Eficacia de la capacidad y la eficiencia pronósticas de la herramienta de inteligencia artificial Thoracic Care Suite de GE aplicada a la radiografía torácica de pacientes con neumonía COVID-19 / Performance in prognostic capacity and efficiency of the Thoracic Care Suite GE AI tool applied to chest radiography of patients with COVID-19 pneumonia

Objetivo: La rápida progresión de la neumonía COVID-19 puede implicar la necesidad de recurrir a sistemas de respiración asistida, como la ventilación mecánica no invasiva o la intubación endotraqueal. La introducción de herramientas que detecten la neumonía COVID-19 puede mejorar la atención sanitaria de los pacientes. Nuestro objetivo es evaluar la eficacia y la eficiencia de la herramienta de inteligencia artificial (IA) Thoracic Care Suite de GE Healthcare (que incorpora Lunit Insight CXR) para predecir la necesidad de recurrir a la respiración asistida en función de la progresión de la neumonía en la COVID-19 en radiografías torácicas consecutivas. Métodos: Se incluyó a pacientes ambulatorios con infección por SARS-CoV-2 confirmada, con hallazgos probables o indeterminados de neumonía COVID-19 en la radiografía torácica (RXT) y que necesitaron una segunda RXT debido a la evolución clínica desfavorable. En las 2RXT se evaluaron el número de campos pulmonares afectados mediante la herramienta de IA. Resultados: Se incluyó a 114 pacientes (57,4±14,2 años; 65 de ellos varones, el 57%) de forma retrospectiva; 15 pacientes (el 13,2%) precisaron respiración asistida. La progresión de la diseminación neumónica ≥0,5 campos pulmonares al día en comparación con el inicio de la neumonía, detectada mediante la herramienta TCS, cuadruplicó el riesgo de precisar respiración asistida. El análisis de los resultados de IA precisó 26 segundos. Conclusiones: Aplicar la herramienta de IA, Thoracic Care Suite, a la RXT de pacientes con neumonía COVID-19 nos permite predecir la necesidad de recurrir a la respiración asistida en menos de medio minuto.(AU)

Objective: Rapid progression of COVID-19 pneumonia may put patients at risk of requiring ventilatory support, such as non-invasive mechanical ventilation or endotracheal intubation. Implementing tools that detect COVID-19 pneumonia can improve the patient's healthcare. We aim to evaluate the efficacy and efficiency of the artificial intelligence (AI) tool GE Healthcare's Thoracic Care Suite (featuring Lunit Insight CXR, TCS) to predict the ventilatory support need based on pneumonic progression of COVID-19 on consecutive chest X-rays. Methods: Outpatients with confirmed SARS-CoV-2 infection, with chest X-ray (CXR) findings probable or indeterminate for COVID-19 pneumonia, who required a second CXR due to unfavorable clinical course, were collected. The number of affected lung fields for the 2CXRs was assessed using the AI tool. Results: One hundred fourteen patients (57.4±14.2 years; 65 of them were men, 57%) were retrospectively collected; and 15 (13.2%) required ventilatory support. Progression of pneumonic extension ≥ 0.5 lung fields per day compared to pneumonia onset, detected using the TCS tool, increased the risk of requiring ventilatory support by 4-fold. Analyzing the AI output required 26seconds of radiological time. Conclusions: Applying the AI tool, Thoracic Care Suite, to CXR of patients with COVID-19 pneumonia allows us to anticipate ventilatory support requirements requiring less than half a minute.(AU)

Inappropriate requests for cranial CT scans in emergency departments increase overuse and reduce test performance.

PURPOSE: The number of non-traumatic urgent cranial computed tomography (NT-UCCT) is exponentially increasing but limited research has been conducted on the quality of clinical justification. Accordingly, we aimed (1) to assess how clinical information in the electronic NT-UCCT request agreed with that provided in the patient's emergency department discharge summary and (2) to analyze the potential effect of those discrepancies on the NT-UCCT overload. MATERIAL AND METHODS: Patients undergoing NT-UCCT in 2017-2021 were randomly selected for this retrospective research-board-approved study. Signs and symptoms (S/S) in electronic request and emergency department discharge summary, acute and relevant computed tomography (CT) findings (acute ischemia or hemorrhage, masses, brain edema, or previously undetected hydrocephalus), and final diagnosis at emergency department discharge summary were collected. Concordance between digital request and emergency department discharge summary and their association with both acute and relevant CT findings and final diagnosis were analyzed. RESULTS: We recruited 156 patients: 80 men; mean age, 55. Acute, relevant CT findings were detected in 28 cases (17.9%). The final diagnosis was neurological disease, non-neurological disease, and no definitive diagnosis in 46 (29.5%), 58 (37.2%), and 51 (32.7%) cases, respectively. Full agreement between the electronic request and emergency department discharge summary occurred in only 36 patients (23.1%). Motor deficit was the most frequent false positive electronic request S/S (18; 11.54%), having low positive predictive value (30.30%; 95%CI 15.59-48.71%) and worst association with acute relevant CT findings than when true positive (OR 2.54; 95%CI 0.04-6.21 vs. OR 6.26, 95%CI 2.21-17.78). Nausea/vomiting was the third most common false negative electronic request S/S (13; 10.26%) and reduced the likelihood of acute and relevant CT findings (OR 0.126; 95%CI 0.016-0.971; p = 0.020). False S/S in electronic request predominated in non-neurological diseases (50-60.2% vs. 33-39.8%; p = 0.068). CONCLUSION: Discrepancies between electronic request and emergency department discharge summary were observed in >75% of patients, leading to unnecessary NT-UCCT tests.

