Research topics
Applied AI

The effectiveness of a cardiac MR scan depends on the ability of the operator to correctly tune the acquisition parameters to the subject being scanned and on the potential occurrence of imaging artefacts such as cardiac and respiratory motion.

Together with colleagues at Imperial College, we developed a learning-based, fully-automated, AI-based quality control pipeline for cardiac MR short-axis image stacks. Our pipeline performs three important quality checks: 1) heart coverage estimation (to detect if the whole left ventricle, LV, has been imaged), 2) inter-slice motion detection (to estimate potential slice misalignments in the stack caused by respiratory motion and correct them), 3) image contrast estimation in the cardiac region (to estimate intensity differences between the LV cavity and myocardium). The pipeline was successfully validated on 3’000 cardiac MR scans of the UK Biobank dataset. It is currently being deployed on a larger UK Biobank cohort (~30’000) scans to detect potential factors associated with lower quality scans.

Featured publications:

Tarroni, G., Oktay, O., Bai, W., Schuh, A., Suzuki, H., Passerat-Palmbach, J. … Rueckert, D. (2019). Learning-based quality control for cardiac MR images. IEEE Transactions on Medical Imaging, 38(5), pp. 1127–1138. doi:10.1109/TMI.2018.2878509.

Tarroni, G., Oktay, O., Sinclair, M., Bai, W., Schuh, A., Suzuki, H. … Rueckert, D. (2018). A comprehensive approach for learning-based fully-automated inter-slice motion correction for short-axis cine cardiac MR image stacks. https://arxiv.org/abs/1810.02201.

We are interested in the application of Recurrent Neural Networks to control connected converters in electric power systems that include smart grids, renewable energy resources and energy storage devices. In collaboration with the University of Alabama and the Missouri S&T, we have implemented a new Forward Accumulation Through Time (FATT) mechanism that is fed into the Levenberg-Marquardt algorithm to train the network, plus a novel input structure. This combination was compared against optimal controllers in close to real-life power converter switching environments. The excellent results indicate the feasibility of using our solution to approximate optimal control in practical applications

Featured publlication:

Fu, X., Li, S., Fairbank, M., Wunsch, D. and Alonso, E. (2015). Training Recurrent Neural Networks with the Levenberg–Marquardt Algorithm for Optimal Control of a Grid-Connected Converter. IEEE Transactions on Neural Networks and Learning Systems, 26(9), pp. 1900–1912. doi:10.1109/TNNLS.2014.2361267.

A wide range of cardiovascular diseases has been long associated with alterations in the shape of the LV myocardium (a phenomenon known as cardiac remodelling). However, current clinical methods to detect these diseases often rely on the visual assessment of cardiac images and on the estimation of global indices (e.g. cardiac volumes and ejection fraction). In recent years, deep learning approaches have achieved remarkable success in the classification of medical images, but typically lack interpretability in the feature extraction and decision processes, limiting their value in clinical diagnosis.

Together with colleagues at Imperial College, we exploited a 3D convolutional Variational Autoencoder (VAE) for the automated classification of 3D high-resolution LV cardiac shapes. Differently from classification networks, our approach allows the visualization and the quantification of the learned disease-specific remodelling patterns, which are of great clinical value.

Feature publication:

We propose an end-to-end deep learning architecture that automatically classifies a subject as ADHD or healthy control and demonstrates the importance of functional connectivity to increase classification accuracy and provide interpretable results. The proposed method, called DeepFMRI, is comprised of three sequential networks, namely, a feature extractor, a functional connectivity network, and a classification network. The model takes fMRI pre-processed time-series signals as input and outputs a diagnosis and is trained end-to-end using backpropagation. Experimental results on the publicly available ADHD-200 dataset demonstrate that this innovative method outperforms previous state-of-the-art. We are interested in applying the model to study other clinical conditions such as autism.

Featured publication:

Riaz, A., Asad, M., Alonso, E. and Slabaugh, G. (2019). DeepFMRI: End-to-end deep learning for functional connectivity and classification of ADHD using fMRI. To appear in Journal of Neuroscience Methods.

We combine data analysis techniques, analytical solutions of mathematical models, and computational simulations to study fundamental aspects of how the brain encodes task-related information of sensory stimuli and uses it for perceptual decision-making.

We have used a combination of signal detection theory methods, generalized linear models, and machine learning classification methods to determine how neuronal responses in sensory areas of the brain are modulated by top-down feedback signals from higher cognitive areas, which are related to behavioral choices. Characterizing the distribution of this feedback across cells, as well as the role of attentional fluctuations, we contributed to identify how interactions between sensory and behavioral neuronal responses reflect the structure of the neural code. Furthermore, we also studied how perceptual abilities are modified during perceptual learning periods. We use machine learning classification methods to determine changes in the tuning of single neuron responses and how sensory discrimination is enhanced by adaptation of the correlation structure of neuronal populations.

Featured publications:

Chicharro D, Panzeri S, Haefner RM. (2021) Stimulus dependent relationships between behavioral choice and sensory neural responses. eLife 10:e54858

Sanayei M, Chen X, Chicharro D, Distler C, Panzeri S, Thiele A. (2018) Perceptual learning of fine contrast discrimination changes neuronal tuning and population coding in macaque V4. Nature Communications 9(1):4238.

We investigate the question of how human brain activity coordinates the linkage of sensory information to actions, studying how neural activity distributed over large-scale brain networks is modulated during the execution of human cognitive functions.

We study the structural and functional connectivity between brain areas and the modulation of information routing associated with specific visuomotor mapping tasks. We used noninvasive recordings of human brain activity such as magnetoencephalographic (MEG) recordings and structural MRI to reveal the cortical networks and corticocortical functional connectivity mediating visuomotor mappings. We characterized information flows between brain areas using a combination of time-series spectral analysis methods, predictive models, and network theory measures. We also use computational models to study how the causal interactions occurring at the level of single neurons are reflected at the macroscopic scale of the MEG and MRI recordings, which capture the aggregated activity of populations of neurons.

Featured publications:

Brovelli A, Chicharro D, Badier JM, Wang H, Jirsa V. (2015) Characterization of cortical networks and corticocortical functional connectivity mediating arbitrary visuomotor mapping. Journal of Neuroscience 35(37):12643-58.

Chicharro D, Panzeri S. (2014) Algorithms of causal inference for the analysis of brain effective connectivity among brain regions. Frontiers in Neuroinformatics 8:64.