Peer Reviewed Journal Article Nursing Neonatal Diaphragmatic Hernia

• Loading metrics

Open up Access

Report Protocol

A maChine and deep Learning Approach to predict pulmoNary hyperteNsIon in newbornS with congenital diaphragmatic Hernia (CLANNISH): Protocol for a retrospective study

• Giorgio De Nunzio ,
• Genny Raffaeli,
• Irene Borzani,
• Alice Griggio,
• Luana Conte,
• Francesco Macchini,
• Valentina Condò,
• Nicola Persico,
• Isabella Fabietti,
• Stefano Ghirardello,
• Maria Pierro,
• Benedetta Tafuri,
• Giuseppe Como,
• Donato Cascio,
• Mariarosa Colnaghi,
• Fabio Mosca,
•  [ ... ],
• Giacomo Cavallaro
• [ view all ]
• [ view less ]

A maChine and deep Learning Arroyo to predict pulmoNary hyperteNsIon in newbornS with congenital diaphragmatic Hernia (CLANNISH): Protocol for a retrospective study

• Ilaria Amodeo,
• Giorgio De Nunzio,
• Genny Raffaeli,
• Irene Borzani,
• Alice Griggio,
• Luana Conte,
• Francesco Macchini,
• Valentina Condò,
• Nicola Persico,
• Isabella Fabietti

x

• Published: November nine, 2021
• https://doi.org/10.1371/journal.pone.0259724

Figures

Abstruse

Introduction

Outcome predictions of patients with built diaphragmatic hernia (CDH) all the same have some limitations in the prenatal gauge of postnatal pulmonary hypertension (PH). We propose applying Machine Learning (ML), and Deep Learning (DL) approaches to fetuses and newborns with CDH to develop forecasting models in prenatal epoch, based on the integrated analysis of clinical data, to provide neonatal PH as the start outcome and, possibly: favorable response to fetal endoscopic tracheal occlusion (FETO), demand for Extracorporeal Membrane Oxygenation (ECMO), survival to ECMO, and death. Moreover, we programme to produce a (semi)automatic fetus lung division organisation in Magnetic Resonance Imaging (MRI), which will exist useful during project implementation simply will likewise exist an important tool itself to standardize lung volume measures for CDH fetuses.

Methods and analytics

Patients with isolated CDH from singleton pregnancies will be enrolled, whose prenatal checks were performed at the Fetal Surgery Unit of measurement of the Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico (Milan, Italy) from the thirty^th week of gestation. A retrospective data collection of clinical and radiological variables from newborns' and mothers' clinical records will be performed for eligible patients born between 01/01/2012 and 31/12/2020. The native sequences from fetal magnetic resonance imaging (MRI) will exist nerveless. Data from dissimilar sources will be integrated and analyzed using ML and DL, and forecasting algorithms will exist developed for each outcome. Methods of data augmentation and dimensionality reduction (feature selection and extraction) volition be employed to increase sample size and avoid overfitting. A software system for automatic fetal lung volume segmentation in MRI based on the DL 3D U-NET approach volition also exist developed.

Ethics and dissemination

This retrospective written report received approval from the local ethics commission (Milan Area 2, Italy). The development of predictive models in CDH outcomes volition provide a primal contribution in affliction prediction, early targeted interventions, and personalized direction, with an overall improvement in intendance quality, resources allocation, healthcare, and family savings. Our findings volition be validated in a time to come prospective multicenter accomplice written report.

Registration

The report was registered at ClinicalTrials.gov with the identifier NCT04609163.

Citation: Amodeo I, De Nunzio G, Raffaeli G, Borzani I, Griggio A, Conte L, et al. (2021) A maChine and deep Learning Approach to predict pulmoNary hyperteNsIon in newbornS with congenital diaphragmatic Hernia (CLANNISH): Protocol for a retrospective study. PLoS 1 16(11): e0259724. https://doi.org/10.1371/journal.pone.0259724

Editor: Rogelio Cruz-Martinez, Medicina Fetal Mexico, United mexican states

Received: March 25, 2021; Accustomed: October 25, 2021; Published: Nov 9, 2021

Copyright: © 2021 Amodeo et al. This is an open access article distributed under the terms of the Artistic Eatables Attribution License, which permits unrestricted use, distribution, and reproduction in whatsoever medium, provided the original author and source are credited.

Funding: The writer(s) received no specific funding for this work.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Congenital Diaphragmatic Hernia (CDH) is a rare congenital malformation that affects ane–4 newborns per 10.000 live births, characterized by a diaphragmatic defect, which allows the herniation of the intestinal organs into the thorax [1, 2].

CDH distinctive features are pulmonary hypoplasia and postnatal pulmonary hypertension (PH). The decreased bronchial branching and reduced gas-exchange surface surface area are invariably associated with impaired vascular development, which is characterized by reduced extension and remodeling of the vascular network and altered vasoreactivity [1, 3–eight]. Yet, the pathogenesis of PH has non been fully clarified notwithstanding [9–eighteen]. In add-on, the caste of postnatal respiratory and cardiovascular compromise are fundamental determinants of prognosis [19, 20].

Neonatal survival is approximately 70% merely varies from over xc% in mild CDH to limited chances of survival (less than 10%) in farthermost forms depending on several factors, such every bit defect side and size, associated anomalies, gestational age at birth, and treatment [21–24]. In improver, survivors may have chronic lung disease, persistent PH, feeding and growth bug, gastroesophageal reflux, hearing loss, neurocognitive delay, thoracic deformations, and hernia recurrence [25].

Early detection of fetuses with limited chances of survival allows cosmetic prenatal intervention. Indeed, fetal endoscopic tracheal occlusion (FETO) performed at 27–29 weeks with an endovascular detachable latex balloon (Goldbal two) counteracts the herniation of abdominal organs and promotes lung development through the accumulation of lung fluid, eventually improving survival. Airship removal is normally scheduled at 34 weeks [24, 26]. Results from a multicenter randomized clinical trial investigating the survival increment later the process in left-sided severe and moderate CDH have recently been published (www.totaltrial.eu) [27, 28]. It can likewise exist clinically offered to astringent right-sided CDH [24].

