Applications of deep learning in disease diagnosis of chest radiographs: A survey on materials and methods

Recent advances in deep learning have given rise to high performance in image analysis operations in healthcare. Lung diseases are of particular interest, as most can be identified using non-invasive image modalities. Deep learning techniques such as convolutional neural networks, convolution autoencoders, and graph convolutional networks have been implemented in several pulmonary disease identification applications, e.g., lung nodule classification, Covid-19, and pneumonia detection. Various sources of medical images such as X-rays, computed tomography scans, magnetic resonance imaging, and positron emission tomography scans make deep learning techniques favorable to identify lung diseases with great accuracy. This paper discusses state-of-the-art methods that use deep learning on various medical imaging modalities to detect and classify diseases in the lungs. A description of a few publicly available databases is included in this study, along with some distinct deep learning techniques developed in recent times. Furthermore, several challenges and open research areas for pulmonary disease diagnosis using deep learning are discussed. The objective of this work is to direct researchers in the field of diagnosis of lung diseases.

Model applied to higher education in times of pandemic (CoVID-19)

This research analyzes the perception that students have about the use of ICT in the work and academic fields, for this the Technology Acceptance model was adopted according to the current context in which the institutions operate, the methodology is of a descriptive type, the study population was 194 students enrolled in face-to-face and distance modality of the administrative area of the UTPL. The results showed that students use ICTs at work because they conceive them as useful tools that facilitate learning, as well as increase productivity and performance. On the other hand, regarding the use of ICTs in the educational field, students relate them to the ease of interacting with those involved in the educational process, and they also agree that they are useful and flexible tools that adapt to the training needs of students. each person, finally its use increases academic performance. © 2022 IEEE Computer Society. All rights reserved.

Impact of information and communication technologies in Higher Education Institutions in times of COVID-19: A look from collaborative work and study modality

Higher education institutions, like other organizations worldwide, have had to adapt their educational model to the pandemic situation, as a human and at the same time forced response, to maintain their educational offer and their enrollment income. Under this context, the research objective of this article focuses in determining the perception of university students in relation to collaborative work through the use of information and communication technologies (ICT). The methodology used is descriptive with the application of statistical tools that allow, on the one hand, to determine the validity of the dimensions evaluated and, on the other, to obtain a description of the university feeling. The main findings focus on confirming the validity of the ACOES questionnaire and the relevance of its structure, as well as ratifying the role of collaborative work as an active methodology in Higher Education that invites university professors to be more flexible and innovative constantly, since regardless of the study modality, personal interaction or ICT-mediated, constitutes the cornerstone of a relevant teaching-learning process. © 2021 AISTI.

Perception of learning mediated by information technologies and communicate in higher education

The use of information and communication technologies (ICT) has become an engine for transforming education around the world. Due to the Covid-19 health crisis, educational institutions have increased the use of ICT to guarantee quality education and face student problems;therefore, the purpose of the research was to analyze the perception of learning mediated by the use of ICT in the training of university students. The methodological approach used was quantitative, the data was contrasted through a descriptive study, where a previously validated questionnaire was applied to 146 students from different careers of the Universidad Técnica Particular de Loja with a closed assessment Likert type (1-5). The results obtained showed that the perception of ICT-mediated learning by students is positive and they contribute to learning in a significant way. © 2021 AISTI.

Renal angina: Risk of acute kidney injury in critically ill adults in fundación cardioinfantil in bogotá using clinical risk factors and changes in serum creatinine

