Refine
Document Type
- Article (6)
Language
- English (6)
Has Fulltext
- yes (6)
Is part of the Bibliography
- no (6)
Keywords
- artificial intelligence (2)
- healthcare (2)
- trustworthy AI (2)
- HBV (1)
- HBV filaments (1)
- HBV genotypes (1)
- HBV surface protein (1)
- PreS1 deletion (1)
- Z-inspection (1)
- arachidonic acid (1)
Institute
- Medizin (5)
- Informatik (1)
- Pharmazie (1)
After the mass-vaccination campaign during the influenza A (H1N1) 2009 pandemic, a significant increase in narcolepsy incidence was observed initially in Scandinavia, later in other European countries and recently also in Canada. Narcolepsy is a sleep disease caused by the loss of hypocretin-producing cells in the hypothalamus. Almost all narcolepsy patients carry the HLA-DQB1*0602 allele, giving a link to an autoimmune-mediated process.
Most of the observed narcolepsy cases were correlated to the vaccination with Pandemrix, the most frequently used vaccine in the EU, and a slight connection to Arepanrix was also detected, which was distributed in Canada. Both vaccines were adjuvanted with AS03, suggesting a possible link between AS03 and narcolepsy. No narcolepsy cases were detected with MF59-adjuvanted or non-adjuvanted influenza vaccines. Recent studies reported differences between Pandemrix and Arepanrix and suggested the vaccine rather than the adjuvant as a suspect for narcolepsy development following vaccination. In addition, in China an increase of narcolepsy cases was reported to occur in absence of vaccination. Possible factors and potential additive effects that may have triggered narcolepsy after the pandemic vaccination are being reviewed in this paper.
Artificial Intelligence (AI) has the potential to greatly improve the delivery of healthcare and other services that advance population health and wellbeing. However, the use of AI in healthcare also brings potential risks that may cause unintended harm. To guide future developments in AI, the High-Level Expert Group on AI set up by the European Commission (EC), recently published ethics guidelines for what it terms “trustworthy” AI. These guidelines are aimed at a variety of stakeholders, especially guiding practitioners toward more ethical and more robust applications of AI. In line with efforts of the EC, AI ethics scholarship focuses increasingly on converting abstract principles into actionable recommendations. However, the interpretation, relevance, and implementation of trustworthy AI depend on the domain and the context in which the AI system is used. The main contribution of this paper is to demonstrate how to use the general AI HLEG trustworthy AI guidelines in practice in the healthcare domain. To this end, we present a best practice of assessing the use of machine learning as a supportive tool to recognize cardiac arrest in emergency calls. The AI system under assessment is currently in use in the city of Copenhagen in Denmark. The assessment is accomplished by an independent team composed of philosophers, policy makers, social scientists, technical, legal, and medical experts. By leveraging an interdisciplinary team, we aim to expose the complex trade-offs and the necessity for such thorough human review when tackling socio-technical applications of AI in healthcare. For the assessment, we use a process to assess trustworthy AI, called 1Z-Inspection® to identify specific challenges and potential ethical trade-offs when we consider AI in practice.
Co-design of a trustworthy AI system in healthcare: deep learning based skin lesion classifier
(2021)
This paper documents how an ethically aligned co-design methodology ensures trustworthiness in the early design phase of an artificial intelligence (AI) system component for healthcare. The system explains decisions made by deep learning networks analyzing images of skin lesions. The co-design of trustworthy AI developed here used a holistic approach rather than a static ethical checklist and required a multidisciplinary team of experts working with the AI designers and their managers. Ethical, legal, and technical issues potentially arising from the future use of the AI system were investigated. This paper is a first report on co-designing in the early design phase. Our results can also serve as guidance for other early-phase AI-similar tool developments.