Later birth, treatment strategies include mechanical ventilation, drug therapies, and surgical correction to restore normal beefcake [25, 29, xxx]. Notwithstanding, in instance of failure of conventional therapies, Extracorporeal Membrane Oxygenation (ECMO) may be required [31, 32]. Indeed, CDH represents the master neonatal ECMO indication, with an overall survival charge per unit of nigh 50% of the cases [24, 31, 32]. Besides, ECMO is an invasive and high-risk procedure, possibly associated with acute complications and long-term morbidities [32–34].

Efforts take been made to identify prenatal and postnatal indicators to better outcome prediction and individualized direction. The combined evaluation of lung size, liver position, and defect side through prenatal imaging is conventionally accustomed to stratify CDH fetuses in different groups, correlating with perinatal mortality and long-term morbidity [22, 35]. The calculation of the observed/expected lung-to-head ratio (O/Eastward LHR) with obstetric ultrasound (US) and the observed/expected full fetal lung book (O/Due east TFLV) with fetal magnetic resonance imaging (MRI) are used to assess lung hypoplasia and subsequently predict neonatal outcomes [36]. The size of the defect and hernia sacs may serve as additional predictive tools [36–40]. However, the prognostic accurateness of these parameters has some limitations, especially for postnatal PH [35]. In addition, associated malformations and genetic anomalies take to be ruled out, as they influence the prognosis [24, 41].

Although fetal lung book reflects lung development, it represents simply an approximation of the vascular network's extent and does not consider all the physiological variables influencing vascular resistance [35, 42]. Ideally, we should predict pulmonary hypertension independently from lung volumetry since PH itself correlates with mortality and long-term morbidity [43]. Withal, techniques available for directly measuring lung vascularization are not easily reproducible, and their predictive values remain unclear [44–51]. Moreover, in addition to increased vascular resistance, decreased pulmonary blood menstruum may be affected by vessel distortion caused past visceral herniation itself, which makes the measurement of prognostic parameters less accurate [42]. Finally, the hemodynamic changes occurring during the transition from fetal to neonatal life make the exact prediction of postnatal pulmonary vascular resistance and perfusion trends challenging [43].

In light of these considerations, the prenatal prediction of postnatal PH still represents the weakest point of the prognostic assessment of CDH [36, 42, 43].

Recently, methodologies based on artificial intelligence (AI) accept been developed to support the analysis of medical data, in detail the traditional Automobile Learning (ML) approach and its modern extension, Deep Learning (DL) [52]. In different only integrated ways, both methods explore the possibility of building forecasting algorithms, starting from acquiring relevant clinical information and using them to predict a specific outcome, anticipate adverse events, guide interventions, and amend the overall quality of intendance [52, 53].

The application of these novel technologies, although still express, is also spreading in the neonatal field [54–lx]. Nevertheless, to the all-time of our knowledge, they have not been successfully applied to CDH newborns nevertheless, and a specific prenatal prediction of the disease event still lacks [61].

Our hypothesis is that combined ML and DL systems in prenatal epoch could let forecasting algorithms for outcome prediction in newborns with CDH.

Methods and assay

Study design

A retrospective data collection and study volition be performed at Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy, involving the Fetal Surgery Center, Pediatric Radiology Service, Pediatric Surgery Unit, and Neonatal Intensive Intendance Unit of measurement (NICU). At the same time, ML and DL data analyses, and the development of prototype predictive models and sectionalization algorithms, volition be started at the Department of Mathematics and Physics of the Università del Salento, Lecce, Italian republic, and at the Department of Physics and Chemical science of the Università degli Studi di Palermo, Palermo, Italy.

Nosotros identified ii main project phases. During the commencement half dozen months, the outset phase includes all authoritative and ethics clearance, the research team'south training, and the collection of retrospective information drove. Moreover, the definition of instrumental parameters for identifying prognostic patterns, the definition of protocols for data management, and the software system design. The second stage involves the car and deep learning analysis and will end eighteen months later (Fig i). The prototype ML- and DL-based predictive prognostic models volition be obtained and integrated into a mixed ML-DL organisation to be then applied to the prospective data. The segmentation software prototype will be used every bit soon as it is available. The software tools volition exist tested and optimized during the project's progress (Fig 1).

Fig 1. Standard protocol items.

Standardized Protocol ItemsRecommendations for Observational Retrospective Written report (SPIROS) menstruum diagram: Schedule for administrative and ethics clearance; research training; retrospective data collection; definition of instrumental parameters for the identification of prognostic patterns; definition of protocols for information management; software system design development of the method for the management of patterns (and data augmentation) in grooming; evolution of the method for optimizing system parameters; fetal MRI information elaboration with manual segmentation; development of semiautomatic sectionalisation software; segmentation operation analysis; fetal MRI information elaboration with semiautomatic segmentation; development of the classification system based on machine learning; development of the classification system based on deep learning; statistical analysis for the coupling of classification methods; statistic and classification analysis; completion of analyses; manuscript writing; result broadcasting; kick-off and project meetings. Unit 1: Milan; Unit two: Lecce; Unit of measurement 3: Palermo.

https://doi.org/10.1371/journal.pone.0259724.g001

Patient interest

Parents of newborns with CDH were non involved in the blueprint, or conduct, or reporting, of our research, just they will exist involved in our dissemination plans.

Study population and sample size

Patients with CDH built-in between 01/01/2012 and 31/12/2020 will be considered for the written report. During the study period, the number of eligible patients will be almost fourscore subjects. Computing a 30% drop-out due to exclusion criteria, we will enroll about 56 subjects with CDH.

Enrollment volition exist performed according to the following criteria (Fig two):

Inclusion criteria (all of these):

• Prenatal diagnosis of CDH;
• Prenatal taking charge of the mother with CDH fetus at a gestational historic period beneath or equal to 30^+vi weeks at our Fetal Surgery Centre;
• Singleton pregnancy;
• Fetuses not enrolled in the TOTAL trial (https://world wide web.totaltrial.eu/);
• Inborn patients admitted to the NICU at birth;

Exclusion criteria:

• Prenatal or postnatal diagnosis of not-isolated CDH, thus associated with other malformations or genetic anomalies.

