scholarly journals Evaluation of an improved tool for non-invasive prediction of neonatal respiratory morbidity based on fully automated fetal lung ultrasound analysis

2019 ◽  
Vol 9 (1) ◽  
Author(s):  
Xavier P. Burgos-Artizzu ◽  
Álvaro Perez-Moreno ◽  
David Coronado-Gutierrez ◽  
Eduard Gratacos ◽  
Montse Palacio
Keyword(s):  
Lung Ultrasound ◽  
Fetal Lung ◽  
Respiratory Morbidity ◽  
Non Invasive

Prenatal Prediction of Neonatal Respiratory Morbidity: A Radiomics Method Based on Imbalanced Few-Shot Fetal Lung Ultrasound Images

2021 ◽  
Author(s):  
Jing Jiao ◽  
Yanran Du ◽  
Xiaokang Li ◽  
Yi Guo ◽  
Yunyun Ren ◽  
...  
Keyword(s):  
Ensemble Learning ◽  
Data Augmentation ◽  
Lung Ultrasound ◽  
Fetal Lung ◽  
Performance Outcomes ◽  
Respiratory Morbidity ◽  
Ultrasound Images ◽  
Evaluation Tool ◽  
Non Invasive ◽  
Fetal Lungs

Abstract Background: To develop a non-invasive method for the prenatal prediction of neonatal respiratory morbidity (NRM) by a novel radiomics method based on imbalanced few-shot fetal lung ultrasound images.Methods: A total of 210 fetal lung ultrasound images were enrolled in this study, including 159 normal newborns and 51 NRM newborns. Fetal lungs were delineated as the region of interest (ROI), where radiomics features were designed and extracted. Integrating radiomics features selected and two clinical features, including gestational age (GA) and gestational diabetes mellitus (GDM), the prediction model was developed and evaluated. The modelling methods used were data augmentation, cost-sensitive learning, and ensemble learning. Furthermore, two methods, which embed data balancing into ensemble learning, were employed to address the problems of imbalance and few-shot simultaneously.Results: Our model achieved sensitivity values of 0.82, specificity values of 0.84, accuracy values of 0.84 and area under the curve values of 0.87 in the test set. The radiomics features extracted from the ROIs at different locations within the lung region achieved similar classification performance outcomes.Conclusion: The feature set we designed can efficiently and robustly describe fetal lungs for NRM prediction. RUSBoost shows excellent performance compared to state-of-the-art classifiers on the imbalanced few-shot dataset. The diagnostic efficacy of the model we developed is similar to that of several previous reports of amniocentesis and can serve as a non-invasive, precise evaluation tool for NRM prediction.

scholarly journals Prenatal prediction of neonatal respiratory morbidity: a radiomics method based on imbalanced few-shot fetal lung ultrasound images

2022 ◽  
Vol 22 (1) ◽  
Author(s):  
Jing Jiao ◽  
Yanran Du ◽  
Xiaokang Li ◽  
Yi Guo ◽  
Yunyun Ren ◽  
...  
Keyword(s):  
Ensemble Learning ◽  
Data Augmentation ◽  
Lung Ultrasound ◽  
Fetal Lung ◽  
Performance Outcomes ◽  
Respiratory Morbidity ◽  
Ultrasound Images ◽  
Evaluation Tool ◽  
Non Invasive ◽  
Fetal Lungs

Abstract Background To develop a non-invasive method for the prenatal prediction of neonatal respiratory morbidity (NRM) by a novel radiomics method based on imbalanced few-shot fetal lung ultrasound images. Methods A total of 210 fetal lung ultrasound images were enrolled in this study, including 159 normal newborns and 51 NRM newborns. Fetal lungs were delineated as the region of interest (ROI), where radiomics features were designed and extracted. Integrating radiomics features selected and two clinical features, including gestational age and gestational diabetes mellitus, the prediction model was developed and evaluated. The modelling methods used were data augmentation, cost-sensitive learning, and ensemble learning. Furthermore, two methods, which embed data balancing into ensemble learning, were employed to address the problems of imbalance and few-shot simultaneously. Results Our model achieved sensitivity values of 0.82, specificity values of 0.84, balanced accuracy values of 0.83 and area under the curve values of 0.87 in the test set. The radiomics features extracted from the ROIs at different locations within the lung region achieved similar classification performance outcomes. Conclusion The feature set we designed can efficiently and robustly describe fetal lungs for NRM prediction. RUSBoost shows excellent performance compared to state-of-the-art classifiers on the imbalanced few-shot dataset. The diagnostic efficacy of the model we developed is similar to that of several previous reports of amniocentesis and can serve as a non-invasive, precise evaluation tool for NRM prediction.