Digital tomosynthesis improves chest radiograph accuracy and reduces microbiological false negatives in COVID-19 diagnosis.

PURPOSE: Diagnosing pneumonia by radiograph is improvable. We aimed (a) to compare radiograph and digital thoracic tomosynthesis (DTT) performances and agreement for COVID-19 pneumonia diagnosis, and (b) to assess the DTT ability for COVID-19 diagnosis when polymerase chain reaction (PCR) and radiograph are negative. METHODS: Two emergency radiologists with 11 (ER1) and 14 experience-years (ER2) retrospectively evaluated radiograph and DTT images acquired simultaneously in consecutively clinically suspected COVID-19 pneumonia patients in March 2020-January 2021. Considering PCR and/or serology as reference standard, DTT and radiograph diagnostic performance and interobserver agreement, and DTT contributions in unequivocal, equivocal, and absent radiograph opacities were analysed by the area under the curve (AUC), Cohen's Kappa, Mc-Nemar's and Wilcoxon tests. RESULTS: We recruited 480 patients (49 ± 15 years, 277 female). DTT increased ER1 (from 0.76, CI95% 0.7-0.8 to 0.79, CI95% 0.7-0.8; P=.04) and ER2 (from 0.77 CI95% 0.7-0.8 to 0.80 CI95% 0.8-0.8, P=.02) radiograph-AUCs, sensitivity, specificity, predictive values, and positive likelihood ratio. In false negative microbiological cases, DTT suggested COVID-19 pneumonia in 13% (4/30; P=.052, ER1) and 20% (6/30; P=.020, ER2) more than radiograph. DTT showed new or larger opacities in 33-47% of cases with unequivocal opacities in radiograph, new opacities in 2-6% of normal radiographs and reduced equivocal opacities by 13-16%. Kappa increased from 0.64 (CI95% 0.6-0.8) to 0.7 (CI95% 0.7-0.8) for COVID-19 pneumonia probability, and from 0.69 (CI95% 0.6-0.7) to 0.76 (CI95% 0.7-0.8) for pneumonic extension. CONCLUSION: DTT improves radiograph performance and agreement for COVID-19 pneumonia diagnosis and reduces PCR false negatives.

[Performance in prognostic capacity and efficiency of the Thoracic Care Suite GE AI tool applied to chest radiography of patients with COVID-19 pneumonia]. / Eficacia de la capacidad y la eficiencia pronósticas de la herramienta de inteligencia artificial Thoracic Care Suite de GE aplicada a la radiografía torácica de pacientes con neumonía por COVID-19.

OBJECTIVE: Rapid progression of COVID-19 pneumonia may put patients at risk of requiring ventilatory support, such as non-invasive mechanical ventilation or endotracheal intubation. Implementing tools that detect COVID-19 pneumonia can improve the patient's healthcare. We aim to evaluate the efficacy and efficiency of the artificial intelligence (AI) tool GE Healthcare's Thoracic Care Suite (featuring Lunit INSIGHT CXR, TCS) to predict the ventilatory support need based on pneumonic progression of COVID-19 on consecutive chest X-rays. METHODS: Outpatients with confirmed SARS-CoV-2 infection, with chest X-ray (CXR) findings probable or indeterminate for COVID-19 pneumonia, who required a second CXR due to unfavorable clinical course, were collected. The number of affected lung fields for the two CXRs was assessed using the AI tool. RESULTS: One hundred fourteen patients (57.4 ± 14.2 years, 65 -57%- men) were retrospectively collected. Fifteen (13.2%) required ventilatory support. Progression of pneumonic extension ≥ 0.5 lung fields per day compared to pneumonia onset, detected using the TCS tool, increased the risk of requiring ventilatory support by 4-fold. Analyzing the AI output required 26 seconds of radiological time. CONCLUSIONS: Applying the AI tool, Thoracic Care Suite, to CXR of patients with COVID-19 pneumonia allows us to anticipate ventilatory support requirements requiring less than half a minute.

A comparison of Covid-19 early detection between convolutional neural networks and radiologists.

