Dataset Information

Ensemble learning for poor prognosis predictions: A case study on SARS-CoV-2.

ABSTRACT:

Objective

Risk prediction models are widely used to inform evidence-based clinical decision making. However, few models developed from single cohorts can perform consistently well at population level where diverse prognoses exist (such as the SARS-CoV-2 [severe acute respiratory syndrome coronavirus 2] pandemic). This study aims at tackling this challenge by synergizing prediction models from the literature using ensemble learning.

Materials and methods

In this study, we selected and reimplemented 7 prediction models for COVID-19 (coronavirus disease 2019) that were derived from diverse cohorts and used different implementation techniques. A novel ensemble learning framework was proposed to synergize them for realizing personalized predictions for individual patients. Four diverse international cohorts (2 from the United Kingdom and 2 from China; N = 5394) were used to validate all 8 models on discrimination, calibration, and clinical usefulness.

Results

Results showed that individual prediction models could perform well on some cohorts while poorly on others. Conversely, the ensemble model achieved the best performances consistently on all metrics quantifying discrimination, calibration, and clinical usefulness. Performance disparities were observed in cohorts from the 2 countries: all models achieved better performances on the China cohorts.

Discussion

When individual models were learned from complementary cohorts, the synergized model had the potential to achieve better performances than any individual model. Results indicate that blood parameters and physiological measurements might have better predictive powers when collected early, which remains to be confirmed by further studies.

Conclusions

Combining a diverse set of individual prediction models, the ensemble method can synergize a robust and well-performing model by choosing the most competent ones for individual patients.

SUBMITTER: Wu H

PROVIDER: S-EPMC7717299 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:ObjectivesTo identify predictors of poor prognosis in previously healthy young individuals admitted to hospital with coronavirus disease 2019 (COVID-19).MethodsWe studied a cohort of patients hospitalized with COVID-19. All patients without co-morbidities, without usual treatments and ≤65 years old were selected from an international registry (HOPE-COVID-19, NCT04334291). We focused on baseline variables-symptoms and signs at admission-to analyse risk factors for poor prognosis. The primary end point was a composite of major adverse clinical events during hospitalization including mortality, mechanical ventilation, high-flow nasal oxygen therapy, prone, sepsis, systemic inflammatory response syndrome and embolic events.ResultsOverall, 773 healthy young patients were included. The primary composite end point was observed in 29% (225/773) and the overall mortality rate was 3.6% (28/773). In the combined event group, 75% (168/225) of patients were men and the mean age was 49 (±11) years, whereas in the non-combined event group, the prevalence of male gender was 43% (238/548) and the mean age was 42 (±13) years (p < 0.001 for both). On admission, respiratory insufficiency and cough were described in 51.4% (114/222) and 76% (170/223) of patients, respectively, in the combined event group, versus 7.9% (42/533) and 56% (302/543) of patients in the other group (p < 0.001 for both). The strongest independent predictor for the combined end point was desaturation (Spo2 <92%) (OR 5.40; 95% CI 3.34-8.75; p < 0.001), followed by tachypnoea (OR 3.17; 95% CI 1.93-5.21; p < 0.001), male gender (OR 3.01; 95% CI 1.96-4.61; p < 0.001) and pulmonary infiltrates on chest X-ray at admission (OR 2.21; 95% CI 1.18-4.16; p 0.014).ConclusionsMajor adverse clinical events were unexpectedly high considering the baseline characteristics of the cohort. Signs of respiratory compromise at admission and male gender, were predictive for poor prognosis among young healthy patients hospitalized with COVID-19.