BACKGROUND AND AIMS: Acute Kidney Injury (AKI) has remarkable cardiovascular and mortality outcomes, both short and long term potentially preventable with adequate ICU support, thus, early diagnosis is mandatory. Full AKI diagnosis according to KDIGO criteria can result in delayed interventions at admission in ICU, giving potential benefits to alternatives in early diagnosis. Cruz and NEFROINT research group described a scale for prediction of severe AKI, based on risk factors and establishing creatinine cuts as markers of kidney distress.1 Our aim is to describe the predictive capacity of small changes in serum creatinine correlating with clinical risk factors in adult critical care patients. 1. Clin J Am Soc Nephrol (2014) 9, 663-672. METHOD: We retrospectively selected from our Critical Care Nephrology database adult patients admitted in any of our hospital's ICU between February to August 2020, excluding those at admission with diagnosis of AKI, serum creatinine > 2.5 mg/dl, or those receiving dialysis (acute or chronic) or kidney transplantation. We defined AKI according to KDIGO criteria. We calculated Cruz et al scale of prediction of severe AKI. The minimally acceptable criteria for this test was a sensitivity of 95%. A point estimate and confidence intervals of sensitivity and specificity were derived from a contingency table. RESULTS: From 1204 new ICU patients, according to selection criteria we found 372 patients (women 40.3%), with mean age of 60.9 years (range 18-98), mainly hospitalized for medical conditions. Mean values of APACHE II was 22.9. Hemodynamic support was required in 41.1% of patients and mechanical ventilation in 58.6% of patients. (Table 1). AKI KDIGO 2-3 was diagnosed in 65 (26.8%) of patients. Creatinine at admission was statistically different in patients that developed AKI (CI 0.95 -0.51 - 0.15 mg/dl, p=0.0004). Requirement of hemodynamic (p = 0.003) and ventilatory support (p = 0.009), sepsis (p = 0.003), and diagnosis of COVID-19 (p = 0.03) were more frequent in patients who developed AKI. Clinical risk for severe AKI was present in 356 patients (95.7%): 66,5% at very high risk, 9,8% at high risk and 19,2% at moderate risk. Patients without risk criteria were classified as low risk (4,3%). In patients with risk factors for AKI, and a significative increase in creatinine adjusted to risks, diagnostic performance for predicting diagnosis of KDIGO 2-3 AKI had a sensitivity, specificity, positive and negative predictive value of 89% (CI95% 79 - 95%), 58% (CI95% 52 - 64%), 0.31 (CI95% 0.25 - 0.39) and 0.96 (CI95% 0.92 - 0.98) respectively (Figure). Renal replacement therapy was required in 39 (60%) of patients with severe AKI (incidence 10.5%). (Table 2) CONCLUSION: Regardless of the risk stratification for AKI, the absence of significant early changes in serum creatinine rules out the possibility of progression to KDIGO 2-3 AKI in the first seven days after ICU admission.

Renal angina index in adult critical care patients in a population from bogotá-Colombia

BACKGROUND AND AIMS: Clinical outcomes of Acute Kidney Injury (AKI) in ICU mainly depend on opportune preventive strategies. Thus, early identification of AKI is mandatory, and alternative diagnostic strategies become plausible: one of them, Renal Angina Index (RAI), described by Matsuura1, predicts the development of AKI KDIGO 2-3, at 7th day after admission to the intensive care unit according to a cut-off point >6 on a scale with a 'creatinine score' (determined by the difference in serum creatinine between that at ICU admission and the first 24 hours in the ICU) and the impact of the patients medical history. 1Kidney Int Rep (2018) 3, 677-683. Our aim is to describe predictive capacity of the Renal Angina Index (RAI) in adult critical care patients in our population. METHOD: We retrospectively selected from our Critical Care Nephrology database adult patients admitted in any of our hospitals ICU between February to August 2020, excluding those at admission with diagnosis of AKI, serum creatinine > 2.5 mg/dl, or those receiving dialysis (acute or chronic) or kidney transplantation. We defined AKI according to KDIGO criteria. The RAI score was defined as the worst condition score multiplied by the creatinine score. The performance of the RAI score was assessed by Receiver Operating Characteristic (ROC) analysis power to detect a difference of 0.2 between the area under the curve (AUC), under the null hypothesis of AUC = 0.5 (no diagnostic accuracy). The optimal cut point was estimated with the Youden method. RESULTS: From 1204 new ICU patients, we included 372 patients (women 40.3%), with mean age 60.9 (18-98) (table 1). Main indication for ICU admission was medical conditions. Mean APACHE II was 22.9, hemodinamic support was required in 41,1% patients, mechanical ventilation in 58.6% patients and diabetes mellitus was present in 21.5% patients. AKI KDIGO 2-3 developed in 26.8% of patients. Mean creatinine at admission was statistically different in patients with AKI (CI 0.95 - 0.51 - -0.15 mg/dl, p=0.0004). The requirement of hemodynamic (p = 0.003) and ventilatory support (p = 0.009), sepsis (p = 0.003), and COVID-19 (p = 0.03) were more frequent in patients who developed AKI. Renal replacement therapy was required in 39 (60%) of patients with severe AKI (incidence 10,5%). RAI cutt-off point determined by Youden method in the overall sample was 24, being significantly higher in patients who developed AKI (16.54 Vs 7.47, CI 0.95 -13.5-4.99, p <0.001). A cut-off point of 24 was required for the Best predictive capacity for severe AKI, with sensitivity, specificity, positive and negative likelihood ratio of 34%, 94%, 5.5 and 0.7 respectively. CONCLUSION: In our population, RAI score requires a cutoff point much higher than that originally described to predict the development of severe AKI. Losing its discriminatory capacity.

