Graphical abstract

No abstract

The year 2020 was characterized by the COVID-19 pandemic that has caused, by the end of March 2021, more than 2.5 million deaths worldwide. Since the beginning, besides the laboratory test, used as the gold standard, many applications have been applying deep learning algorithms to chest X-ray images to recognize COVID-19 infected patients. In this context, we found out that convolutional neural networks perform well on a single dataset but struggle to generalize to other data sources. To overcome this limitation, we propose a late fusion approach where we combine the outputs of several state-of-the-art CNNs, introducing a novel method that allows us to construct an optimum ensemble determining which and how many base learners should be aggregated. This choice is driven by a two-objective function that maximizes, on a validation set, the accuracy and the diversity of the ensemble itself. A wide set of experiments on several publicly available datasets, accounting for more than 92000 images, shows that the proposed approach provides average recognition rates up to 93.54% when tested on external datasets.

No abstract

This is the first study on gut microbiota (GM) in children affected by coronavirus disease 2019 (COVID-19). Stool samples from 88 patients with suspected severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection and 95 healthy subjects were collected (admission: 3–7 days, discharge) to study GM profile by 16S rRNA gene sequencing and relationship to disease severity. The study group was divided in COVID-19 (68), Non–COVID-19 (16), and MIS-C (multisystem inflammatory syndrome in children) (4). Correlations among GM ecology, predicted functions, multiple machine learning (ML) models, and inflammatory response were provided for COVID-19 and Non–COVID-19 cohorts. The GM of COVID-19 cohort resulted as dysbiotic, with the lowest α-diversity compared with Non–COVID-19 and CTRLs and by a specific β-diversity. Its profile appeared enriched in Faecalibacterium, Fusobacterium, and Neisseria and reduced in Bifidobacterium, Blautia, Ruminococcus, Collinsella, Coprococcus, Eggerthella, and Akkermansia, compared with CTRLs (p < 0.05). All GM paired-comparisons disclosed comparable results through all time points. The comparison between COVID-19 and Non–COVID-19 cohorts highlighted a reduction of Abiotrophia in the COVID-19 cohort (p < 0.05). The GM of MIS-C cohort was characterized by an increase of Veillonella, Clostridium, Dialister, Ruminococcus, and Streptococcus and a decrease of Bifidobacterium, Blautia, Granulicatella, and Prevotella, compared with CTRLs. Stratifying for disease severity, the GM associated to “moderate” COVID-19 was characterized by lower α-diversity compared with “mild” and “asymptomatic” and by a GM profile deprived in Neisseria, Lachnospira, Streptococcus, and Prevotella and enriched in Dialister, Acidaminococcus, Oscillospora, Ruminococcus, Clostridium, Alistipes, and Bacteroides. The ML models identified Staphylococcus, Anaerostipes, Faecalibacterium, Dorea, Dialister, Streptococcus, Roseburia, Haemophilus, Granulicatella, Gemmiger, Lachnospira, Corynebacterium, Prevotella, Bilophila, Phascolarctobacterium, Oscillospira, and Veillonella as microbial markers of COVID-19. The KEGG ortholog (KO)–based prediction of GM functional profile highlighted 28 and 39 KO-associated pathways to COVID-19 and CTRLs, respectively. Finally, Bacteroides and Sutterella correlated with proinflammatory cytokines regardless disease severity. Unlike adult GM profiles, Faecalibacterium was a specific marker of pediatric COVID-19 GM. The durable modification of patients’ GM profile suggested a prompt GM quenching response to SARS-CoV-2 infection since the first symptoms. Faecalibacterium and reduced fatty acid and amino acid degradation were proposed as specific COVID-19 disease traits, possibly associated to restrained severity of SARS-CoV-2–infected children. Altogether, this evidence provides a characterization of the pediatric COVID-19–related GM.

