Prognostic models in COVID-19 infection that predict severity: a systematic review

European Journal of Epidemiology, Journal Year: 2023, Volume and Issue: 38(4), P. 355 - 372

Published: Feb. 25, 2023

Abstract Current evidence on COVID-19 prognostic models is inconsistent and clinical applicability remains controversial. We performed a systematic review to summarize critically appraise the available studies that have developed, assessed and/or validated of predicting health outcomes. searched six bibliographic databases identify published articles investigated univariable multivariable adverse outcomes in adult patients, including intensive care unit (ICU) admission, intubation, high-flow nasal therapy (HFNT), extracorporeal membrane oxygenation (ECMO) mortality. identified 314 eligible from more than 40 countries, with 152 these presenting mortality, 66 progression severe or critical illness, 35 mortality ICU admission combined, 17 only, while remaining 44 reported prediction for mechanical ventilation (MV) combination multiple The sample size included varied 11 7,704,171 participants, mean age ranging 18 93 years. There were 353 investigated, area under curve (AUC) 0.44 0.99. A great proportion (61.5%, 193 out 314) internal external validation replication. In 312 (99.4%) studies, be at high risk bias due uncertainties challenges surrounding methodological rigor, sampling, handling missing data, failure deal overfitting heterogeneous definitions severity While several been described literature, they are limited generalizability deficiencies addressing fundamental statistical concerns. Future large, multi-centric well-designed prospective needed clarify uncertainties.

Language: Английский

---

Artificial intelligence and discrete-event simulation for capacity management of intensive care units during the Covid-19 pandemic: A case study

Journal of Business Research, Journal Year: 2023, Volume and Issue: 160, P. 113806 - 113806

Published: March 3, 2023

Language: Английский

---

Assessing the accuracy of lightweight gradient boosting and J48 decision tree for creating effective dietary recommendations

AIP conference proceedings, Journal Year: 2025, Volume and Issue: 3252, P. 020203 - 020203

Published: Jan. 1, 2025

Language: Английский

---

Multimodal deep learning for COVID-19 prognosis prediction in the emergency department: a bi-centric study

Scientific Reports, Journal Year: 2023, Volume and Issue: 13(1)

Published: July 5, 2023

Abstract Predicting clinical deterioration in COVID-19 patients remains a challenging task the Emergency Department (ED). To address this aim, we developed an artificial neural network using textual (e.g. patient history) and tabular laboratory values) data from ED electronic medical reports. The predicted outcomes were 30-day mortality ICU admission. We included consecutive Humanitas Research Hospital San Raffaele Milan area between February 20 May 5, 2020. 1296 patients. Textual predictors consisted of history, physical exam, radiological Tabular age, creatinine, C-reactive protein, hemoglobin, platelet count. TensorFlow tabular-textual model performance indices compared to those models implementing only data. For mortality, combined yielded slightly better performances than fastai XGBoost models, with AUC 0.87 ± 0.02, F1 score 0.62 0.10 MCC 0.52 0.04 ( p < 0.32). As for admission, was superior 0.024) models. Our results suggest that can effectively predict prognosis which may assist physicians their decision-making process.

Language: Английский

---

Research on the application and effect of flipped-classroom combined with TBL teaching model in WeChat-platform-based biochemical teaching under the trend of COVID-19

BMC Medical Education, Journal Year: 2023, Volume and Issue: 23(1)

Published: Sept. 19, 2023

Abstract Background Biochemistry is a core subject in clinical medical education. The traditional classroom teaching model led by teachers often limited to the knowledge transfer of and passive acceptance students. It lacks interactive efficient methods not enough meet learning needs educational goals modern combination WeChat public platform, flipped TBL closer real life workplace, helping students cultivate comprehensive literacy ability solve practical problems. At same time, this has yet be used biochemistry courses. Objective To explore influence mixed combining based on platform upon undergraduates biochemistry. Methods Using research method quasi-experimental design descriptive qualitative research, 68 were selected into blended groups. Among them, group adopts combined with learn biochemical In study, an independent sample t-test was intended analyze differences final scores, chi-square test served satisfaction questionnaires, thematic analysis semi-structured interview data. Results Compared model, significantly improved students' exam scores ( P < 0.05). also higher than that statistical significance results interviews eight summarized three topics: (1) Stimulating interest learning; (2) Improving autonomous (3) Recommendations for improvement. Conclusions positive effect improving problem-solving ability. shows mode effective feasible.

Language: Английский

---

What Is (Not) Big Data Based on Its 7Vs Challenges: A Survey

Big Data and Cognitive Computing, Journal Year: 2022, Volume and Issue: 6(4), P. 158 - 158

Published: Dec. 14, 2022

Big Data has changed how enterprises and people manage knowledge make decisions. However, when talking about Data, so many times there are different definitions what it is used for, as interpretations disagreements. For these reasons, we have reviewed the literature to compile provide a possible solution existing discrepancies between terms Analysis, Mining, Knowledge Discovery in Databases, Data. In addition, gathered patterns phases of some according important companies organisations. Moreover, challenges that sometimes same its own characteristics. These characteristics known Vs. Nonetheless, depending on author, Vs can be more or less, from 3 5, even 7. Furthermore, 4Vs 5Vs not every time. Therefore, this survey, explain been detected explained problems. 7Vs, three which had subtypes.