Fig ii. Study flow chart.

CDH: Congenital diaphragmatic hernia; GA: Gestational age; NICU: Neonatal intensive care unit; MRI: Magnetic resonance imaging; PH: Pulmonary hypertension; FETO: Fetal endoscopic tracheal occlusion; ECMO: Extracorporeal membrane oxygenation.

https://doi.org/10.1371/journal.pone.0259724.g002

Data collection

Clinical and instrumental information regarding the prenatal history and the medical and surgical postnatal course of each patient, based on mothers' and newborns' medical records, volition be collected (Astraia, Astraia Software GmbH; NeoCare, GPI SpA) (S1 Table).

Prenatal ultrasound performed between 25⁺⁰ and 30⁺⁶ weeks of gestation (earlier FETO procedure, in case of prenatal handling) for the following data will exist considered: estimated fetal weight; amniotic fluid; defect side; herniated organs; O/Eastward LHR% (tracing method); grading of hernia severity; doppler parameters (umbilical artery pulsatility index, pulmonary artery pulsatility index, pulmonary artery elevation systolic velocity, pulmonary artery height early diastolic reverse catamenia) [22, 62–65]. Gestational historic period at diagnosis, details about FETO procedure, and pregnancy course will too be recorded. In particular, a favorable response to FETO in terms of survival will exist considered.

Regarding the neonatal course, PH will be the principal focus. In the setting of CDH, PH is divers as elevated pulmonary vascular resistance relative to systemic blood pressure level, based on echocardiographic and clinical parameters [66]. Systolic pulmonary artery force per unit area from tricuspid valve regurgitation, hateful pulmonary avenue force per unit area from pulmonary valve regurgitation, pulmonary artery period, characteristics of the interventricular sept, presence, and characteristics of shunts, will exist obtained from the systematic review of cardiac The states performed bedside during NICU stay. In particular, we volition consider: the primeval echocardiogram after NICU access performed inside the first 24-hour interval of life (T0), the pre- (T1) and postoperative (T2) assessment performed closest to CDH repair, and echocardiogram performed one week after surgery (T3) [67]. Physiological parameters such as systemic arterial pressure, centre rate, oxygen pre-and post-ductal saturation volition be obtained from the electronic monitoring systems records throughout the hospitalization and matched with the echocardiographic assessment to define the presence of CDH-associated PH. Patients with systemic or suprasystemic pulmonary artery pressure volition be categorized every bit having CDH-associated PH.

From each newborn'south electronic medical tape, data regarding the following treatments volition be extracted: mechanical ventilation, oxygen supplementation, pulmonary vasodilators, vasoactive and inotropic support, antibiotics, blood product transfusions.

The vasoactive inotropic score (VIS) will exist calculated hourly, and the maximum VIS volition be recorded at each of the 4 fourth dimension points mentioned above [68, 69].

Trends of laboratory parameters will also be collected, similar complete blood count and differential, hemoglobin, C-reactive poly peptide, bilirubin.

The surgical class, day of the intervention, type of surgical repair, prosthetic patch use, intra- or postoperative complications will be noted. In addition, all ECMO cases and deaths during the hospital stay will be reported. In detail, a favorable response to ECMO in terms of survival to procedure and/or discharge volition be considered.

Multiple checks will be performed to appraise data quality, integrity, and accuracy. Whatever incorrect or inconsistent data will exist reviewed and verified through the revision of medical records.

Data extraction

Data used for the study will be extracted from NeoCare electronic medical record (GPI SpA). The relational database will provide access to data points related to each other. The relational database will exist based on the relational model, and the data will exist stored in related tables. Starting from the initial dataset of the patients enrolled for the written report using Sequential Query Language (SQL), data of interest for the analysis for each patient will exist extracted. Two different types of data will exist extracted:

1. structured data per patient, for which at that place is only i data for each patient;
2. time-serial data for which multiple information per patient will exist extracted, associated with recording engagement and time.

Radiological parameters

The software imaging in employ is Synapse PACS and Synapse 3D (FUJIFILM Medical Systems Us, Inc.). All fetal MRIs performed before long after the The states diagnosis of CDH and before fetal intervention in those fetuses eligible for the procedure (25⁺⁰–35⁺⁶ weeks) volition be considered. In addition, lung volumes, liver volume, mediastinal shift angle (MSA), and apparent diffusion coefficient (ADC) for each patient will be calculated (Figs 3–half-dozen) [69–72].

Lung volumes will be calculated on the T2 HASTE sequences, selecting the plane corresponding to the best image quality, covering the whole thorax on a unmarried acquisition without motion-induced artifacts. On each section, left and right lung areas will exist independently adamant by cartoon freehand regions of involvement (ROIs). The main vessels of the pulmonary hila and mediastinal structures will be excluded. The areas volition be automatically added, multiplied past the sum of slice thickness and intergap by the software, to obtain the unabridged book of each lung. Left and correct volumes will be added to obtain the total fetal lung volume (TFLV). The TFLV volition be expressed as a percentage of the mean standard value expected for gestational age (O/East TFLV%), as determined by Rypens et al. [73].

The liver volume will be calculated with the same technique. The proportion between the liver volume above the diaphragmatic level and the full volume will exist calculated and expressed as a pct of liver herniation (%LH) [74, 75].

The ADC will exist calculated using a freehand ROI on each lung, considering the median, minimum, and maximum values [72].

The mediastinal shift angle (MSA) will exist measured on an axial True-Fisp image at the level of the four-chamber view of the fetal heart. A sagittal midline landmark line will be fatigued from the posterior face of the vertebral body to the mid of the posterior surface of the sternum, dividing the fetal thorax into two symmetric parts. A second landmark line will be traced from the same vertebral point, representing the angle vertex, tangentially to the external wall of the right atrium [69, 71, 76, 77]. The software will automatically summate the corresponding MSA.

