Welcome to the 4th DeepHealth newsletter! Throughout these communications, we will be providing more relevant news about the project in the last months.
The summer holidays are gone, and we are working hard to finalise the DeepHealth activity in the last year of the project. Navigate through this newsletter to get a glimpse of our highlights in the latest months!
Main Goals and Results
We are honored that the DeepHealth project has attracted the interest of the EC Innovation Radar. More specifically, both the core libraries developed in the DeepHealth project, have been analyzed by the European Commission's Innovation Radar and categorised as "Exploring" technology, in terms of actively exploring value creation opportunities, that "address the needs of existing markets/existing customers".
The core libraries developed in the DeepHealth project are:
ECVL (European Computer Vision Library), an image processing software stack that supports most common image formats (medical & non-medical) and whole-slide images,
EDDLL (European Distributed Deep Learning Library), an open -source library for Distributed Deep Learning and Tensor Operations,
EDDL and ECVL are two of the major outcomes of the DeepHealth project, forming together with the associated front-end and back-end application, the DeepHealth toolkit. Take a look at the DeepHealth github for technical details.
The University of Torino released the open-access dataset UniToPatho, collected for the homonymous DeepHealth Use Case 2. UniToPatho is a dataset of annotated high-resolution hematoxylin and eosin stained images, comprising different histological samples of colorectal polyps, collected from patients undergoing cancer screening. It includes the most relevant patch images extracted from 292 whole-slide images.
The University of Torino released the open-access dataset UniTOBrain collected for the homonymous DeepHealth Use Case 3. UniToBrain is a dataset of Computed Tomography (CT) perfusion images (CTP). The dataset includes 100 training subjects and 15 testing subjects used in a submitted publication for the training and the testing of a Convolutional Neural Network.
On 15th April, Prof. Marco Aldinucci (UniTo) was invited at the NVIDIA GPU Technology Conference to present "The universal cloud-HPC pipeline for the AI-assisted explainable diagnosis of covid-19 pneumonia", illustrating the results from the EU COVID-CLAIRE (Confederation of Laboratories for Artificial Intelligence Research in Europe) task force on image analysis achieved during the 1st Italian lockdown.
The COUGHVID dataset is a corpus for the study of large-scale cough analysis algorithms collected and analyzed by EPFL. COUGHVID provides over 25,000 crowdsourced cough recordings representing a wide range of participant ages, genders, geographic locations, and COVID-19 statuses. EPFL also contributed the open-sourced cough detection algorithm to the research community to assist in data robustness assessment.
On 12th February, Prof. Marco Aldinucci (UniTo) participated to the External Research Seminar Series at the University of Cambridge - Institute of Metabolic Science. He discussed a technical presentation about “Pipeline-as-a-Service with StreamFlow” demonstrating how AI pipelines are can be managed on the OpenDeepHealth infrastructure using StreamFlow, a workflow management system developed in DeepHealth project.
On 19th April, Prof. David Atienza (EPFL) gave a keynote speech at 2021 IEEE Latin America Electron Devices Conference (LAEDC) with the title "Energy-Scalable Many-Core Servers: Follow Your Brain!". The keynote concerned the continuous advances in manufacturing technologies enabling the development of more powerful and compact high-performance servers made of many-core processing architectures and stacked memories.
The “BDVA/DAIRO Workshop on HPC, Big Data, IoT and AI future industry-driven collaborative strategic topics” workshop launched on 23rd March 2021, was held as a part of EuroHPC Summit Week 2021. Monica Caballero (everis), as the DeepHealth Project coordinator, and Jon A. Gomez (UPV), as the Technical Manager of DeepHealth, gave a pitch on the key facts of the project and detailed the perspective of DeepHealth for identifying research and innovation needs in terms of convergence and alignment of the basic technologies and their utilization by industry.
DeepHealth actively participated in the Dataweek 2021 held online from 25th to the 27th of May. Jon A. Gómez (UPV) presented the "DeepHealth AI Deep Learning pipelines" in the "Big Data and AI Pipelines in Big Data PPP Projects from a Technology Analysis and Benchmarking Perspective" session and Marco Aldinucci (UniTo) presented the project at the "Future challenges in IoT, AI, and convergence of HPC & Cloud & Big Data" workshop, and together with Eduardo Quiñones (BSC) discussed about technical challenges with other ICT 11 projects representatives.