Your browser doesn't support javascript.
Show: 20 | 50 | 100
Results 1 - 11 de 11
Filter
1.
Algorithms ; 15(10), 2022.
Article in English | Web of Science | ID: covidwho-2099287

ABSTRACT

With the increased popularity of social media platforms such as Twitter or Facebook, sentiment analysis (SA) over the microblogging content becomes of crucial importance. The literature reports good results for well-resourced languages such as English, Spanish or German, but open research space still exists for underrepresented languages such as Romanian, where there is a lack of public training datasets or pretrained word embeddings. The majority of research on Romanian SA tackles the issue in a binary classification manner (positive vs. negative), using a single public dataset which consists of product reviews. In this paper, we respond to the need for a media surveillance project to possess a custom multinomial SA classifier for usage in a restrictive and specific production setup. We describe in detail how such a classifier was built, with the help of an English dataset (containing around 15,000 tweets) translated to Romanian with a public translation service. We test the most popular classification methods that could be applied to SA, including standard machine learning, deep learning and BERT. As we could not find any results for multinomial sentiment classification (positive, negative and neutral) in Romanian, we set two benchmark accuracies of approximate to 78% using standard machine learning and approximate to 81% using BERT. Furthermore, we demonstrate that the automatic translation service does not downgrade the learning performance by comparing the accuracies achieved by the models trained on the original dataset with the models trained on the translated data.

2.
Journal of Clinical Oncology ; 40(16):E18647-E18647, 2022.
Article in English | Web of Science | ID: covidwho-2092897
5.
Journal of Clinical Oncology ; 40(16), 2022.
Article in English | EMBASE | ID: covidwho-2009601

ABSTRACT

Background: Patients with comorbidities especially those with oncological diseases could have severe COVID-19 outcomes. OBJECTIVE: The aim of this study was to evaluate the role of prolonged positivity of SARS-Cov2 in evolution of patients with various neoplasia. Methods: We analysed clinical and laboratory (hematological parameters and inflammatory markers: interleukin-6 and ferritin) data of COVID-19 patients admitted in intensive care unit (ICU) Department of our hospital, during 2020-2021 and presented in medical history solid tumors or haematological neoplasia . The cohort of patients included 78 patients with severe and critical form of COVID-19, 31 patients with solid tumors and 47 patients with hematologic malignancies. We consider long COVID-19 all cases with SARS Cov2 positivity more than 14 days. Results: The frequency of long COVID-19 was quite equal between patients groups with solid tumors and hematologic malignancies, incidence rate 1:2, the incidence rate differences was 1:18, p = 0.75. Long COVID-19 was observed in 60% cases with favourable evolution, Chi-squared 5.35%, p = 0.02, these patients had moderate form and was admitted in hospital in 1 or more days after the onset, median value 3 (min 1, max 55) compared with patients with normal duration of positivity of SARS-Cov2 test-median value 2 (min 1, max 8), p = 0.01. The Kaplan Meyer survival analyses indicated long COVID-19 as predictive factor for unfavourable evolution, Chi-squared 17.97, p < 0.0001. Although we have not obtain significant differences, we observed more severe lymphopenia in patient without long COVID-19, probably because a part of these patients group died in the first 14 days of COVID-19 (0.765 (min 0.04, max 297.64) vs. 1.01 (min 0.09, max 254.35), p = 0.09). The rest of hematological and biochemistry parameters was not significant different between groups. Infectious and thrombotic complication was most frequent in patients with long COVID-19, Chi squared 8.6, p = 0.003. Conclusions: Long Covid-19 is predictable for unfavourable evolution and is associated with sepsis and thrombotic complication. This diagnosis is frequent in patients who was admitted in hospital after the onset of COVID-19 symptoms, early treatment of COVID-19 in oncological patients being very important for favourable evolution.