478: Imaging before 24 weeks gestation can predict neonatal respiratory morbidity in pregnancies complicated by fetal lung masses

2018 ◽  
Vol 218 (1) ◽  
pp. S287-S288
Author(s):  
Katie Sherwin ◽  
Anna I. Girsen ◽  
Safwan S. Halabi ◽  
Ariana M. Spiegel ◽  
Christine J. Lee ◽  
...  
Keyword(s):  
Fetal Lung ◽  
Respiratory Morbidity

scholarly journals Lung ultrasound reaeration score: a useful tool to predict non-invasive ventilation effectiveness

Critical Care ◽  
2014 ◽  
Vol 18 (Suppl 1) ◽  
pp. P255 ◽  
Author(s):  
L Nobile ◽  
P Beccaria ◽  
M Zambon ◽  
L Cabrini ◽  
G Landoni ◽  
...  
Keyword(s):  
Lung Ultrasound ◽  
Invasive Ventilation ◽  
Non Invasive ◽  
Non Invasive Ventilation ◽  
Ventilation Effectiveness

scholarly journals Lung ultrasound as a predictor of mortality of patients with COVID-19

2021 ◽  
pp. e20210092
Author(s):  
Fernando A. Sosa1 ◽  
Agustín Matarrese1 ◽  
Santiago Saavedra1 ◽  
Javier Osatnik1 ◽  
Javier Roberti2 ◽  
...  
Keyword(s):  
Prospective Observational Study ◽  
Lung Ultrasound ◽  
Area Under The Curve ◽  
Invasive Ventilation ◽  
High Flow Nasal Cannula ◽  
Short Term ◽  
Non Invasive ◽  
Non Invasive Ventilation ◽  
Reproducible Method ◽  
Score Range

Objective: To evaluate the performance of lung ultrasound to determine short-term outcomes of patients with COVID-19 admitted to the intensive care unit. Methods: This is a Prospective, observational study. Between July and November 2020, 59 patients were included and underwent at least two LUS assessments using LUS score (range 0-42) on day of admission, day 5th, and 10th of admission. Results: Age was 66.5±15 years, APACHE II was 8.3±3.9, 12 (20%) patients had malignancy, 46 (78%) patients had a non-invasive ventilation/high-flow nasal cannula and 38 (64%) patients required mechanical ventilation. The median stay in ICU was 12 days (IQR 8.5-20.5 days). ICU or hospital mortality was 54%. On admission, the LUS score was 20.8±6.1; on day 5th and day 10th of admission, scores were 27.6±5.5 and 29.4±5.3, respectively (P=0.007). As clinical condition deteriorated the LUS score increased, with a positive correlation of 0.52, P <0.001. Patients with worse LUS on day 5th versus better score had a mortality of 76% versus 33% (OR 6.29, 95%CI 2.01-19.65, p. 0.003); a similar difference was observed on day 10. LUS score of 5th day of admission had an area under the curve of 0.80, best cut-point of 27, sensitivity and specificity of 0.75 and 0.78 respectively. Conclusion: These findings position LUS as a simple and reproducible method to predict the course of COVID-19 patients.

scholarly journals Lung ultrasound monitoring of CPAP effectiveness on SARS-CoV-2 pneumonia: A case report

2020 ◽  
Vol 16 (3) ◽  
Author(s):  
Marco Montanari ◽  
Pierpaolo De Ciantis ◽  
Andrea Boccatonda ◽  
Marta Venturi ◽  
Giuseppe D'Antuono ◽  
...  
Keyword(s):  
Pressure Support ◽  
Lung Ultrasound ◽  
High Peep ◽  
Imaging Method ◽  
Non Invasive ◽  
Ultrasound Monitoring ◽  
Air Interface ◽  
Non Invasive Ventilation ◽  
Heterogeneous Features ◽  
Risk Of Exposure

SARS-CoV-2 infection is characterized by extremely heterogeneous features, going from cases with few symptoms to severe respiratory failures. Chest Computed Tomography (CT) is currently the gold-standard imaging method, although burdened by the risk of exposure to ionizing radiation and management / organizational concerns. In particular, the critical patient undergoing ventilation (invasive or not) seems to be difficult to monitor by repeated CT scan over time. We report the case of a 55-year-old male patient subjected to Continuous Positive Airway Pressure (CPAP) and prone positioning, in which the use of ultrasound monitoring allowed to verify the effectiveness of the pressure support used in recruiting previously atelectasis lung areas. Lung ultrasound can guide pulmonary recruitment and pronation maneuvers in patients undergoing non-invasive ventilation. Ultrasound can identify atelectatic lung areas, which demonstrate an alveolar re-expansion following the setting of high PEEP values, as underlined by the reappearance of pleural/air interface.

Sign in / Sign up

Export Citation Format

Share Document