Finally, the radiographic pulmonary surface area will be calculated on a digital chest 10-ray performed within 24 hours later birth past tracing the perimeter of the lung outlined by the rib cage and the diaphragm, excluding the mediastinal structures and the herniated organs (Fig seven) [67, 78, 79].

Consequence

Chief event. To build an AI system able to identify prenatally patients who volition develop CDH-associated PH at T0, T1, T2, T3.

Secondary outcome.

1. To build models able to predict:
   1. the favorable response to FETO: survival to NICU belch in patients undergoing the procedure;
   2. the need for ECMO;
   3. the favorable response to ECMO treatment: survival to ECMO and survival to NICU discharge in patients undergoing the procedure;
   4. death.
2. To develop a software system for (semi)automated fetal lung segmentation and volume measurement in MRI.

Analysis

Data will exist analyzed using ML, and DL approaches. This computational analysis belongs to the domain of AI and supervised pattern recognition. ML is a traditional technique that provides algorithms that parse information, learn from data without being explicitly programmed, so apply what they have learned to make informed decisions. The typical ML pipeline consists of multiple sequential steps that perform data acquisition and preprocessing, calculation from the information of a ordinarily big number of variables (also called attributes or features, both domain-specific and "agnostic"), complexity reduction, model training and validation, deployment [eighty]. In the training stage, the ML model learns the possible relationship between the features and a target variable (the consequence of involvement), and in principle, generalizes, becoming able to give the correct value of the target variable even for unseen samples.

DL methods are a newer application of ML. While in classical ML techniques, the most effective features demand to be identified by a domain skillful and/or by laborious trials, the biggest advantage of DL algorithms is that they attempt to larn loftier-level features from data in a direct and incremental way. This eliminates the need of domain expertise and difficult-cadre feature extraction.

The purpose of ML and DL classification methods will exist fetus stratification, i.e., to build models in which patients are classified co-ordinate to the different outcomes of interest (primarily the possible evolution of PH, but likewise response to FETO, need and response to ECMO, decease).

Computing environments will be Python and Matlab/Octave. The computational organization will be based on the combined and multivariate apply of features from clinical and instrumental data acquired in the pre-and postnatal menstruum. For the analysis of MRI images, nosotros will create advert hoc calculating tools that will perform the post-obit steps: import of the native image, (semi)automatic contouring and segmentation of the volumes of interest, calculation of semantic and agnostic feature descriptors, classification by ML and DL methods, using commercial software or preferably open-source libraries.

Later contouring/segmentation, feature descriptors for supervised pattern recognition by ML techniques will be calculated in the volumes of interest (particularly the fetus lung and liver). Initially, expert radiologists will manually execute segmentation, but in progress, we volition also develop a (semi)automatic segmentation software useful to supplement the radiologists' work.

For the partitioning phase, a 3D U-Cyberspace volition be implemented [81]. The 3D U-Net has recently been proposed and has been widely used for volumetric segmentation in medical images due to its outstanding performance. It is an extended version of the previously proposed second U-Cyberspace. Information technology must be said that fully convolutional neural networks (CNNs) like U-Cyberspace have been the dominant approach in automated medical imaging segmentation [82, 83]. This architecture has been designed to work with a very small number of preparation images, and this allows a high-functioning application to many medical imaging problems in which it is like shooting fish in a barrel to have a small number of data. It consists of two architectural parts: contracting path and expanding path. To learn and employ local information, high-resolution 3D features in the contracting path are concatenated with upsampled versions of global depression-resolution 3D features in the expanding path. Through this chain, the network learns to use both high-resolution local features and low-resolution global features. The network volition exist trained and optimized for fetal lung volume sectionalisation in MRI.

To cope with the small sample size effect, we plan to implement data augmentation techniques [84]. Data augmentation increases the amount of available data by adding slightly modified copies of existing data or synthetic data derived from existing ones to the dataset. Usually, data augmentation makes the models more robust and reduces overfitting. Several methods volition be used, such as: exporting the images of each patient from the PACS with different spatial orientations, rototranslations of the images, slightly modifying the computed feature descriptors. Through information augmentation, the sample size—although originally limited past the pocket-sized number of patients–volition abound to the advantage of predictive model grooming.

In the ML arroyo, the calculated features will comprise specific clinical variables (as described in the "Data Collection" and "Radiological Parameters" sections), further semantic features computed from the diagnostic images, and Radiomics agnostic features [85]. As the well-nigh axiomatic effect of herniation is the baloney of organ forms, quantitative shape assay will be used to precisely locate and measure morphological changes.

One time a clinical and imaging characteristic vector is defined, the result will necessarily be a dataset with many variables compared to the number of cases analyzed. This may lead to a peculiar phenomenon known equally the "curse of dimensionality", in which a large number of parameters (features) gives origin to poor classification quality because, as it is well known, the number of samples needed to gauge an capricious function with a high level of accuracy grows exponentially with the number of variables [86]. Therefore, a dimensionality reduction phase will be performed in order to eliminate correlated or irrelevant data and to preserve more discriminant variables, overall reducing dimensionality [87]. Methods such equally the Primary Component Assay (PCA), Contained Component Analysis (ICA), Fisher Linear Discriminant Analysis (FLDA), and various characteristic selection techniques volition exist used. PCA, ICA, and FLDA are known equally extraction methods and allow the computation of a reduced-size set of novel features starting from a large initial fix. On the other manus, feature selection reduces dimensionality by choosing a set of "best" features, i.e., the ones which maximize classification quality, discarding redundant or useless ones. An advantage of feature selection, also dimensionality reduction, is also that by identifying the best features it gets human-understandable insights, to the reward of the so-called ML "explainability" (i.e., reducing the "black box" character that ML applications often accept) [88].