Unsupervised Identification of SARS-CoV-2 Target Cell Groups via Nonlinear Dimensionality Reduction on Single-cell RNA-Seq Data

Recent emergence of a new coronavirus, SARS-CoV2, has caused the disease COVID-19 and has been declared a worldwide pandemic. Identification of relevant modules such as target cells is a significant step for characterizing diseases and consequently leads to better diagnosis, treatment and prognosis. High-throughput single-cell RNA-Seq (scRNA-seq) technologies have advanced in recent years, enabling researchers to investigate cells individually and understand their biological mechanisms. Computational techniques such as data clustering, which are categorized via unsupervised learning methods, are the more suitable for the pre-processing step in scRNA-seq data analysis. They can be used to identify a group of genes that belong to a specific cell type based on similar gene expression patterns. However, due to the sparsity and high-dimensional nature of this type of data, classical clustering methods are not efficient. Therefore, the use of nonlinear dimensionality reduction techniques to improve clustering results is crucial. In this work, we aim to find representative clusters of SARS-CoV-2 target cell lung by combining dimensionality reduction and clustering techniques. We first perform upstream analysis on data, including normalization and filtering using quality control metrics. We then assess the impact of different dimensionality reduction techniques on the clustering results. Our results show that modified Locally Linear Embedding combined with Independent Component Analysis have a very positive impact on clustering large-scale COVID19 scRNA-seq data. To validate our findings, we identified target cell types involved in immune system functionality and a list of overlapping marker genes among COVID-19, Influenza A and HSV-1 infection. © 2020 IEEE.

Data in times of pandemic: The urgency of a new covenant. Reflections from Latin America and the Caribbean

Advances in Information and Communication Technologies (ICT) provide real-time access to a vast amount of data, through which it is possible to know the behavior of social facts. In this scenario, the current SARS-CoV-2 pandemic has allowed, under questionable criteria of immediacy and urgency, to circulate information that generates reality and impacts on decision-making;and has also favored the appropriation of the data, exposing people to violations of their fundamental rights. Both issues are sensitive to Latin America and the Caribbean, a region that today is presented itself not only as the epicenter of the pandemic but also of inequalities. The contribution that bioethical reflection and deliberation can make in this matter, acquires special relevance with a view to generating a new covenant for the treatment of data. Copyright © 2020 Patricia Sorokin, María Angélica Sotomayor Saavedra, Blanca Bórquez Polloni, Myrna Martí, Alejandro Duro, Estela Quiroz Malca, Águeda Muñoz del Carpio Toia, Eduardo A. Duro, Fabiola Czubaj, Laura Rueda, Elizabeth M. Benites Estupiñan, Paula Romina Putallaz, Santiago A. Resett, Ida Cristina Gubert, Luis M. López Dávila, Alejandra Mpolás Andreadis, Mirtha Andreau de Bennato, Claude Vergès.