Background Prostate magnetic resonance imaging (MRI) is technically demanding, requiring high image quality to reach its full diagnostic potential. An automated method to identify diagnostically inadequate images could help optimize image quality. Purpose To develop a convolutional neural networks (CNNs) based analysis pipeline for the classification of prostate MRI image quality. Study Type Retrospective. Subjects Three hundred sixteen prostate mpMRI scans and 312 men (median age 67). Field Strength/Sequence A 3 T; fast spin echo T2WI, echo planar imaging DWI, ADC, gradient‐echo dynamic contrast enhanced (DCE). Assessment MRI scans were reviewed by three genitourinary radiologists (V.P., M.D.M., S.C.) with 21, 12, and 5 years of experience, respectively. Sequences were labeled as high quality (Q1) or low quality (Q0) and used as the reference standard for all analyses. Statistical Tests Sequences were split into training, validation, and testing sets (869, 250, and 120 sequences, respectively). Inter‐reader agreement was assessed with the Fleiss kappa. Following preprocessing and data augmentation, 28 CNNs were trained on MRI slices for each sequence. Model performance was assessed on both a per‐slice and a per‐sequence basis. A pairwise t ‐test was performed to compare performances of the classifiers. Results The number of sequences labeled as Q0 or Q1 was 38 vs. 278 for T2WI, 43 vs. 273 for DWI, 41 vs. 275 for ADC, and 38 vs. 253 for DCE. Inter‐reader agreement was almost perfect for T2WI and DCE and substantial for DWI and ADC. On the per‐slice analysis, accuracy was 89.95% ± 0.02% for T2WI, 79.83% ± 0.04% for DWI, 76.64% ± 0.04% for ADC, 96.62% ± 0.01% for DCE. On the per‐sequence analysis, accuracy was 100% ± 0.00% for T2WI, DWI, and DCE, and 92.31% ± 0.00% for ADC. The three best algorithms performed significantly better than the remaining ones on every sequence ( P ‐value < 0.05). Data Conclusion CNNs achieved high accuracy in classifying prostate MRI image quality on an individual‐slice basis and almost perfect accuracy when classifying the entire sequences. Evidence Level 4 Technical Efficacy Stage 1

The availability of the epidemiological data strongly affects the reliability of several mathematical models in tracing and forecasting COVID-19 pandemic, hampering a fair assessment of their relative performance. The marked difference between the lethality of the virus when comparing the first and second waves is an evident sign of the poor reliability of the data, also related to the variability over time in the number of performed swabs. During the early epidemic stage, swabs were made only to patients with severe symptoms taken to hospital or intensive care unit. Thus, asymptomatic people, not seeking medical assistance, remained undetected. Conversely, during the second wave of infection, total infectives included also a percentage of detected asymptomatic infectives, being tested due to close contacts with swab positives and thus registered by the health system. Here, we compared the outcomes of two SIR-type models (the standard SIR model and the A-SIR model that explicitly considers asymptomatic infectives) in reproducing the COVID-19 epidemic dynamic in Italy, Spain, Germany, and France during the first two infection waves, simulated separately. We found that the A-SIR model overcame the SIR model in simulating the first wave, whereas these discrepancies are reduced in simulating the second wave, when the accuracy of the epidemiological data is considerably higher. These results indicate that increasing the complexity of the model is useless and unnecessarily wasteful if not supported by an increased quality of the available data.

Investigating the frequency and distribution of small subgraphs with a few nodes/edges, i.e., motifs, is an effective analysis method for static networks. Motif-driven analysis is also useful for temporal networks where the number of motifs is significantly larger due to the additional temporal information on edges. This variety makes it challenging to design a temporal motif model that can consider all aspects of temporality. In the literature, previous works have introduced various models that handle different characteristics. In this work, we compare the existing temporal motif models, and evaluate the facets of temporal networks that are overlooked in the literature. We first survey four temporal motif models and highlight their differences. Then, we evaluate the advantages and limitations of these models with respect to the temporal inducedness and timing constraints. In addition, we suggest a new lens, event pairs, to study temporal motifs to investigate cause-effect relations. We believe that our comparative survey and extensive evaluation will catalyze the research on temporal network motif models.