Extracellular vesicles (EVs) are increasingly recognized as important mediators of intercellular communication. In this study, we aimed to further characterize the role of macrophage-derived EVs in immune responses against hepatitis C virus (HCV) and the potential of polyunsaturated fatty acids (PUFAs) to modulate this modality of innate immunity. To this end, EVs were isolated from interferon-stimulated macrophage cultures or from serum of patients with acute or chronic hepatitis C. EVs were characterized by electron microscopy, flow cytometry, RNA-sequencing, and Western blot analysis. The effect of EVs on replication of HCV was assessed in coculture models. Functional analyses were performed to assess the impact of PUFAs on EV-mediated antiviral immunity. We found that macrophages secreted various cytokines shortly after stimulation with type I and II IFN, which orchestrated a fast but short-lasting antiviral state. This rapid innate immune answer was followed by the production of macrophage-derived EVs, which induced a late, but long-lasting inhibitory effect on HCV replication. Of note, exposure of macrophages to PUFAs, which are important regulators of immune responses, dampened EV-mediated antiviral immune responses. Finally, EVs from patients with hepatitis C exhibited long-lasting antiviral activities during IFN therapy as well. The antiviral effect of EVs from Caucasian and Japanese patients differed, which may be explained by different nutritional uptake of PUFAs. In conclusion, our data indicate that macrophage-derived EVs mediate long-lasting inhibitory effects on HCV replication, which may bridge the time until efficient adaptive immune responses are established, and which can be blunted by PUFAs.
Background & Aims: HBV genotype G (HBV/G) is mainly found in co-infections with other HBV genotypes and was identified as an independent risk factor for liver fibrosis. This study aimed to analyse the prevalence of HBV/G co-infections in healthy European HBV carriers and to characterize the crosstalk of HBV/G with other genotypes.
Methods: A total of 560 European HBV carriers were tested via HBV/G-specific PCR for HBV/G co-infections. Quasispecies distribution was analysed via deep sequencing, and the clinical phenotype was characterized regarding qHBsAg-/HBV-DNA levels and frequent mutations. Replicative capacity and expression of HBsAg/core was studied in hepatoma cells co-expressing HBV/G with either HBV/A, HBV/D or HBV/E using bicistronic vectors.
Results: Although no HBV/G co-infection was found by routine genotyping PCR, HBV/G was detected by specific PCR in 4%-8% of patients infected with either HBV/A or HBV/E but only infrequently in other genotypes. In contrast to HBV/E, HBV/G was found as the quasispecies major variant in co-infections with HBV/A. No differences in the clinical phenotype were observed for HBV/G co-infections. In vitro RNA and DNA levels were comparable among all genotypes, but expression and release of HBsAg was reduced in co-expression of HBV/G with HBV/E. In co-expression with HBV/A and HBV/E expression of HBV/G-specific core was enhanced while core expression from the corresponding genotype was markedly diminished.
Conclusions: HBV/G co-infections are common in European inactive carriers with HBV/A and HBV/E infection, but sufficient detection depends strongly on the assay. HBV/G regulated core expression might play a critical role for survival of HBV/G in co-infections.
The N-terminus of the hepatitis B virus (HBV) large surface protein (LHB) differs with respect to genotypes. Compared to the amino terminus of genotype (Gt)D, in GtA, GtB and GtC, an additional identical 11 amino acids (aa) are found, while GtE and GtG share another similar 10 aa. Variants of GtB and GtC affecting this N-terminal part are associated with hepatoma formation. Deletion of these amino-terminal 11 aa in GtA reduces the amount of LHBs and changes subcellular accumulation (GtA-like pattern) to a dispersed distribution (GtD-like pattern). Vice versa, the fusion of the GtA-derived N-terminal 11 aa to GtD causes a GtA-like phenotype. However, insertion of the corresponding GtE-derived 10 aa to GtD has no effect. Deletion of these 11aa decreases filament size while neither the number of released viral genomes nor virion size and infectivity are affected. A negative regulatory element (aa 2–8) and a dominant positive regulatory element (aa 9–11) affecting the amount of LHBs were identified. The fusion of this motif to eGFP revealed that the effect on protein amount and subcellular distribution is not restricted to LHBs. These data identify a novel region in the N-terminus of LHBs affecting the amount and subcellular distribution of LHBs and identify release-promoting and -inhibiting aa residues within this motive.
