Suchergebnisse

BACKGROUND: Immigration has taken the central stage in world politics, especially in the developed countries like Germany, where the continuous flow of immigrants has been well documented since 1960s. Strikingly, emerging data suggest that migrant patients have a poorer response to the treatment and lower survival rates in their new host country, raising concerns about health disparities. Herein, we present our investigation on the treatment response rate and cancer survival in German patients with and without an immigrant background that were treated at our comprehensive cancer center in Germany. METHODS: Initially, we considered 8162 cancer patients treated at the Center for Integrated Oncology (CIO), University Hospital Bonn, Germany (April 2002–December 2015) for matched-pair analysis. Subsequently, the German patients with a migration background and those from the native German population were manually identified and catalogued using a highly specific name-based algorithm. The clinical parameters such as demographic characteristics, tumor characteristics, defined staging criteria, and primary therapy were further adjusted. Using these stringent criteria, a total of 422 patients (n = 211, Germans with migration background; n = 211, native German population) were screened to compare for the treatment response and survival rates (i.e., 5-year overall survival, progression-free survival, and time to progression). RESULTS: Compared to the cohort with migration background, the cohort without migration background was slightly older (54.9 vs. 57.9 years) while having the same sex distribution (54.5% vs. 55.0% female) and longer follow-up time (36.9 vs. 42.6 months). We did not find significant differences in cancer survival (5-year overall survival, P = 0.771) and the response rates (Overall Remission Rate; McNemar's test, P = 0.346) between both collectives. CONCLUSION: Contrary to prior reports, we found no significant differences in cancer survival between German patients with immigrant background and native ...

Fast and reliable detection of patients with severe and heterogeneous illnesses is a major goal of precision medicine1,2 . Patients with leukaemia can be identifed using machine learning on the basis of their blood transcriptomes3 . However, there is an increasing divide between what is technically possible and what is allowed, because of privacy legislation4,5 . Here, to facilitate the integration of any medical data from any data owner worldwide without violating privacy laws, we introduce Swarm Learning—a decentralized machine-learning approach that unites edge computing, blockchain-based peer-to-peer networking and coordination while maintaining confdentiality without the need for a central coordinator, thereby going beyond federated learning. To illustrate the feasibility of using Swarm Learning to develop disease classifers using distributed data, we chose four use cases of heterogeneous diseases (COVID-19, tuberculosis, leukaemia and lung pathologies). With more than 16,400 blood transcriptomes derived from 127 clinical studies with non-uniform distributions of cases and controls and substantial study biases, as well as more than 95,000 chest X-ray images, we show that Swarm Learning classifers outperform those developed at individual sites. In addition, Swarm Learning completely fulfls local confdentiality regulations by design. We believe that this approach will notably accelerate the introduction of precision medicine.

Fast and reliable detection of patients with severe and heterogeneous illnesses is a major goal of precision medicine(1,2). Patients with leukaemia can be identified using machine learning on the basis of their blood transcriptomes(3). However, there is an increasing divide between what is technically possible and what is allowed, because of privacy legislation(4,5). Here, to facilitate the integration of any medical data from any data owner worldwide without violating privacy laws, we introduce Swarm Learning-a decentralized machine-learning approach that unites edge computing, blockchain-based peer-to-peer networking and coordination while maintaining confidentiality without the need for a central coordinator, thereby going beyond federated learning. To illustrate the feasibility of using Swarm Learning to develop disease classifiers using distributed data, we chose four use cases of heterogeneous diseases (COVID-19, tuberculosis, leukaemia and lung pathologies). With more than 16,400 blood transcriptomes derived from 127 clinical studies with non-uniform distributions of cases and controls and substantial study biases, as well as more than 95,000 chest X-ray images, we show that Swarm Learning classifiers outperform those developed at individual sites. In addition, Swarm Learning completely fulfils local confidentiality regulations by design. We believe that this approach will notably accelerate the introduction of precision medicine.