Classification of feature vectors will exist performed through conventional feed-forward Artificial Neural Networks, Support Vector Machines (SVM), and other classification systems [80]. In addition, the usage of Ensemble Learning systems is planned, such as Random Forests. In social club to brand maximum employ of the examples available, training/validation will be performed with a Get out One Patient Out Cantankerous Validation scheme (LOPO-CV). The performance of the trained system will be measured by the Receiver Operating Characteristic (ROC) bend and the sensitivity and specificity of the system (Fig 8).

Fig 8. Machine learning pipeline.

The flowchart illustrates the radiomics workflow starting from multimodal paradigm acquisition. Later on manual or (semi)automatic segmentation/contouring of the volumes of interest (such as fetus lungs and liver), feature descriptors are calculated from VOI shape and texture. Together with a choice of prenatal clinical parameters, the obtained feature vectors are labeled with relevant output variables (due east.1000., presence of postnatal PH, or demand for ECMO…) and, after dimensionality reduction performed to become rid of redundant and useless descriptors, enter the supervised classification/regression model-construction pace, here represented past one of the possible choices, i.e., a multi-layer perceptron bogus neural network. Because of the relatively low number of samples, model training and validation will exist obtained by leave-one-out cross-validation (LOO-CV), and diverse methods, such as the ROC bend, will quantify model quality. The trained and validated ML model will be the pipeline output to be employed in the Decision Support Organization.

https://doi.org/ten.1371/journal.pone.0259724.g008

The nomenclature effectiveness of DL volition as well be evaluated. In this project, four of the all-time-known pre-trained convolutional neural networks (CNNs) will be tested (AlexNet, SqueezeNet, ResNet18, GoogLeNet) (Fig 9) [89]. The fine-tuning training method will be implemented. In detail, different training modalities will be analyzed: three levels of freezing weights and scratch. Furthermore, the CNNs volition be used as characteristic extractors; in this manner, the CNNs will be coupled to linear Back up Vector Machine (SVM) classifiers. The main characteristic of SVM classification is their simplicity in terms of parameters, which allows them to face up circuitous nomenclature problems in which, as in our case, a large number of input features are present.

Fig nine. Schematic of the compages of a convolutional neural network (CNN).

A CNN is composed of several kinds of layers, namely the convolution layers and the pooling layers. One of the well-nigh significant differences between deep networks and other ML algorithms is the use of ReLU equally a transfer role to brand the algorithm faster. So, the outputs generated by the previous levels are "flattened" to transform them into a single vector that tin can be used as an input for the next level. The fully continued layer applies weights to the input generated by the feature analysis to predict an accurate label. Finally, the fully connected output layer and softmax produce the final outputs in order to determine the class to associate with the image.

https://doi.org/10.1371/journal.pone.0259724.g009

Discussion

In the era of precision medicine research, integrating a wide variety of data from unlike sources is of primal importance for the complete take accuse of the patients [90]. The proposed retrospective project will be the first to explore the awarding of AI methods to CDH. Due to the complexity and heterogeneity of CDH patients, this disorder is well suited to an integrated, ML- and DL-based analysis of the multitude of clinical, instrumental, and imaging data deriving from multiple sources.

The apply of AI methods will generate a unique advancement in managing fetal/neonatal patients with CDH. This innovative arroyo will maximize data collection from a rare clinical entity and allow the development of forecasting algorithms to predict neonates' outcomes with CDH prenatally. The early detection of clinical and radiological prognostic factors will provide a peculiar opportunity to implement risk stratification, which is key for proper and timely prenatal counseling and guide resource-intensive procedures such as fetal surgery and neonatal ECMO support. Eventually, the selection of the right candidate for the right process in a timely mode will have a huge bear upon in terms of patient outcomes and resources resource allotment. Identifying the relationship betwixt clinical-radiologic variables with patients' outcomes will further clarify the main determinants of CDH-associated PH and improve sympathize PH pathophysiology.

Moreover, imaging has increased its role in modern medicine over time, concerning the response to handling, assessment of toxicity, and more recently in predicting results that let for guiding therapeutic choices. Although medical imaging is chop-chop growing, bias is still linked to homo estimation of the exam [91]. The sectionalization task is fundamental for organ volume and shape assessment, but commonly used medical imaging software does not generally provide the physician with specific segmentation options; thus, contouring work is manual. Transmission segmentation is a time-consuming procedure that tends to be operator-dependent and prone to errors, making volume and shape measurements made past different physicians often challenging to compare.

The definition of an automatic, or at least a semiautomatic, contouring and segmentation software tool specifically designed for the fetal lung would be relevant in clinical practice for fetal lung book assessment, with obvious touch on for healthcare in terms of standardization upgrade and simplification of the diagnostic process. As it will be the first software specifically created for fetal lung assessment, this innovative tool will better data collection accuracy and create solid AI algorithms. This will optimize medical practice and resource, improve quality of care and outcomes, and benefit patients, families, healthcare professionals, and the public health system. If projection evolution is favorable, nosotros shall extend the segmentation software to liver contouring.

The multidisciplinary approach of this study requires the collaboration of experts and research units from different fields. Indeed, information technology gathers mathematicians, physicists, and clinical perinatal physicians and scientists, with different backgrounds and specific competencies, all equally of import for the purpose of the study.

In lite of these considerations, the present retrospective study could correspond the first step of a possible new enquiry line regarding patients with CDH. Indeed, from a research perspective, nosotros intend to consolidate our results in a prospective, multicenter cohort study based on a larger CDH population. Soon, we also intend to establish a Fetal Surgery/Neonatal Network supported by innovative software tools to enable data-driven decisions.

Our retrospective study is the first to explore the awarding of AI methods to CDH. While ML and DL fall under the broad category of AI, ML is a classical technique that provides algorithms that parse information, learn from data, and then employ what they have learned to brand informed decisions. However, the limits of this technique are almost ever related to the limits of the examples bachelor. In addition to being correct, the latter should be consummate, representing the physical problem to be analyzed in all its modalities. DL methods are a newer application of ML. The DL model usually employs the deep classification skills developed on some other classification problem (pre-trained networks), and this acquired cognition is "tuned" to the examples (usually limited in number) of the specific problem to be analyzed.

