Laboratorio de Procesado de Imagen

(Image Processing Lab)
E.T.S.I. Telecomunicación
Campus Miguel Delibes s/n
Universidad de Valladolid
47011 Valladolid, Spain
contact@lpi.tel.uva.es


Preprint available: Microstructure Diffusion Scalar Measures from Reduced MRI Acquisitions

[19 Sep 2019]

We propose the so-called "Apparent Measures Using Reduced Acquisitions" (AMURA) to drastically reduce the number of samples needed for the estimation of q-space measures such as RTOP, RTAP and RTPP. AMURA avoids the calculation of the whole EAP by assuming the diffusion anisotropy is roughly independent from the radial direction.

Available to download as a preprint ar [bioRxiv]

PhD Thesis defended

[16 Sep 2019]

Last 13th September 2019 Iñali Rabanillo defended his PhD Thesis entitled "Artifact reduction in Magnetic Resonance Imaging: noise modelling in 2D/3D GRAPPA
accelerated acquisitions and motion-induced ghosting correction in multishot diffusion MRI
", getting a qualification of "Sobresaliente" and being proposed for "cum laude" distinction.

Un trabajo fin de grado realizado en el LPI consigue un 98% de acierto en el diagnóstico del TDAH

[30 Aug 2019]

El TFG realizado por Patricia Amado Caballero "Ayuda al diagnóstico del TDAH en la infancia mediante técnicas de procesado de señal y aprendizaje" ha dado lugar a un sistema que alcanza un 98% de acierto en el diagnóstico del TDAH. El TFG aplica técnicas de aprendizaje profundo (deep learning) para analizar firmas espectrales en patrones de movimiento. Para adquirir estos patrones se han utilizado pulseras de actividad que no interfieren para nada en la vida diaria del niño.

OpenCLIPER: An OpenCL-Based C++ Framework for Overhead-Reduced Medical Image Processing and Reconstruction on Heterogeneous Devices

OpenCLIPER performance vs. alternatives
[03 Jul 2019]

OpenCLIPER allows us to take advantage of any parallel computing device (CPUs, GPUs, FPGAs, DSPs, etc., as long as there is an OpenCL implementation for it), while developing medical imaging methods, but without the burden that OpenCL generally conveys. The work has been published in the IEEE Journal of Biomedical and Health Informatics journal.

Pages