Language: Английский

---

Concordance and generalization of an AI algorithm with real-world clinical data in the pre-omicron and omicron era

Heliyon, Journal Year: 2024, Volume and Issue: 10(3), P. e25410 - e25410

Published: Feb. 1, 2024

All viruses, including SARS-CoV-2, the virus responsible for COVID-19, continue to evolve, which can lead new variants. The objective of this study is assess agreement between real-world clinical data and an algorithm that utilizes laboratory markers age predict progression disease severity in COVID-19 patients during pre-Omicron Omicron variant periods. evaluated performance a deep learning (DL) predicting scores using from USA, Spain, Turkey (Ankara City Hospital (ACH) set). was developed validated era tested on both Omicron-era data. predictions were compared actual outcomes multidisciplinary approach. concordance index values all datasets ranged 0.71 0.81. In ACH cohort, negative predictive value (NPV) 0.78 or higher observed severe eras, consistent with algorithm's development cohort.

Language: Английский

---

Machine Learning-Driven Prediction of ICU Admissions for COVID-19 Patients

Published: April 18, 2024

In the wake of COVID-19 pandemic, efficiently allocating ICU resources for critical patients has become crucial, especially those with chronic conditions.This study harnesses machine learning (ML) to forecast admissions among in Kuwait, analyzing a dataset 4399 identify pivotal predictors needs.Employing cross-validation and Synthetic Minority Over-sampling Technique (SMOTE) tackle data imbalance, predictive variables were refined using backward feature selection logistic regression evaluated model interpretability Shapley additive explanations (SHAP).The Support Vector Machine (SVM) outperformed other models an area under curve (AUC) 0.91, Extra Tree (ET) showed better performance accuracy 96.42%.Critical included demographics, clinical outcomes like shortness breath, elevated d-dimer levels, abnormal chest X-rays.This research not only underscores potential ML healthcare decision-making during pandemics but also highlights its role discovery science, suggesting broader applications scientific domains.The advances medical informatics by integrating healthcare, offering insights into disease dynamics improving resource allocation strategies.

Language: Английский

---

The Potential Use of Digital Health in Iran: A Systematic Mapping Review

Frontiers in Health Informatics, Journal Year: 2024, Volume and Issue: 13, P. 198 - 198

Published: March 25, 2024

Introduction: Digital health technologies are transforming healthcare delivery globally. The purpose of the current study was to identify and map status digital applications in Iran through providing graphical/tabular classifications on studies conducted this field.Material Methods: Following PRISMA guidelines, relevant English-language papers published from 2012 until 2023 online scientific databases, including PubMed, Scopus, Web Science IEEE Xplore were screened. A total 97 selected for data extraction heath technologies, medical fields, application areas users.Results: number publications has grown considerably since 2016. most common artificial intelligence machine learning (34%), mobile (25%) telehealth (16%). These mostly applied infections (16%), nutrition/metabolism disorders (13%), mental (20%) cancers (12%). key education (21%), therapy (16%) diagnosis (15%). primary users patients (45%) professionals (42%).Conclusion: continuously evolving. activities focused a few like intelligence, with diverse subfields objectives diagnosis. results help research gaps future directions advancing Iran.

Language: Английский

---

Use of machine learning to identify protective factors for death from COVID-19 in the ICU: a retrospective study

PeerJ, Journal Year: 2024, Volume and Issue: 12, P. e17428 - e17428

Published: June 12, 2024

Background Patients in serious condition due to COVID-19 often require special care intensive units (ICUs). This disease has affected over 758 million people and resulted 6.8 deaths worldwide. Additionally, the progression of may vary from individual individual, that is, it is essential identify clinical parameters indicate a good prognosis for patient. Machine learning (ML) algorithms have been used analyzing complex medical data identifying prognostic indicators. However, there still an urgent need model elucidate predictors related patient outcomes. Therefore, this research aimed verify, through ML, variables involved discharge patients admitted ICU COVID-19. Methods In study, 126 were collected with information on demography, hospital length stay outcome, chronic diseases tumors, comorbidities risk factors, complications adverse events, health care, vital indicators southern Brazil. These filtered then selected by ML algorithm known as decision trees optimal set predicting using logistic regression. Finally, confusion matrix was performed evaluate model’s performance variables. Results Of 532 evaluated, 180 discharged: female (16.92%), central venous catheter (23.68%), bladder (26.13%), average 8.46- 23.65-days submitted mechanical ventilation, respectively. addition, chances increase 14% each additional day hospital, 136% patients, 716% when no catheter, 737% used. decrease 3% year age 9% other ventilation. The training presented balanced accuracy 0.81, sensitivity 0.74, specificity 0.88, kappa value 0.64. test had 0.85, 0.75, 0.95, 0.73. McNemar found significant differences error rates data, suggesting classification. work showed female, absence shorter duration associated greater chance discharge. results help develop measures lead

Language: Английский

---