Although CDH disease is well suited to an integrated, ML- and DL-based assay, the retrospective design of the study and the rarity of the affliction could limit our findings. In add-on, our results may non represent the illness's intrinsic variability because of the limited sample size, and data collection might be incomplete or non-homogeneous. Nevertheless, data will be mostly obtained from electronic medical records and computerized monitoring systems, thus mitigating the limits of a retrospective data drove.

A loftier-dimensional multivariate problem, with a large number of variables compared to the number of cases analyzed, could be hard to be solved through an ML arroyo due to a peculiar phenomenon known equally the "curse of dimensionality" because the number of samples needed to estimate an capricious function with a high level of accuracy grows exponentially with the number of variables (i.e., dimensionality). The dimensionality reduction approaches listed in the "Analysis" section will contrast these issues.

In recent years, deep neural networks, particularly convolutional neural networks (CNN), have angry bully interest in the field of medical imaging [92]. This is undoubtedly due not simply to the loftier classification performance demonstrated past these methods but also to the ease of carrying out a classification process using these tools. Indeed, the traditional concatenation consisting of preprocessing, characteristic extraction, and training models is entirely replaced by CNNs, including feature extraction in their training process. Still, due to the express number of training examples, a failure of the nomenclature performance of the implemented ML methods could occur.

Despite being limited in size, which is intrinsically related to the nature of the disease, our accomplice volition accept the advantage of presenting a broad spectrum of disease severity, including patients undergoing prenatal treatment and postnatal extracorporeal back up. These patients stand for a deep involvement category due to the illness rarity and the limited number of Institutions performing FETO and neonatal ECMO. Therefore, the heterogeneity of our report population will represent a strength betoken for the report's purposes.

Ethics and broadcasting

The nowadays written report will exist in accord with the principles of skilful clinical practice and the Helsinki Declaration. This written report was canonical past the local ethics commission (Milan Surface area two, Italy) with approval number/ID 800_2020bis. All the same, due to the retrospective nature of the study, informed consent was waived by the Ethics Committee.

The report was registered at ClinicalTrials.gov with the identifier NCT04609163. Confidentiality of information will be guaranteed in accord with the regulation in force.

Once the investigation is completed, nosotros intend to publish our results in a peer-reviewed journal. Moreover, findings will be presented at relevant national and international conferences for fetal surgery, neonatology, pediatrics surgery, pediatric radiology, and computer science. Finally, the total database volition be made free and available to share with the scientific community to farther advance the knowledge in this field and promote collaborations (preserving medical confidentiality and in total respect for patients' privacy).

Somewhen, the ML predicting tools may be presented worldwide to CDH fetal and neonatal referral centers, with the ultimate goal to ease the direction of patients and their family from the prenatal to the postnatal epoch.

Conclusions

The predictive algorithms based on AI will provide an essential contribution for precise result prediction, early targeted interventions, and personalized management of fetuses and newborns with CDH. Furthermore, developing a ML- and DL-based prognostic pattern could enable a more than rational resources allocation, non-invasive diagnostic evaluation, cost-effective and timely therapeutic management with a potential bear upon in terms of improved outcomes, reduced social burden, and economic savings. This volition further advance the development of a precision medicine arroyo in the high-risk perinatal setting.

Supporting data

Acknowledgments

The authors would like to thank all the Neonatal ECMO Team Mangiagalli of the Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico: nurses and neonatologists of the NICU, surgeons of the Section of Pediatric Surgery, anesthesiologists of the Pediatric Anesthesiology and Intensive Intendance Unit, nurses of the operating room.

