2024 - Deep learning for volumetric assessment of primary lung cancer on computed tomography
Title: Deep learning for volumetric assessment of primary lung cancer on computed tomography
Authors: Liliana Petrychenko, Miriam Cobo
Coauthors: Regina Beets-Tan, Kevin Groot Lipman, Laurens Topff
2024 - Deep Learning and Explainability for Multimodal Medical Data
Title: Deep Learning and Explainability for Multimodal Medical Data
Authors: Miriam Cobo
Coauthors: Wilson Silva, Lara Lloret Iglesias
2023/07 - Deep learning to detect coronary artery tortuosity in coronary angiography
Escuela de verano AIHUB CSIC
Author: Miriam Cobo Cano.
July 3 - 7 2023. CaixaForum Macaya Barcelona, Spain.
2023/07 - Application of federated learning to medical imaging scenarios
AIHUB CSIC Summer School 2023
Authors: Judith Sáinz-Pardo Díaz and Álvaro López García.
July 3 - 7 2023. Barcelona, Spain.
2023/05 - Reliable machine learning algorithms for medical imaging
X Jornadas doctorales y V jornadas de divulgación. Grupo de Universidades G9
Authors: Miriam Cobo Cano, Pablo Menéndez Fernández-Miranda (codirector), Lara Lloret Iglesias (director).
May 31 - 2 June 2023. University of Oviedo, Spain.
2023/05 - Privacy preserving techniques for data science
X Jornadas doctorales y V jornadas de divulgación. Grupo de Universidades G9
Authors: Judith Sáinz-Pardo Díaz and Álvaro López García (director).
May 31 - 2 June 2023. University of Oviedo, Spain.
2022/09 - pyCANON: A Python library to check the level of anonymity of a dataset
The unstoppable improvements in data analysis techniques for knowledge extraction and decision-making make necessary the evolution of techniques for the secure publication of data. Moreover, the need for collaboration between different institutions, research centers or companies makes it necessary to be able to share data with certain security guarantees. There are numerous attacks that can be carried out on databases: re-identification, linkage, skewness and semantic attacks among others. For this the implementation of pyCANON, a Python library and CLI that can be used to know the level of anonymity of a dataset (and thus publish or share it while being aware of the risks involved), is presented. Nine different techniques will be used for this purpose.
2nd Inria-DFKI European Summer School on AI, IDESSAI 2022.
August 29 - September 2, 2022. Saarbrücken, Germany.
2022/09 - Frouros: A Python library for drift detection in Machine Learning problems
In real-world ML problems, models tend to suffer some degradation in terms of performance over time, due to changes in the concept of what is first learned by the model (concept drift), or due to significant changes in the variables used by the model (data/covariate drift). We present Frouros, a Python library capable of detecting drift using both supervised and unsupervised drift detection methods, and which can be easily integrated with scikit-learn.
2nd Inria-DFKI European Summer School on AI, IDESSAI 2022.
August 29 - September 2, 2022. Saarbrücken, Germany.
2022/06 - SQAaaS - Ensuring Software, Service and Data quality automatically in open repositories
SQAaaS - Ensuring Software, Service and Data quality automatically in open repositories
Authors: Fernando Aguilar
Coauthors: Pablo Orviz, Isabel Bernal
Conference: Open Repositories 2022, Denver, Colorado (USA) 6-10 junio 2022
URL: https://or2022.openrepositories.org/
2020/07 - The ethical and legal issues arising from the implantation of AI systems in the radiological practice
Title: The ethical and legal issues arising from the implantation of AI systems in the radiological practice
Authors: Menéndez Fernández-Miranda, Pablo ;Sanz Bellón, Pablo ;Pérez del Barrio, Amaia ;Marco de Lucas, Enrique ;Rodríguez González, David ;Lloret Iglesias, Lara ;Vega Álvarez, José Antonio
Conference: European Congress of Radiology-ECR 2020
2020/07 - Radiologists in the regulation of the responsibilities of artificial intelligence applied to radiodiagnosis: Let's make it real.
Title: Radiologists in the regulation of the responsibilities of artificial intelligence applied to radiodiagnosis: Let's make it real
Authors: Sanz Bellón, Pablo ;Pérez del Barrio, Amaia ;Menéndez Fernández-Miranda, Pablo ;Pérez, J. Domingo ;Lloret Iglesias, Lara ;Rodríguez González, David
Conference: European Congress of Radiology-ECR 2020
2019/04 - Deep Learning techniques for Satellite data: Atmospheric Corrections
Deep Learning techniques for Satellite data: Atmospheric Corrections
Remote sensing is a technique that allows obtaining information about an object through the analysis of data acquired by sensors that are not in contact with it, and the detection of changes from a multi-temporal sequence of satellite data is one of its most important applications.
As is well known, this process requires the application of atmospheric corrections to the images, so that the changes detected are only attributable to real landscape modifications. One of the techniques often recommended for applications of classification and detection of changes is the Dark Object Subtraction (DOS).
The technological advances in the sensors and the spatial and temporal resolutions now available open the door to new ways of monitoring the Earth by applying the latest techniques in machine learning or deep learning to these data.
EGU ( European Geosciences Union) 2019
Vienna | Austria | 7–12 April 2019
2019/04 - Deep Learning as a Service for Satellite Imagery Super-resolution
Deep Learning as a Service for Satellite Imagery Super-resolution
The problem of performing we try to solve is what we call multi-spectral imagery super-resolution. Satellites usually take images at different light frequencies and these different bands usually have different resolutions. For example in the case of ESA’s Sentinel-2 satellite it has 4, 6 and 3 bands with 10m, 20m and 60m resolutions respectively.
The idea is to take these low resolution bands (20m and 60m) and to super-resolve them to high resolution (10m). Several methods have been implemented to do this, but we choose to follow [1] that is based on deep neural networks and has shown to outperform previous state-of-the-art pansharpening methods. The results of using this method can be seen in the figure below:
We have modified the original code of [1] in order to make it general enough to be trained on any satellite. In fact we plan to launch services to perform super-resolution on other satellites like LandSat 8 in addition to the current service for Sentinel 2.
EOSC-hub week 2019
10 - 12 April , Prague
2018 - Water quality forecasting system
Title: Water quality forecasting system
Authors: García, Daniel; Aguilar, Fernando; Castrillo, Maria; Marco, Jesús; Monteoliva, Agustín
Conference: Digital Infrastructures for Research (2018)
URL: https://digital.csic.es/handle/10261/171640
2013/10 - Experience with a frozen computational framework from LEP age
Experience with a frozen computational framework from LEP age
By the end of the LEP era there was no well established strategy for the long term preservation of the physics results and data processing framework.
This happening at a time before the generalization of the virtualization techniques, several alternatives were studied.
Among them the setup of a dedicated computational cluster to be preserved medium-long tem in its original state that would provide the capability to reanalyse LEP experiments’ data in case the LHC results, and in particular the Higgs boson search, indicated that was required.
In 2001 we implemented such strategy for the DELPHI experiment at the Institute of Physics of Cantabria, IFCA, placed in Santander (SPAIN).
CHEP 2013
14 - 18 October , Amsterdam
