benign tissues and as an agent for drug delivery. Magnetic resonance imaging (MRI) has been proposed as guidance and monitoring method for the therapy. The identification of regions of interest is a crucial procedure in HIFU therapy planning. This procedure is performed in the MR images. The purpose of the present research work is to implement a time-efficient and functional segmentation scheme, based on the watershed segmentation algorithm, for the MR images used for the HIFU therapy planning. The achievement of a segmentation process with functional results is feasible, but preliminary image processing steps are required in order to define the markers for the segmentation algorithm. Moreover, the segmentation scheme is applied in parallel to an MR image data set through the use of a thread pool, achieving a near real-time execution and making a contribution to solve the time-consuming problem of the HIFU therapy planning.
Subject(s)
High-Intensity Focused Ultrasound Ablation , Magnetic Resonance Imaging, Interventional/methods , Algorithms , Humans , Magnetic Resonance Imaging, Interventional/instrumentationABSTRACT
Resumen: La Parálisis Cerebral (PC) es un grupo de trastornos pre, post y perinatales permanentes del desarrollo, movimiento y postura debidos a alteraciones no progresivas ocurridas durante el desarrollo cerebral, producto de lesiones del Sistema Nervioso Central. Debido a la importancia del uso del miembro superior en las actividades de la vida diaria, es importante considerar formas eficientes de medir el desempeño motor de este miembro en los pacientes con PC. Una forma de obtener la evaluación del miembro torácico es grabando movimientos definidos y calculando la suavidad de los mismos, utilizando un tablero seleccionador de figuras instrumentado. Nuestro objetivo es desarrollar un protocolo de valoración para el miembro superior, que a su vez sea objetivo, eficiente y que otorgue una medición cuantitativa del grado de afectación motora de los niños con PC en un entorno clínico.
Abstract: Cerebral Palsy (CP) is a group of permanent pre, post and perinatal disorders of the motor and posture development due to non-progressive alterations in brain's natural development caused by injuries in the Central Nervous System. Due to the importance of the daily use of the upper limb members, it's important to consider more efficient ways to evaluate the performance in patients diagnosed with CP. One way to obtain an evaluation of the performance of the thoracic member is recording defined movements and calculating the smoothness, using an instrumented sorting block box. Our objective is to create a protocol of valuation for the upper member that is objective, efficient and that gives a quantitative feedback of the grade of the motor affectation of child with PC in a clinical environment.
ABSTRACT
We investigated the use of a sorting box to obtain a quantitative assessment of upper limb motor function in children with cerebral palsy. In our study, children with and without cerebral palsy placed and removed geometrical objects of a sorting-box while their wrist position was monitored by a camera-based, motion-tracking system. We analyzed three different smoothness metrics (logarithmic dimensionless jerk, spectral arc-length and number of peaks) together with time to task completion. Our results suggest that smoothness metrics are an effective tool to distinguish between impaired and non-impaired subjects, as well as to quantify differences between the affected and less-affected sides in children with hemiparetic cerebral palsy.