References

1. 1. Fraga MV, Guttentag Due south. Lung development: embryology, growth, maturation, and developmental biological science. In: Gleason C, Devaskar S, editors. Avery's Diseases of the Newborn. ninth ed. Philadelphia: Elsevier; 2012. p. 571–83.
2. 2. Russo FM, De Coppi P, Allegaert K, Toelen J, Van Der Veeken L, Attilakos G, et al. Current and future antenatal management of isolated congenital diaphragmatic hernia. Seminars in Fetal and Neonatal Medicine. 2017;22(half-dozen):383–90. pmid:29169875
   • View Commodity
   • PubMed/NCBI
   • Google Scholar
3. iii. Kluth D, Kangah R, Reich P, Tenbrinck R, Tibboel D, Lambrecht Westward. Nitrofen-induced diaphragmatic hernias in rats: an animal model. J Pediatr Surg. 1990;25(8):850–4. pmid:2401939
   • View Article
   • PubMed/NCBI
   • Google Scholar
4. 4. Keijzer R, Liu J, Deimling J, Tibboel D, Postal service M. Dual-hit hypothesis explains pulmonary hypoplasia in the nitrofen model of congenital diaphragmatic hernia. The American periodical of pathology. 2000;156(4):1299–306. pmid:10751355
   • View Commodity
   • PubMed/NCBI
   • Google Scholar
5. five. Thébaud B, Tibboel D, Rambaud C, Mercier J-C, Bourbon JR, Dinh-Xuan AT, et al. Vitamin A decreases the incidence and severity of nitrofen-induced built diaphragmatic hernia in rats. American Journal of Physiology-Lung Cellular and Molecular Physiology. 1999;277(ii):L423–L9. pmid:10444537
   • View Article
   • PubMed/NCBI
   • Google Scholar
6. 6. Beurskens LW, Tibboel D, Lindemans J, Duvekot JJ, Cohen-Overbeek TE, Veenma DC, et al. Retinol condition of newborn infants is associated with congenital diaphragmatic hernia. Pediatrics. 2010;126(iv):712–20. pmid:20837596
   • View Article
   • PubMed/NCBI
   • Google Scholar
7. 7. Beurskens LW, Schrijver LH, Tibboel D, Wildhagen MF, Knapen MF, Lindemans J, et al. Dietary vitamin A intake below the recommended daily intake during pregnancy and the chance of congenital diaphragmatic hernia in the offspring. Nascence Defects Res Part A: Clin Mol Teratol. 2013;97(1):lx–vi. pmid:23283831
   • View Article
   • PubMed/NCBI
   • Google Scholar
8. 8. Massaro GD, Massaro D. Germination of pulmonary alveoli and gas-commutation surface area: quantitation and regulation. Annual review of physiology. 1996;58(1):73–92. pmid:8815817
   • View Article
   • PubMed/NCBI
   • Google Scholar
9. 9. Pierro G, Thebaud B. Understanding and treating pulmonary hypertension in congenital diaphragmatic hernia. Seminars in Fetal and Neonatal Medicine. 2014;xix(vi):357–63. pmid:25456753
   • View Article
   • PubMed/NCBI
   • Google Scholar
10. x. Sluiter I, van der Horst I, van der Voorn P, Boerema-de Munck A, Buscop-van Kempen M, de Krijger R, et al. Premature differentiation of vascular shine muscle cells in human congenital diaphragmatic hernia. Experimental and Molecular Pathology. 2013;94(one):195–202. pmid:23018129
    • View Article
    • PubMed/NCBI
    • Google Scholar
11. 11. Alphonse RS, Vadivel A, Fung M, Shelley WC, Critser PJ, Ionescu L, et al. Existence, functional impairment, and lung repair potential of endothelial colony-forming cells in oxygen-induced arrested alveolar growth. Apportionment. 2014;129(21):2144–57. pmid:24710033
    • View Article
    • PubMed/NCBI
    • Google Scholar
12. 12. Acker SN, Mandell EW, Sims-Lucas S, Gien J, Abman SH, Galambos C. Histologic identification of prominent intrapulmonary anastomotic vessels in severe congenital diaphragmatic hernia. The Periodical of pediatrics. 2015;166(1):178–83. pmid:25306189
    • View Article
    • PubMed/NCBI
    • Google Scholar
13. 13. Acker SN, Seedorf GJ, Abman SH, Nozik-Grayck E, Partrick DA, Gien J. Pulmonary artery endothelial jail cell dysfunction and decreased populations of highly proliferative endothelial cells in experimental congenital diaphragmatic hernia. American Journal of Physiology-Lung Cellular and Molecular Physiology. 2013;305(12):L943–L52. pmid:24124189
    • View Article
    • PubMed/NCBI
    • Google Scholar
14. xiv. Baker CD, Blackness CP, Ryan SL, Balasubramaniam V, Abman SH. Cord blood endothelial colony-forming cells from newborns with built diaphragmatic hernia. The Journal of pediatrics. 2013;163(3):905–7. pmid:23684109
    • View Article
    • PubMed/NCBI
    • Google Scholar
15. 15. Sbragia L, Nassr A, Gonçalves F, Schmidt A, Zuliani C, Garcia P, et al. VEGF receptor expression decreases during lung development in built diaphragmatic hernia induced by nitrofen. Brazilian Journal of Medical and Biological Research. 2014;47(2):171–eight. pmid:24519134
    • View Article
    • PubMed/NCBI
    • Google Scholar
16. 16. Chang R, Andreoli South, Ng Y-S, Truong T, Smith SR, Wilson J, et al. VEGF expression is downregulated in nitrofen-induced congenital diaphragmatic hernia. J Pediatr Surg. 2004;39(6):825–8. pmid:15185205
    • View Commodity
    • PubMed/NCBI
    • Google Scholar
17. 17. Shehata SM, Mooi WJ, Okazaki T, El-Banna I, Sharma HS, Tibboel D. Enhanced expression of vascular endothelial growth factor in lungs of newborn infants with built diaphragmatic hernia and pulmonary hypertension. Thorax. 1999;54(5):427–31. pmid:10212108
    • View Article
    • PubMed/NCBI
    • Google Scholar
18. 18. Boucherat O, Franco-Montoya M-L, Delacourt C, Martinovic J, Masse 5, Elie C, et al. Defective angiogenesis in hypoplastic human fetal lungs correlates with nitric oxide synthase deficiency that occurs despite enhanced angiopoietin-2 and VEGF. American Periodical of Physiology-Lung Cellular and Molecular Physiology. 2010;298(6):L849–L56. pmid:20348277
    • View Article
    • PubMed/NCBI
    • Google Scholar
19. nineteen. Snoek KG, Greenough A, Van Rosmalen J, Capolupo I, Schaible T, Ali K, et al. Congenital diaphragmatic hernia: 10-year evaluation of survival, extracorporeal membrane oxygenation, and foetoscopic endotracheal occlusion in four high-volume centres. Neonatology. 2018;113(ane):63–8. pmid:29080897
    • View Article
    • PubMed/NCBI
    • Google Scholar
