Research

Our research interest are varied and span several medicine use-cases.

An overview of our expertise and interests in Medicine and Artificial Intelligence.

We have extensive experience in working with MRI images (2D and 3D), with the software used and in how to build computer vision pipelines to work on large and small image datasets. In Figure 1 you can see an example of the results of a pipeline we built (by using our internal code) to separate different parts of the brain and skull from MRI images.

We bring together different skills: medical data analysis, machine learning, statistics and software engineering. We have also an extensive network of partners that can support in different phases of research projects (for example web application development, or integration of algorithms in an extensive and existing software landscape).

In Figure 2 you can see an animation of hands and where the eyes are looking from a guided reaching task done with the Kinarm exoskeleton. We developed dedicated python libraries to extract the information from proprietary files saved by the exoskeleton. The library is freely available on GitHub. The information visualised is not available out of the box in standard reports.

Here is a list of projects and interests that is not exhaustive.

• MRI imaging (2D and 3D): to link brain features to development disorders in children (typically developed and pre-term).

  Partners: KU Leuven (BE), University of Cadiz (ES), University of Ljubljana (SI), Politecnico di Torino (IT).

• Eye-gaze: data in structured coordination tests to try to understand how hands-coordination is linked to eye-movements in children with brain development disorders.

  Partners: KU Leuven (BE), University of Ljubljana (SI).

• Psychology: We work with data coming from phsychology assessments to determine how families are coping with children with disabilities.

  Partners: KU Leuven (BE).

• ECG Data: We work with different kind of ECG data

  Partners: Politecnico di Torino (IT).

• Cardiology data: We have worked with various Cardiology datasets (F.D’Ascenzo et al., 2021) to predict onset of cardiological conditions

  Partners: Politecnico di Torino (IT).

• We have built models and studied COVID-19

  Partners: Istituto dalle Molle for artificial Intelligence (IDSIA) (CH), Politecnico di Torino (IT), cardiology department of the Guglielmo Da Saliceto Hospital in Italy (IT).

• We have developed Python libraries to work with the Kinarm exoskeleton lab (US) that is freely available on GitHub.

DETERMINED 2022 - International Workshop in Neurodevelopmental Impairments in Preterm Children - Computational Advancements

Umberto Michelucci has edited (and authored two of the papers) the proceedings of the DETERMINED2022 conference in Ljubljana.

Proceedings of the International Workshop in Neurodevelopmental Impairments in Preterm Children - Computational Advancements (DETERMINED 2022): https://ceur-ws.org/Vol-3363/

Publications

References cited in this page

1. F.D’Ascenzo, Filippo, O. D., Gallone, G., Mittone, G., Deriu, M. A., Iannaccone, M., Ariza-Solé, A., Liebetrau, C., Manzano-Fernández, S., Quadri, G., Kinnaird, T., Campo, G., Henriques, J. S., Hughes, J., Dominguez-Rodriguez, A., Aldinucci, M., Morbiducci, U., Patti, G., Raposeiras-Roubin, S., … Arfat, Y. (2021). Machine learning-based prediction of adverse events following an acute coronary syndrome (PRAISE): a modelling study of pooled datasets. The Lancet, 397(10270), 199–207.
   • [Bibtex]
   • [Link]

   @article{2021machine,
     title = {Machine learning-based prediction of adverse events following an acute coronary syndrome (PRAISE): a modelling study of pooled datasets},
     author = {F.D'Ascenzo and Filippo, O. De and Gallone, G. and Mittone, G. and Deriu, M.A. and Iannaccone, M. and Ariza-Solé, A. and Liebetrau, C. and Manzano-Fernández, S. and Quadri, G. and Kinnaird, T. and Campo, G. and Henriques, J. S. and Hughes, J. and Dominguez-Rodriguez, A. and Aldinucci, M. and Morbiducci, U. and Patti, G. and Raposeiras-Roubin, S. and Abu-Assi, E. and Ferrari, G.M. De and Piroli, F. and Saglietto, A. and Conrotto, F. and Omedé, P. and Montefusco, A. and Pennone, M. and Bruno, F. and Bocchino, P. Paolo and Boccuzzi, G. and Cerrato, E. and F. and Varbella, O. and Sperti, M. and Wilton, S. B. and Velicki, L. and Xanthopoulou, I. and Cequier, A. and Iniguez-Romo, A. and Pousa, I. Munoz and Fern, M. Cespon and Queija, B. Caneiro and Cobas-Paz, R. and Lopez-Cuenca, A. and Garay, A. and Blanco, P. Flores and Rognoni, A. and Zoccai, G. Biondi and Biscaglia, S. and Nunez-Gil, I. and Fujii, T. and A. and ro Durante and Song, X. and Kawaji, T. and Alexopoulos, D. and Huczek, Z. and Juanatey, J. R. G. and Nie, S.-P. and Kawashiri, M.-aki and Colonnelli, I. and Cantalupo, B. and Esposito, R. and Leonardi, S. and Marra, W. G. and Chieffo, A. and Michelucci, U. and Piga, D. and Malavolta, M. and Gili, S. and Mennuni, M. and Montalto, C. and Visconti, L. O. and Arfat, Y.},
     journal = {The Lancet},
     volume = {397},
     number = {10270},
     pages = {199--207},
     year = {2021},
     bibtex_show = {true},
     abbr = {Medicine},
     topic = {medicine},
     link = {https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)32519-8/fulltext}
   }
   

Partners