Refine
Language
- English (16)
Has Fulltext
- yes (16)
Is part of the Bibliography
- no (16)
Keywords
- 2019-nCoV (1)
- ACLF (1)
- Aromatase Inhibitors (1)
- CLIF-C ACLF score (1)
- CLIF-C ACLF-R score (1)
- COVID-19 (1)
- Carcinoma, Ductal, Breast (1)
- Colchicum (1)
- ERSCongress (1)
- Estrogen Receptor Modulators (1)
Institute
- Medizin (15)
- E-Finance Lab e.V. (1)
- Wirtschaftswissenschaften (1)
This article discusses the counterpart of interactive machine learning, i.e., human learning while being in the loop in a human-machine collaboration. For such cases we propose the use of a Contradiction Matrix to assess the overlap and the contradictions of human and machine predictions. We show in a small-scaled user study with experts in the area of pneumology (1) that machine-learning based systems can classify X-rays with respect to diseases with a meaningful accuracy, (2) humans partly use contradictions to reconsider their initial diagnosis, and (3) that this leads to a higher overlap between human and machine diagnoses at the end of the collaboration situation. We argue that disclosure of information on diagnosis uncertainty can be beneficial to make the human expert reconsider her or his initial assessment which may ultimately result in a deliberate agreement. In the light of the observations from our project, it becomes apparent that collaborative learning in such a human-in-the-loop scenario could lead to mutual benefits for both human learning and interactive machine learning. Bearing the differences in reasoning and learning processes of humans and intelligent systems in mind, we argue that interdisciplinary research teams have the best chances at tackling this undertaking and generating valuable insights.
Recently, a 15-valent (PCV15) and a 20-valent pneumococcal conjugate vaccine (PCV20) have been licensed by the US Food and Drug Administration and are under evaluation by the European Medicines Agency. PCV15 contains all serotypes of the 13-valent conjugate vaccine (PCV13) plus serotype 22F and 33F and PCV20 includes PCV13 serotypes plus serotypes 8, 10A, 11A, 12F, 15B, 22F, 33F. We investigated pneumococcal serotype distribution, secular trends and proportion of pneumonia caused by serotypes included in PCV13, PCV15, PCV20, and the 23-valent pneumococcal polysaccharide vaccine (PPV23) among adult patients with all-cause community-acquired pneumonia (CAP) between 2013 and 2019. We applied logistic mixed regression modelling to assess annual trends. Urine samples from adult patients with CAP treated in the community or hospital in Germany and included in the CAPNETZ study, a prospective multi-centre cohort study, were analysed by two serotype-specific multiplex urinary antigen detection assays (UAD1/UAD2) at Pfizer’s Vaccines Research and Development Laboratory. UAD1 detects serotypes in PCV13, UAD2 detects additional serotypes in PCV20 plus serotypes 2, 9N, 17F and 20. Out of 1,831 patients screened, urine samples with a valid UAD test result were available for 1,343 patients (73.3%). Among those patients, 829 patients (61.7%) were male, 792 patients (59.0%) were aged ≥60 years, 1038 patients (77.3%) had at least one comorbidity and 1,204 patients (89.7%) were treated in the hospital. The overall proportion of vaccine-type pneumonia among all-cause CAP for PCV13, PCV15, PCV20 and PPV23 was 7.7% (n=103), 9.1% (n=122), 12.3% (n=165) and 13.3% (n=178). Over the entire observation period, we did not observe evidence for significant annual trends in pneumococcal vaccine serotype coverage against pneumonia in adults (PCV13: OR 0.94, 95% CI 0.83-1.05; PCV15: OR 0.93, 95% CI 0.84-1.03; PCV20: OR 0.95, 95% CI 0.86-1.04; PPV23: OR 0.99, 95% CI 0.90-1.08). In conclusion, our data show that i) the infant vaccination program of PCV13, which started in Germany 2010 did not result in a relevant and sustained decrease of PCV13 serotypes in pneumonia in adults and ii) that the gap in the coverage between PCV20 and PPV23 was small and did not increase over the entire observation time.
