Winners of the 500k Pandemic Response Challenge

Our team VALENCIA IA4COVID, coleaded by Nuria Oliver and me, has won 500k Pandemic Response Challenge, organized by XPRIZE Foundation and supported by Cognizant. The $500K Pandemic Response Challenge, required teams to build effective data-driven AI systems capable of accurately predicting COVID-19 transmission rates and prescribing intervention and mitigation measures that, with testing in “what-if” scenarios, were shown to minimize infection rates as well as negative economic impacts.

Our group is made up of fourteen experts from the Universities and research centers of the Valencian Community. Our model successfully forecasted epidemiological evolution through their use of AI and data science and provided decision makers with the best prescriptor models to produce non-pharmaceutical intervention plans that minimize the number of infections while minimizing the stringency of the interventions.

You can find the details of the prize here  and of our model here .

Finalists of the 500k Pandemic Challenge Response XPRIZE!!!!

Our team VALENCIA IA4COVID has progressed to the 2nd phase of the Pandemic Response Challenge, organized by XPRIZE Foundation and supported by Cognizant. This is a $500K, a four-month challenge that focuses on the development of data-driven AI systems to predict COVID-19 infection rates and prescribe Intervention Plans (IPs) that regional governments, communities, and organizations can implement to minimize harm when reopening their economies. Our group is made up of fourteen experts from the Universities and research centers of the Valencian Community and it is leaded by Nuria Oliver and me. We have all been working intensively since the beginning of the pandemic, altruistically and using the resources available to us in our respective institutions and with the occasional philanthropic collaboration of some companies.

Our model is among the three best models in the competition in MAE Mean Rank, leading in ASIA and in the top 5 of EUROPE in MAE per 100k habitants.

You can see our predictions here. The model has not been updated since its release on December 22nd.

Potential limitations in COVID-19 machine learning due to data source variability

Our recent paper Potential limitations in COVID-19 machine learning due to data source variability: A case study in the nCov2019 dataset has been accepted for publication in . (JAMIA, IF 4.112). We study whether the lack of representative coronavirus disease 2019 (COVID-19) data is a bottleneck for reliable and generalizable machine learning. Data sharing is insufficient without data quality, in which source variability plays an important role. We showcase and discuss potential biases from data source variability for COVID-19 machine learning. Our results are based in the publicly available nCov2019 dataset, including patient-level data from several countries. We aimed to the discovery and classification of severity subgroups using symptoms and comorbidities. We show that cases from the 2 countries with the highest prevalence were divided into separate subgroups with distinct severity manifestations. This variability can reduce the representativeness of training data with respect the model target populations and increase model complexity at risk of overfitting.

Article d’opinió en VIA EMPRESA sobre la disponibilitat de dades

Article d’opinió al voltant de la disponibilitat de dades al voltant de la evolució de la pandèmia del COVID-19. Després de portar uns mesos analitzant dades sobre l’evolució de la COVID-19 hem comprovat que algunes coses han millorat i que ha augmentat la disponibilitat de les dades i l’accés a aquests.

Comptar cada vegada amb ciutadans amb una major cultura de l’ús de les dades ens permet ser més corresponsables en la presa de decisions individuals i poder recolzar amb major criteri les decisions dels organismes oficials.

Pots trobar l’article sencer en https://www.viaempresa.cat/opinio/conejero-upv-dades-dades-dades_2142814_102.html

 

 

Investigadores de la UPV trabajan en la lucha contra la COVID-19 a través de la Ciencia de Datos

We Alberto Conejero (Instituto Universitario de Matemática Pura y Aplicada, IUMPA) and Miguel Rebollo (of the Valencian Research Institute for Artificial Intelligence, VRAIN) of UPV are part of the Data Science Working Group in the Fight against COVID-19 , of the Commissioner for the Presidency of the Generalitat Valenciana on Strategy for Artificial Intelligence and Data Sciences against COVID-19. News appeared in UPV, in El Periodic newspaper and in RUVID website.