Deutsche Telekom (Spain)

Transfer learning aims at adapting a model learned from source dataset to target dataset. It is a beneficial approach especially when annotating on the target dataset is expensive or infeasible. Transfer learning has demonstrated its powerful learning capabilities in various vision tasks. Despite transfer learning being a promising approach, it is still an open question how to adapt the model learned from the source dataset to the target dataset. One big challenge is to prevent the impact of category bias on classification performance. Dataset bias exists when two images from the same category, but from different datasets, are not classified as the same. To address this problem, a transfer learning algorithm has been proposed, called negative back‐dropout transfer learning (NB‐TL), which utilizes images that have been misclassified and further performs back‐dropout strategy on them to penalize errors. Experimental results demonstrate the effectiveness of the proposed algorithm. In particular, the authors evaluate the performance of the proposed NB‐TL algorithm on UCF 101 action recognition dataset, achieving 88.9% recognition rate.

Introduction: The International Registry in Organ Donation and Transplantation (IRODaT) has been developed by the Transplant Procurement Management (TPM) since 2001. It provides periodical overview of international data on organ donation and transplantation. In 2010 the registry was transferred to the Donation and Transplantation Institute (DTI foundation) with the aim to improve and innovate its methodology. A new data base is currently being developed to collect the required information on three different levels: National - Regional-Hospital, respectively. Purpose: IRODaT aims at supporting the transplant community by providing up-to-date data on organ donation and transplantation worldwide. Hence, any similarities or differences between countries, regions and/or centers could be identified. Materials and methods: The database, easily to be accessed and managed, it gives, up-to-date information provided by a network of professionals directly involved in the various stages of the donation and transplantation process. The registry provides numbers of Donors after Brain death (DBD), Donors after Cardiac Death (DBD) and Living Donors (LD) as well as specific organ transplantation activities related to the three types of organ donation. Follow-up has been continuously performed. All numbers are constantly checked and updated. New countries are invited to participate in the project and they have already started to report their organ donation activity to the international community. Results: Data on organ donation and transplantation from 65 countries has been collected and compiled for 2009 and 2010. New countries have reported their statistics: Bolivia, Belarus, Canada, Cuba, Nicaragua, Lebanon and Peru. Data analysis revealed a remarkable increase in donation rate in such countries as Croatia, Italy, Slovenia, Czech Republic, Germany, Hungary, Australia, Luxemburg, Poland, Brazil, Singapore, Iran, Saudi Arabia, Venezuela, Romania, Bulgaria, Mexico, Russia and Argentina.[Fig 1][Fig. 2][Fig. 3 & 4]Conclusions: IRODaT is able to provide data and basic statistics with a short timeframe, based on an extensive network of key persons and experts involved in the organ donation and transplantation process worldwide. National and comparative statistics carried out on an international basis can be provided. They have turned out to be of an extreme value to scientific programs, social and governmental bodies. Thanks to these initiatives it is possible to analyze current practices in organ and tissue donation in any country or region of the world.