310: Prediction of neonatal respiratory morbidity by quantitative ultrasound lung texture analysis: a multicenter study

Palacio, Montse; Bonet‐Carne, Elisenda; Cobo, Teresa; Perez‐Moreno, Alvaro; Sabrià, J.; Richter, Jute; Kacerovský, Marian; Jacobsson, Bo; García-Posada, Raúl Alejandro; Bugatto, Fernando; Santisteve, Ramon; Vives, Àngels; Parra-Cordero, Mauro; Hernández‐Andrade, Edgar; Bartha, José Luis; Carretero-lucena, Pilar; Lit, Kai; Cruz, R Armas; Burke, Minke; Vavilala, Suseela; Iruretagoyena, Igor; Delgado, Juan Luis; Schenone, Mauro; Vilanova, Josep

doi:10.1016/j.ajog.2016.11.568

Cited by 7 publications

(10 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These methods, mostly based on very well-known textural descriptors, aimed to assess the respiratory status [6], [7] or predict the newborn diseases [8]. Texture analysis has been also employed to determine the respiratory morbidity in newborns [18], while recent fully automatic methods focused on predicting the neonatal maturation degree [19] as well as the respiratory morbidity [20].…”

Section: Related Workmentioning

confidence: 99%

Deep learning in the ultrasound evaluation of neonatal respiratory status

Gravina

Gragnaniello

Verdoliva

et al. 2021

2020 25th International Conference on Pattern Recognition (ICPR)

View full text Add to dashboard Cite

Lung ultrasound imaging is reaching growing interest from the scientific community. On one side, thanks to its harmlessness and high descriptive power, this kind of diagnostic imaging has been largely adopted in sensitive applications, like the diagnosis and follow-up of preterm newborns in neonatal intensive care units. On the other side, state-of-the-art image analysis and pattern recognition approaches have recently proven their ability to fully exploit the rich information contained in these data, making them attractive for the research community. In this work, we present a thorough analysis of recent deep learning networks and training strategies carried out on a vast and challenging multicenter dataset comprising 87 patients with different diseases and gestational ages. These approaches are employed to assess the lung respiratory status from ultrasound images and are evaluated against a reference marker. The conducted analysis sheds some light on this problem by showing the critical points that can mislead the training procedure and proposes some adaptations to the specific data and task. The achieved results sensibly outperform those obtained by a previous work, which is based on textural features, and narrow the gap with the visual score predicted by the human experts.

show abstract

Section: Related Workmentioning

confidence: 99%

Deep learning in the ultrasound evaluation of neonatal respiratory status

Gravina

Gragnaniello

Verdoliva

et al. 2021

2020 25th International Conference on Pattern Recognition (ICPR)

View full text Add to dashboard Cite

show abstract

“…Prenatal evaluation of fetal lung maturity (FLM) is still controversial and different approaches have been described with invasive biological tools and fetal imaging [1][2][3][4][5][6][7][8][9][10]. FLM is mainly correlated with surfactant level and different tests have been developed to quantify this production through detection of components in the amniotic fluid [1].…”

Section: Introductionmentioning

confidence: 99%

“…FLM is mainly correlated with surfactant level and different tests have been developed to quantify this production through detection of components in the amniotic fluid [1]. However, performances of these tests to predict neonatal respiratory morbidity are limited for a routine clinical application because they need invasive procedures and all have a positive predictive value ranging from 18 to 34% [2,[4][5][6]. Thus, development of new and non-invasive methods to assess FLM is challenging to predict respiratory distress syndrome (RDS).…”

Section: Introductionmentioning

confidence: 99%

“…RDS complicates 1% of all births and the incidence increases with prematurity. RDS ranges from 26% to 30% in preterm neonates before 34 weeks of gestation (WG) and from 5% to 20% after 34 WG [2,11].…”

Section: Introductionmentioning

confidence: 99%

“…These studies demonstrated imaging changes associated with lung maturation but weak correlations between echogenicity and clinical outcomes at birth. Recently, a quantitative fetal lung texture analysis has been described and demonstrated a strong correlation with gestational age, biological test on amniotic fluid and risk of neonatal respiratory morbidity [2,16]. Prediction of the pulmonary function through quantitative fetal lung stiffness measurement seems innovative and could be complementary to quantitative ultrasound texture analysis.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Feasibility of two-dimensional ultrasound shear wave elastography of human fetal lungs and liver: A pilot study

Mottet

Vidal

et al. 2020

Diagnostic and Interventional Imaging

View full text Add to dashboard Cite

Fetal lung growth predicts the risk for early-life respiratory infections and childhood asthma

Zazara,

Giannou,

Schepanski

et al. 2024

World J Pediatr

View full text Add to dashboard Cite

Background Early-life respiratory infections and asthma are major health burdens during childhood. Markers predicting an increased risk for early-life respiratory diseases are sparse. Here, we identified the predictive value of ultrasound-monitored fetal lung growth for the risk of early-life respiratory infections and asthma. Methods Fetal lung size was serially assessed at standardized time points by transabdominal ultrasound in pregnant women participating in a pregnancy cohort. Correlations between fetal lung growth and respiratory infections in infancy or early-onset asthma at five years were examined. Machine-learning models relying on extreme gradient boosting regressor or classifier algorithms were developed to predict respiratory infection or asthma risk based on fetal lung growth. For model development and validation, study participants were randomly divided into a training and a testing group, respectively, by the employed algorithm. Results Enhanced fetal lung growth throughout pregnancy predicted a lower early-life respiratory infection risk. Male sex was associated with a higher risk for respiratory infections in infancy. Fetal lung growth could also predict the risk of asthma at five years of age. We designed three machine-learning models to predict the risk and number of infections in infancy as well as the risk of early-onset asthma. The models’ R2 values were 0.92, 0.90 and 0.93, respectively, underscoring a high accuracy and agreement between the actual and predicted values. Influential variables included known risk factors and novel predictors, such as ultrasound-monitored fetal lung growth. Conclusion Sonographic monitoring of fetal lung growth allows to predict the risk for early-life respiratory infections and asthma. Graphical abstract

show abstract

310: Prediction of neonatal respiratory morbidity by quantitative ultrasound lung texture analysis: a multicenter study

Cited by 7 publications

References 0 publications

Deep learning in the ultrasound evaluation of neonatal respiratory status

Deep learning in the ultrasound evaluation of neonatal respiratory status

Feasibility of two-dimensional ultrasound shear wave elastography of human fetal lungs and liver: A pilot study

Fetal lung growth predicts the risk for early-life respiratory infections and childhood asthma

Contact Info

Product

Resources

About