Correction: ai-corona: Radiologist-assistant deep learning framework for COVID-19 diagnosis in chest CT scans.

[This corrects the article DOI: 10.1371/journal.pone.0250952.].

ai-corona: Radiologist-assistant deep learning framework for COVID-19 diagnosis in chest CT scans.

The development of medical assisting tools based on artificial intelligence advances is essential in the global fight against COVID-19 outbreak and the future of medical systems. In this study, we introduce ai-corona, a radiologist-assistant deep learning framework for COVID-19 infection diagnosis using chest CT scans. Our framework incorporates an EfficientNetB3-based feature extractor. We employed three datasets; the CC-CCII set, the MasihDaneshvari Hospital (MDH) cohort, and the MosMedData cohort. Overall, these datasets constitute 7184 scans from 5693 subjects and include the COVID-19, non-COVID abnormal (NCA), common pneumonia (CP), non-pneumonia, and Normal classes. We evaluate ai-corona on test sets from the CC-CCII set, MDH cohort, and the entirety of the MosMedData cohort, for which it gained AUC scores of 0.997, 0.989, and 0.954, respectively. Our results indicates ai-corona outperforms all the alternative models. Lastly, our framework's diagnosis capabilities were evaluated as assistant to several experts. Accordingly, We observed an increase in both speed and accuracy of expert diagnosis when incorporating ai-corona's assistance.

Highest risk abandoned, lost and discarded fishing gear.

Derelict abandoned, lost and discarded fishing gear have profound adverse effects. We assessed gear-specific relative risks from derelict gear to rank-order fishing methods based on: derelict gear production rates, gear quantity indicators of catch weight and fishing grounds area, and adverse consequences from derelict gear. The latter accounted for ghost fishing, transfer of microplastics and toxins into food webs, spread of invasive alien species and harmful microalgae, habitat degradation, obstruction of navigation and in-use fishing gear, and coastal socioeconomic impacts. Globally, mitigating highest risk derelict gear from gillnet, tuna purse seine with fish aggregating devices, and bottom trawl fisheries achieves maximum conservation gains. Locally, adopting controls following a sequential mitigation hierarchy and implementing effective monitoring, surveillance and enforcement systems are needed to curb derelict gear from these most problematic fisheries. Primary and synthesis research are priorities to improve future risk assessments, produce the first robust estimate of global derelict gear quantity, and assess the performance of initiatives to manage derelict gear. Findings from this first quantitative estimate of gear-specific relative risks from derelict gear guide the allocation of resources to achieve the largest improvements from mitigating adverse effects of derelict gear from the world's 4.6 million fishing vessels.

Gene co-expression network analysis reveals immune cell infiltration as a favorable prognostic marker in non-uterine leiomyosarcoma.

The present study aimed to improve the understanding of non-uterine leiomyosarcoma (NULMS) prognostic genes through system biology approaches. This cancer is heterogeneous and rare. Moreover, gene interaction networks have not been reported in NULMS yet. The datasets were obtained from the public gene expression databases. Seven co-expression modules were identified from 5000 most connected genes; using weighted gene co-expression network analysis. Using Cox regression, the modules showed favorable (HR = 0.6, 95% CI = 0.4-0.89, P = 0.0125), (HR = 0.65, 95% CI = 0.44-0.98, P = 0.04) and poor (HR = 1.55, 95% CI = 1.06-2.27, P = 0.025) prognosis to the overall survival (OS) (time = 3740 days). The first one was significant in multivariate HR estimates (HR = 0.4, 95% CI = 0.28-0.69, P = 0.0004). Enriched genes through the Database for Annotation, Visualization, and Integrated Discovery (DAVID) revealed significant immune-related pathways; suggesting immune cell infiltration as a favorable prognostic factor. The most significant protective genes were ICAM3, NCR3, KLRB1, and IL18RAP, which were in one of the significant modules. Moreover, genes related to angiogenesis, cell-cell adhesion, protein glycosylation, and protein transport such as PYCR1, SRM, and MDFI negatively affected the OS and were found in the other related module. In conclusion, our analysis suggests that NULMS might be a good candidate for immunotherapy. Moreover, the genes found in this study might be potential candidates for targeted therapy.