6.
Research and Practice in Thrombosis and Haemostasis ; 5(SUPPL 2), 2021.
Article in English | EMBASE | ID: covidwho-1509096

ABSTRACT

Background : COVID-19 frequently associated thrombotic complication that could determine severe evolution. Inflammation was proved as important pathogenic mechanism of thrombosis. Aims : The main objective was to evaluate the role of inflammation in increased risk of thrombosis in COVID 19 patients. Methods : Our study was prospective and included all patients diagnosed with COVID 19 between April-September 2020 in Hematology, Pneumology and Intensive Care Unit from Colentina Clinical Hospital (285 patients). The diagnosis was established using molecular test for SARS-Cov2. Results : Thrombotic complication was presented in 56 COVID-19 patients (19, 65%), The higher incidence of thrombosis was observed in severe form of COVID-19: stage 3 (66%) and stage 2 (26.3%), Comorbidities: diabetes mellitus, obesity and arterial hypertension were presented in majority of COVID 19 patients with thrombosis. Acute thrombosis (stroke, myocardial infarction or pulmonary embolism) was diagnosed in 14 patients;all of them were admitted in Intensive care unit due severe form of COVID-19. Inflammatory markers including C reactive protein (CRP), procalcitonin, ferritin are significantly increased in COVID-19 group with acute thrombosis compared with COVID -19 patients with thrombosis in medical history CRP 148.86 mg/L (2.96-386.5) vs. 58.24 mg/L (min 0.25, max 212.98) P = 0.005;procalcitonin 0.93 ng/ml (0.04-784) vs 0.18 (min 0.02, max 14.1) P = 0.02;ferritin 702 ng/ml (min 102, max 4070) vs. 1195 ng/ml (min 358, max 12800) P = 0.03. There is no significant difference between haematological parameters in COVID-19 patients with acute thrombosis or in their medical history. D Dimers are significant increased in patients with acute thrombosis 4.79 ug/ml (0.51-20) vs patients with medical history of thrombosis 2.12 (0.31-20), P = 0.02. The level of protein C, protein S and antitrombine III, antiphospholipid antibodies are not significant modified in the both groups. Conclusions : The assessment of inflammation parameters are very important in COVID-19 patients especially those with a history of thrombosis or who have significant comorbidities (diabetes mellitus, arterial hypertension or obesity).

8.
HemaSphere ; 5(SUPPL 2):383, 2021.
Article in English | EMBASE | ID: covidwho-1393444

ABSTRACT

Background: The severe acute respiratory syndrome coronavirus-2 (SARSCoV-2) was first reported in Wuhan, China in December2019 and represented the pathogen agent that induced COVID-19. The onset and evolution of COVID -19 is severe when is associated with another comorbidities. Patients with acute leukemia present high risk for severe form of COVID-19 Aims: The main objective was to evaluate the particularities of COVID-19 in patients with acute leukemia. Methods: Our study was prospective and included 49 patients with acute leukemia (27 male median age 64 and22 female median age 54,5) who also were SARS CoV2 positive between April2020- February2021 admitted in Hematology and Intensive Care Unit Departments of Colentina Clinical Hospital Bucharest. The diagnosis was established using molecular test for SARS-Cov2 Results: In the group was included 32 patients diagnosed with acute myeloid leukemia (AML), 9 patients with acute lymphoid leukemia (ALL), 6 patients with acut promyelocytic leukemia and2 patients with acute bifenotypic leukemia. Severe form of COVID-19 with ICU addmission was diagnosed in16 patients (32,17%), almost all of them (15 patients) had unfavourable evolution compared with non-ICU patients group with only1 deceased patient, p<0.0001. The recent chemotherapy followed by severe aplasia was the main negative factor that impacted patient evolution (rho=0.508, p=0.0002),13 patients admitted in ICU Department and12 patients in non-ICU. Severe pneumonia (more than 30% lung field) was diagnosed in17 patients with recent chemotherapy and 4 untreated patients. The type of leukemia or refractory status have not any impact of patient evolution. Antiviral therapy - Remdesivir rapidly introduced in patient's therapy was followed by favourable evolution. Summary/Conclusion: Patients with acute leukemia are negatively impacted by intensive chemotherapy during COVID-19 evolution. The key for good prognosis of these patients during COVID-19 are rapid diagnosis and antiviral therapy at the onset of the disease.