Recently, a 15-valent (PCV15) and a 20-valent pneumococcal conjugate vaccine (PCV20) have been licensed by the US Food and Drug Administration and are under evaluation by the European Medicines Agency. PCV15 contains all serotypes of the 13-valent conjugate vaccine (PCV13) plus serotype 22F and 33F and PCV20 includes PCV13 serotypes plus serotypes 8, 10A, 11A, 12F, 15B, 22F, 33F. We investigated pneumococcal serotype distribution, secular trends and proportion of pneumonia caused by serotypes included in PCV13, PCV15, PCV20, and the 23-valent pneumococcal polysaccharide vaccine (PPV23) among adult patients with all-cause community-acquired pneumonia (CAP) between 2013 and 2019. We applied logistic mixed regression modelling to assess annual trends. Urine samples from adult patients with CAP treated in the community or hospital in Germany and included in the CAPNETZ study, a prospective multi-centre cohort study, were analysed by two serotype-specific multiplex urinary antigen detection assays (UAD1/UAD2) at Pfizer’s Vaccines Research and Development Laboratory. UAD1 detects serotypes in PCV13, UAD2 detects additional serotypes in PCV20 plus serotypes 2, 9N, 17F and 20. Out of 1,831 patients screened, urine samples with a valid UAD test result were available for 1,343 patients (73.3%). Among those patients, 829 patients (61.7%) were male, 792 patients (59.0%) were aged ≥60 years, 1038 patients (77.3%) had at least one comorbidity and 1,204 patients (89.7%) were treated in the hospital. The overall proportion of vaccine-type pneumonia among all-cause CAP for PCV13, PCV15, PCV20 and PPV23 was 7.7% (n=103), 9.1% (n=122), 12.3% (n=165) and 13.3% (n=178). Over the entire observation period, we did not observe evidence for significant annual trends in pneumococcal vaccine serotype coverage against pneumonia in adults (PCV13: OR 0.94, 95% CI 0.83-1.05; PCV15: OR 0.93, 95% CI 0.84-1.03; PCV20: OR 0.95, 95% CI 0.86-1.04; PPV23: OR 0.99, 95% CI 0.90-1.08). In conclusion, our data show i) no decline of PCV13 serotypes in all-cause CAP between 2013-2019 mainly due to a persistently high proportion of serotype 3 suggesting no meaningful effect of childhood PCV13 vaccination on PCV13 coverage in pneumonia in adults during this time period and ii) that the gap in the coverage between PCV20 and PPV23 was small and did not increase over the entire observation time.
Severe acute respiratory syndrome virus 2 (SARS-CoV-2) is the cause of the current coronavirus disease 19 (COVID-19) pandemic. Protease inhibitors are under consideration as virus entry inhibitors that prevent the cleavage of the coronavirus spike (S) protein by cellular proteases. Herein, we showed that the protease inhibitor aprotinin (but not the protease inhibitor SERPINA1/alpha-1 antitrypsin) inhibited SARS-CoV-2 replication in therapeutically achievable concentrations. An analysis of proteomics and translatome data indicated that SARS-CoV-2 replication is associated with a downregulation of host cell protease inhibitors. Hence, aprotinin may compensate for downregulated host cell proteases during later virus replication cycles. Aprotinin displayed anti-SARS-CoV-2 activity in different cell types (Caco2, Calu-3, and primary bronchial epithelial cell air–liquid interface cultures) and against four virus isolates. In conclusion, therapeutic aprotinin concentrations exert anti-SARS-CoV-2 activity. An approved aprotinin aerosol may have potential for the early local control of SARS-CoV-2 replication and the prevention of COVID-19 progression to a severe, systemic disease.
Cystic fibrosis (CF) lung disease is aggravated by recurrent and ultimately chronic bacterial infections. One of the key pathogens in adult CF lung disease is P. aeruginosa (PA). In addition to bacteria, respiratory viral infections are suggested to trigger pulmonary exacerbations in CF. To date, little is known on how chronic infections with PA influence susceptibility and response to viral infection. We investigated the interactions between PA, human rhinovirus (HRV) and the airway epithelium in a model of chronic PA infection using differentiated primary bronchial epithelial cells (pBECs) and clinical PA isolates obtained from the respiratory sample of a CF patient. Cells were repeatedly infected with either a mucoid or a non-mucoid PA isolate for 16 days to simulate chronic infection, and subsequently co-infected with HRV. Key cytokines and viral RNA were quantified by cytometric bead array, ELISA and qPCR. Proteolytic degradation of IL-6 was analyzed by Western Blots. Barrier function was assessed by permeability tests and transepithelial electric resistance measurements. Virus infection stimulated the production of inflammatory and antiviral mediators, including interleukin (IL)-6, CXCL-8, tumor necrosis factor (TNF)-α, and type I/III interferons. Co-infection with a non-mucoid PA isolate increased IL-1β protein concentrations (28.88 pg/ml vs. 6.10 pg/ml), but in contrast drastically diminished levels of IL-6 protein (53.17 pg/ml vs. 2301.33 pg/ml) compared to virus infection alone. Conditioned medium obtained from co-infections with a non-mucoid PA isolate and HRV was able to rapidly degrade recombinant IL-6 in a serine protease-dependent manner, whereas medium from individual infections or co-infections with a mucoid isolate had no such effect. After co-infection with HRV and the non-mucoid PA isolate, we detected lower mRNA levels of Forkhead box J1 (FOXJ1) and Cilia Apical Structure Protein (SNTN), markers of epithelial cell differentiation to ciliated cells. Moreover, epithelial permeability was increased and barrier function compromised compared to single infections. These data show that PA infection can influence the response of bronchial epithelial cells to viral infection. Altered innate immune responses and compromised epithelial barrier function may contribute to an aggravated course of viral infection in PA-infected airways.
