Your browser doesn't support javascript.
Show: 20 | 50 | 100
Results 1 - 7 de 7
Filter
3.
Eur Radiol ; 32(7): 4446-4456, 2022 Jul.
Article in English | MEDLINE | ID: covidwho-1707890

ABSTRACT

OBJECTIVES: We aimed to develop deep learning models using longitudinal chest X-rays (CXRs) and clinical data to predict in-hospital mortality of COVID-19 patients in the intensive care unit (ICU). METHODS: Six hundred fifty-four patients (212 deceased, 442 alive, 5645 total CXRs) were identified across two institutions. Imaging and clinical data from one institution were used to train five longitudinal transformer-based networks applying five-fold cross-validation. The models were tested on data from the other institution, and pairwise comparisons were used to determine the best-performing models. RESULTS: A higher proportion of deceased patients had elevated white blood cell count, decreased absolute lymphocyte count, elevated creatine concentration, and incidence of cardiovascular and chronic kidney disease. A model based on pre-ICU CXRs achieved an AUC of 0.632 and an accuracy of 0.593, and a model based on ICU CXRs achieved an AUC of 0.697 and an accuracy of 0.657. A model based on all longitudinal CXRs (both pre-ICU and ICU) achieved an AUC of 0.702 and an accuracy of 0.694. A model based on clinical data alone achieved an AUC of 0.653 and an accuracy of 0.657. The addition of longitudinal imaging to clinical data in a combined model significantly improved performance, reaching an AUC of 0.727 (p = 0.039) and an accuracy of 0.732. CONCLUSIONS: The addition of longitudinal CXRs to clinical data significantly improves mortality prediction with deep learning for COVID-19 patients in the ICU. KEY POINTS: • Deep learning was used to predict mortality in COVID-19 ICU patients. • Serial radiographs and clinical data were used. • The models could inform clinical decision-making and resource allocation.


Subject(s)
COVID-19 , Deep Learning , Humans , Intensive Care Units , Radiography , X-Rays
4.
NPJ Digit Med ; 5(1): 5, 2022 Jan 14.
Article in English | MEDLINE | ID: covidwho-1625359

ABSTRACT

While COVID-19 diagnosis and prognosis artificial intelligence models exist, very few can be implemented for practical use given their high risk of bias. We aimed to develop a diagnosis model that addresses notable shortcomings of prior studies, integrating it into a fully automated triage pipeline that examines chest radiographs for the presence, severity, and progression of COVID-19 pneumonia. Scans were collected using the DICOM Image Analysis and Archive, a system that communicates with a hospital's image repository. The authors collected over 6,500 non-public chest X-rays comprising diverse COVID-19 severities, along with radiology reports and RT-PCR data. The authors provisioned one internally held-out and two external test sets to assess model generalizability and compare performance to traditional radiologist interpretation. The pipeline was evaluated on a prospective cohort of 80 radiographs, reporting a 95% diagnostic accuracy. The study mitigates bias in AI model development and demonstrates the value of an end-to-end COVID-19 triage platform.

5.
Front Pediatr ; 8: 579512, 2020.
Article in English | MEDLINE | ID: covidwho-902431

ABSTRACT

Objective: This work aims to investigate the clinical features and the temporal changes of RT-PCR and CT in COVID-19 pediatric patients. Methods: The clinical, RT-PCR, and CT features of 114 COVID-19 pediatric in-patients were retrospectively reviewed from January 21 to March 14, 2020. All patients had chest CT on admission and were identified as positive by pharyngeal swab nucleic acid test. The clinical features were analyzed, as well as the features and the temporal changes of RT-PCR and CT. Results: Fever (62, 54%) and cough (61, 54%) were the most common symptoms. There were 34 (30%) cases of concurrent infections. The most common imaging features on CT were ground-glass opacities (46, 40%) and consolidation (46, 40%). The bilateral lower lobes were the most common pattern of involvement, with 63 cases (55%) involving one to two lobes, and in 32 (28%) cases CT was normal. Throughout the whole duration of COVID-19 in children, the diagnostic positive rate of RT-PCR has been far higher than that of CT (all P < 0.05). For RT-PCR follow-up, reliable negative results were obtained only 7 days after the onset of symptoms. Though lung involvement on chest CT progressed rapidly in several cases, lung involvement in children with COVID-19 is mild, with a median value of 2 on CT score. Conclusions: RT-PCR is more reliable than CT in the initial diagnosis of pediatric patients with COVID-19. On follow-up, reliable negative RT-PCR results are available 7 days after the initial symptoms. The use of CT should be considered for follow-up purposes only if necessary.

6.
Obesity (Silver Spring) ; 28(11): 2040-2048, 2020 11.
Article in English | MEDLINE | ID: covidwho-654908

ABSTRACT

OBJECTIVE: This study aimed to assess the association between adipose tissue distribution and severity of clinical course in patients with severe acute respiratory syndrome coronavirus 2. METHODS: For this retrospective study, 143 hospitalized patients with confirmed coronavirus disease 2019 (COVID-19) who underwent an unenhanced abdominal computed tomography (CT) scan between January 1, 2020, and March 30, 2020, were included. Univariate and multivariate logistic regression analyses were performed to identify the risk factors associated with the severity of COVID-19 infection. RESULTS: There were 45 patients who were identified as critically ill. High visceral to subcutaneous adipose tissue area ratio (called visceral adiposity) (odds ratio: 2.47; 95% CI: 1.05-5.98, P = 0.040) and low mean attenuation of skeletal muscle (called high intramuscular fat [IMF] deposition) (odds ratio: 11.90; 95% CI: 4.50-36.14; P < 0.001) were independent risk factors for critical illness. Furthermore, visceral adiposity or high IMF deposition increased the risk of mechanical ventilation (P = 0.013, P < 0.001, respectively). High IMF deposition increased the risk of death (P = 0.012). CONCLUSIONS: COVID-19 patients with visceral adiposity or high IMF deposition have higher risk for critical illness. Therefore, patients with abdominal obesity should be monitored more carefully when hospitalized.


Subject(s)
Betacoronavirus/pathogenicity , Coronavirus Infections/complications , Obesity, Abdominal/complications , Pneumonia, Viral/complications , Subcutaneous Fat/physiopathology , Adiposity/physiology , Aged , COVID-19 , Critical Illness , Female , Humans , Male , Middle Aged , Pandemics , Retrospective Studies , Risk Factors , SARS-CoV-2
7.
Clin Imaging ; 69: 72-74, 2021 Jan.
Article in English | MEDLINE | ID: covidwho-628456

ABSTRACT

Typical chest CT findings in COVID-19 have been described as bilateral peripheral ground glass opacities, with or without consolidation. Halo sign and reversed halo sign have been reported as atypical imaging findings in this disease. However, to the best of our knowledge, combined presence of these signs has never been reported before. Herein, we present a COVID-19 patient with numerous atypical target-shaped, combined halo and reversed halo pulmonary lesions, in the absence of any other underlying disease.


Subject(s)
COVID-19 , Coronavirus Infections , Pneumonia, Viral , Coronavirus Infections/epidemiology , Humans , Lung/diagnostic imaging , Pandemics , Pneumonia, Viral/epidemiology , SARS-CoV-2 , Thorax , Tomography, X-Ray Computed
SELECTION OF CITATIONS
SEARCH DETAIL