9.
Journal of Interdisciplinary Medicine ; 6(2):48-52, 2021.
Article in English | EMBASE | ID: covidwho-1334898

ABSTRACT

The World Health Organization (WHO) declared the transmission of SARS-CoV-2 a Public Health Emergency of International Concern. The novel coronavirus has diverse manifestations, usually similar to a common cold or influenza. The majority of patients with coronavirus disease have typical imaging features. The typical CT characteristics of patients with COVID-19 pneumonia are ground-glass opacities and consolidative lesions with a peripheral and posterior distribution. Noninvasive imaging methods are precise and rapid means of diagnosing pneumonia and cardiovascular complications caused by COVID-19 infection. Therefore, it is important for clinicians to understand the implications of this pandemic and to be familiar with the different imaging aspects of the novel coronavirus disease. This review focuses on the most commonly reported imaging findings of COVID-19 infection in different patients from different countries, the expert recommendations, and the cardiac manifestations of SARS-CoV-2 infection.

11.
Clinical Lymphoma, Myeloma and Leukemia ; 20:S217, 2020.
Article in English | EMBASE | ID: covidwho-989492

ABSTRACT

Context: Coronavirus disease 2019 (COVID-19) is a highly infectious disease. Severity of this disease is associated with comorbidities present (hypertension, obesity, pulmonary disease) or with age. Objective: In this study, we evaluate haematological and biochemistry parameters in order to obtain indications for unfavourable evolution of the patient. Design and Setting: We performed a prospective study that included all patients admitted in our hospital in Hematology, Pneumology, and ICU at Department Colentina Clinical Hospital during April and May 2020. Patients or other participants: The study group included 80 patients that was split into ICU and non-ICU patients. All patients were SARS-CoV-2-positive by molecular test. The distribution according to gender was: 47 male with median age: 73 (min 35, max 88) and 33 female with median age: 50 (min 17, max 84). Results: Age is an important risk factor for the severity, as the median age of patients admitted in ICU was 73 (min 43, max 88) compared with non-ICU patients 41 (min 17, max 64), p=0.00004. Comorbidities associated were important but were present in both groups. In ICU patients, we obtained lower level of lymphocytes compared with non ICU group median: 0.87× 103/L (min 0.09 × 103/L max 7.04 × 103/L) vs 2.17 × 103/L (min 0.19 × 103/L max 3.28 × 103/L), p=0.01. There are no significant differences between groups for the rest of haematological parameters. The biochemistry markers ferritin, AST, ALT, LDH, and D Dimers are important in evaluation of COVID-19 patients;there are statistical differences between ICU and non ICU patients (median value: LDH 405.5 UI/l vs 215 U/l, p=0.001;ferritin 1275 ng/ml vs. 161 ng/ml, p=0.002;D Dimers 2.61 mg/ml FEU vs 0.39 mg/ml FEU, p=0.002;AST 70.9 U/l vs. 19.9 U/l, p=0.0003;ALT 50.05 U/l vs. 18.5 U/l, p=0.009). The ICU patients with unfavourable evolution had a higher level of D-Dimers at the admission in hospital compared with ICU patients who was discharged from the hospital (3.42 mg/ml FEU vs 1.09 mg/ml FEU, p=0.01). We did not obtain statistical significance between ICU groups for all haematological and biochemistry parameters. Conclusions: We conclude that lymphocyte count, LDH, AST, ALT, and ferritin at the time of hospital admission is important to evaluate in COVID-19 patient in order to expect a severe evolution of the disease. D-Dimer should be an important parameter to evaluate for all COVID-19 patients. Anti-thrombotic therapy is important to be introduced in COVID-19 patients.

SELECTION OF CITATIONS
SEARCH DETAIL