BACKGROUND: The role of chest radiography in COVID-19 disease has changed since the beginning of the pandemic from a diagnostic tool when microbiological resources were scarce to a different one focused on detecting and monitoring COVID-19 lung involvement. Using chest radiographs, early detection of the disease is still helpful in resource-poor environments. However, the sensitivity of a chest radiograph for diagnosing COVID-19 is modest, even for expert radiologists. In this paper, the performance of a deep learning algorithm on the first clinical encounter is evaluated and compared with a group of radiologists with different years of experience. METHODS: The algorithm uses an ensemble of four deep convolutional networks, Ensemble4Covid, trained to detect COVID-19 on frontal chest radiographs. The algorithm was tested using images from the first clinical encounter of positive and negative cases. Its performance was compared with five radiologists on a smaller test subset of patients. The algorithm's performance was also validated using the public dataset COVIDx. RESULTS: Compared to the consensus of five radiologists, the Ensemble4Covid model achieved an AUC of 0.85, whereas the radiologists achieved an AUC of 0.71. Compared with other state-of-the-art models, the performance of a single model of our ensemble achieved nonsignificant differences in the public dataset COVIDx. CONCLUSION: The results show that the use of images from the first clinical encounter significantly drops the detection performance of COVID-19. The performance of our Ensemble4Covid under these challenging conditions is considerably higher compared to a consensus of five radiologists. Artificial intelligence can be used for the fast diagnosis of COVID-19.

Early radiological worsening of SARS-CoV-2 pneumonia predicts the need for ventilatory support.

OBJECTIVES: Identifying early markers of poor prognosis of coronavirus disease 2019 (COVID-19) is mandatory. Our purpose is to analyze by chest radiography if rapid worsening of COVID-19 pneumonia in the initial days has predictive value for ventilatory support (VS) need. METHODS: Ambispective observational ethically approved study in COVID-19 pneumonia inpatients, validated in a second outpatient sample. Brixia score (BS) was applied to the first and second chest radiography required for suspected COVID-19 pneumonia to determine the predictive capacity of BS worsening for VS need. Intraclass correlation coefficient (ICC) was previously analyzed among three radiologists. Sensitivity, specificity, likelihood ratios, AUC, and odds ratio were calculated using ROC curves and binary logistic regression analysis. A value of p < .05 was considered statistically significant. RESULTS: A total of 120 inpatients (55 ± 14 years, 68 men) and 112 outpatients (56 ± 13 years, 61 men) were recruited. The average ICC of the BS was between 0.812 (95% confidence interval 0.745-0.878) and 0.906 (95% confidence interval 0.844-0.940). According to the multivariate analysis, a BS worsening per day > 1.3 points within 10 days of the onset of symptoms doubles the risk for requiring VS in inpatients and 5 times in outpatients (p < .001). The findings from the second chest radiography were always better predictors of VS requirement than those from the first one. CONCLUSION: The early radiological worsening of SARS-CoV-2 pneumonia after symptoms onset is a determining factor of the final prognosis. In elderly patients with some comorbidity and pneumonia, a 48-72-h follow-up radiograph is recommended. KEY POINTS: • An early worsening on chest X-ray in patients with SARS-CoV-2 pneumonia is highly predictive of the need for ventilatory support. • This radiological worsening rate can be easily assessed by comparing the first and the second chest X-ray. • In elderly patients with some comorbidity and SARS-CoV-2 pneumonia, close early radiological follow-up is recommended.

Prognostic value of computed tomography pulmonary angiography indices in patients with cancer-related pulmonary embolism: Data from a multicenter cohort study.

OBJECTIVE: To analyze the prognostic value of pulmonary artery obstruction versus right-ventricle (RV) dysfunction radiologic indices in cancer-related pulmonary embolism (PE). METHODS: We enrolled 303 consecutive patients with paraneoplastic PE, evaluated by computed tomography pulmonary angiography (CTPA) between 2013 and 2014. The primary outcome measure was serious complications at 15days. Multivariate analyses were conducted by using binary logistic and robust regressions. Radiological features such as the Qanadli index (QI) and RV dysfunction signs were analyzed with Spearman's partial rank correlations. RESULTS: RV diameter was the only radiological variable associated with an adverse outcome. Subjects with enlarged RV (diameter>45mm) had more 15-day complications (58% versus 40%, p=0.001). The QI correlated with the RV diameter (r=0.28, p<0.001), left ventricle diameter (r=-0.19, p<0.001), right ventricular-to-left ventricular diameter ratio (r=0.39, p<0.001), pulmonary artery diameter (r=0.22, p<0.001), and pulmonary artery/ascending aorta ratio (r=0.27, p<0.001). A QI≥50% was only associated with 15-day complications in subjects with enlarged RV, inverted intraventricular septum, or chronic cardiopulmonary diseases. The central or peripheral PE location did not affect the correlations among radiological variables and was not associated with clinical outcomes. CONCLUSIONS: Right ventricular dysfunction signs in CTPA are more useful than QI in predicting cancer-related PE outcome.

Gastroenteritis eosinofílica y obstrucción intestinal, ¿es la cirugía evitable? / Eosinophilic gastroenteritis and bowel obstruction. Can surgery be avoided?

No disponible