Background: Cirrhosis is known to have a high prevalence and mortality worldwide. However, in Europe, the epidemiology of cirrhosis is possibly undergoing demographic changes, and etiologies may have changed due to improvements in standard of care. The aim of this population-based study was to analyze the trends and the course of liver cirrhosis and its complications in recent years in Germany.
Methods: We analyzed the data of all hospital admissions in Germany within diagnosis-related groups from 2005 to 2018. The diagnostic records of cirrhosis and other categories of diseases were based on ICD-10-GM codes. The primary outcome measurement was in-hospital mortality. Trends were analyzed through Poisson regression of annual number of admissions. The impact of cirrhosis on overall in-hospital mortality were assessed through the multivariate multilevel logistic regression model adjusted for age, sex, and comorbidities.
Findings: Of the 248,085,936 admissions recorded between 2005 and 2018, a total of 2,302,171(0•94%) were admitted with the diagnosis of cirrhosis, mainly as a comorbidity. Compared with other chronic diseases, patients admitted with cirrhosis were younger, mainly male and had the highest in-hospital mortality rate. Diagnosis of cirrhosis was an independent risk factor of in-hospital mortality with the highest odds ratio (OR:6•2[95%CI:6.1-6•3]) among all diagnoses. The prevalence of non-alcoholic fatty liver disease has increased four times from 2005 to 2018, while alcoholic cirrhosis is 20 times than other etiologies. Bleeding was found to be decreasing over time, but ascites remained the most common complication and was increasing.
Interpretation: This nationwide study demonstrates that cirrhosis represents a considerable healthcare burden, as shown by the increasing in-hospital mortality, also in combination with other chronic diseases. Alcohol-related cirrhosis and complications are on the rise. More resources and better management strategies are warranted.
COVID-19 HAS AGAIN TIGHTENED ITS GRIP AROUND THE WORLD AND THE HEALTH SYSTEM. THIS ARTICLE GIVES AN INTRODUCTION TO EXPLAINABLE INTERACTIVE MACHINE LEARNING AND PROVIDES INSIGHTS ON HOW THIS METHOD MAY NOT ONLY HELP IN ENGINEERING MORE POWERFUL AI SYSTEMS, BUT ALSO HOW IT MAY HELP TO EASE THE BURDEN OF VIRAL STRAINS ON THE HEALTHCARE SYSTEM.
A case of Lymphangioleiomyomatosis (LAM) of the lung in a patient with a history of breast cancer
(2019)
Background: Lymphangioleiomyomatosis (LAM) is a rare progressive cystic and nodular disease of the lung characterized by smooth muscle cell proliferation. LAM predominantly affects young premenopausal women. This report is of a case of LAM presenting in a 47-year-old woman with a past history of breast cancer and discusses the possibility of an association between the two conditions.
Case report: A 47-year-old woman presented as an emergency with an exacerbation of a four-month history of shortness of breath and dry cough. Her symptoms began following the start of anti-hormonal treatment with letrozole and goserelin acetate for a moderately differentiated (grade 2) invasive ductal carcinoma of the breast (pT2, pN0, M0) which was positive for expression of estrogen receptor (ER+), progesterone receptor (PR+), and human epidermal growth factor receptor 2 (HER2+). Until the previous four months, she had breast-conserving treatment with radiotherapy and tamoxifen therapy. Following hospital admission, she was found to be in type I respiratory failure. Chest X-ray, lung computed tomography (CT), and positron-emission tomography (PET) showed diffuse cystic and nodular lung lesions, consistent with a diagnosis of LAM, and antihormonal therapy was discontinued. She developed pericarditis that was treated with the anti-inflammatory agent, colchicine. Treatment with letrozole and sirolimus improved her respiratory symptoms.
Conclusions: A rare case of LAM is presented in a woman with a recent history of breast cancer. Because both tumors were hormone-dependent, this may support common underlying gene associations and signaling pathways between the two types of tumor.