20. 20. Coughlin MA, Werner NL, Gajarski R, Gadepalli South, Hirschl R, Barks J, et al. Prenatally diagnosed severe CDH: bloodshed and morbidity remain high. J Pediatr Surg. 2016;51(7):1091–5. pmid:26655216
    • View Commodity
    • PubMed/NCBI
    • Google Scholar
21. 21. Harting MT, Lally KP. The congenital diaphragmatic hernia report group registry update. Seminars in Fetal and Neonatal Medicine. 2014;19(half dozen):370–5. pmid:25306471
    • View Article
    • PubMed/NCBI
    • Google Scholar
22. 22. Jani J, Nicolaides One thousand, Keller R, Benachi A, Peralta C, Favre R, et al. Observed to expected lung surface area to caput circumference ratio in the prediction of survival in fetuses with isolated diaphragmatic hernia. Ultrasound in obstetrics & gynecology. 2007;thirty(1):67–71. pmid:17587219
    • View Article
    • PubMed/NCBI
    • Google Scholar
23. 23. Russo FM, De Coppi P, Allegaert One thousand, Toelen J, van der Veeken L, Attilakos G, et al. Current and time to come antenatal management of isolated congenital diaphragmatic hernia. Semin Fetal Neonatal Med. 2017;22(half-dozen):383–90. Epub 2017/11/21. pmid:29169875.
    • View Article
    • PubMed/NCBI
    • Google Scholar
24. 24. Van der Veeken L, Russo FM, De Catte L, Gratacos East, Benachi A, Ville Y, et al. Fetoscopic endoluminal tracheal occlusion and reestablishment of fetal airways for congenital diaphragmatic hernia. Gynecol Surg. 2018;xv(one):ix. pmid:29770109
    • View Article
    • PubMed/NCBI
    • Google Scholar
25. 25. Surgery AAoPSo. Postdischarge follow-up of infants with congenital diaphragmatic hernia. Pediatrics. 2008;121(3):627. pmid:18310215
    • View Article
    • PubMed/NCBI
    • Google Scholar
26. 26. Deprest J, Brady P, Nicolaides K, Benachi A, Berg C, Vermeesch J, et al. Prenatal management of the fetus with isolated congenital diaphragmatic hernia in the era of the TOTAL trial. Seminars in Fetal and Neonatal Medicine. 2014;19(half-dozen):338–48. pmid:25447987
    • View Article
    • PubMed/NCBI
    • Google Scholar
27. 27. Deprest JA, Nicolaides KH, Benachi A, Gratacos E, Ryan G, Persico N, et al. Randomized Trial of Fetal Surgery for Severe Left Diaphragmatic Hernia. N Engl J Med. 2021;385(two):107–18. Epub 20210608. pmid:34106556.
    • View Article
    • PubMed/NCBI
    • Google Scholar
28. 28. Deprest JA, Benachi A, Gratacos E, Nicolaides KH, Berg C, Persico N, et al. Randomized Trial of Fetal Surgery for Moderate Left Diaphragmatic Hernia. N Engl J Med. 2021;385(two):119–29. Epub 20210608. pmid:34106555.
    • View Article
    • PubMed/NCBI
    • Google Scholar
29. 29. Snoek KG, Reiss IK, Greenough A, Capolupo I, Urlesberger B, Wessel L, et al. Standardized postnatal management of infants with congenital diaphragmatic hernia in Europe: the CDH EURO consortium consensus-2015 update. Neonatology. 2016;110(1):66–74. pmid:27077664
    • View Article
    • PubMed/NCBI
    • Google Scholar
30. xxx. Hollinger LE, Buchmiller TL. Long term follow-up in congenital diaphragmatic hernia. Semin Perinatol. 2020;44(1):151171. pmid:31451197
    • View Article
    • PubMed/NCBI
    • Google Scholar
31. 31. Abman SH, Hansmann Yard, Archer SL, Ivy DD, Adatia I, Chung WK, et al. Pediatric Pulmonary Hypertension: Guidelines from the American Heart Association and American Thoracic Gild. Circulation. 2015;132(21):2037–99. pmid:26534956
    • View Article
    • PubMed/NCBI
    • Google Scholar
32. 32. Amodeo I, Di Nardo Yard, Raffaeli G, Kamel S, Macchini F, Amodeo A, et al. Neonatal respiratory and cardiac ECMO in Europe. Eur J Pediatr. 2021. pmid:33547504
    • View Article
    • PubMed/NCBI
    • Google Scholar
33. 33. Short BL. The event of extracorporeal life support on the brain: a focus on ECMO. Semin Perinatol. 2005;29(i):45–50. pmid:15921152
    • View Article
    • PubMed/NCBI
    • Google Scholar
34. 34. Van Meurs Thou, Robbins S, Reed Five, Karr S, Wagner A, Glass P, et al. Congenital diaphragmatic hernia: long-term outcome in neonates treated with extracorporeal membrane oxygenation. The Periodical of pediatrics. 1993;122(half-dozen):893–9. pmid:8501565
    • View Article
    • PubMed/NCBI
    • Google Scholar
35. 35. Russo FM, Eastwood MP, Keijzer R, Al‐Maary J, Toelen J, Van Mieghem T, et al. Lung size and liver herniation predict need for extracorporeal membrane oxygenation but not pulmonary hypertension in isolated built diaphragmatic hernia: systematic review and meta‐analysis. Ultrasound Obstet Gynecol. 2017;49(6):704–13. pmid:27312047
    • View Commodity
    • PubMed/NCBI
    • Google Scholar
36. 36. Cordier A-G, Russo FM, Deprest J, Benachi A. Prenatal diagnosis, imaging, and prognosis in Congenital Diaphragmatic Hernia. Semin Perinatol. 2020;44(1):51163. pmid:31439324
    • View Article
    • PubMed/NCBI
    • Google Scholar
37. 37. Slavotinek AM. The genetics of mutual disorders–congenital diaphragmatic hernia. Eur J Med Genet. 2014;57(viii):418–23. pmid:24793812
    • View Article
    • PubMed/NCBI
    • Google Scholar
38. 38. Kardon One thousand, Ackerman KG, McCulley DJ, Shen Y, Wynn J, Shang L, et al. Congenital diaphragmatic hernias: from genes to mechanisms to therapies. Dis Model Mech. 2017;10(8):955–lxx. pmid:28768736
    • View Article
    • PubMed/NCBI
    • Google Scholar
39. 39. Zamora IJ, Cass DL, Lee TC, Welty S, Cassady CI, Mehollin-Ray AR, et al. The presence of a hernia sac in built diaphragmatic hernia is associated with better fetal lung growth and outcomes. J Pediatr Surg. 2013;48(vi):1165–71. pmid:23845602
    • View Article
    • PubMed/NCBI
    • Google Scholar
40. twoscore. Bouchghoul H, Marty O, Fouquet Five, Cordier AG, Senat MV, Saada J, et al. Congenital diaphragmatic hernia has a better prognosis when associated with a hernia sac. Prenatal Diagnosis. 2018;38(9):638–44. pmid:29956346
    • View Commodity
    • PubMed/NCBI
    • Google Scholar
41. 41. Russo FM, Cordier AG, De Catte L, Saada J, Benachi A, Deprest J, et al. Proposal for standardized prenatal ultrasound cess of the fetus with congenital diaphragmatic hernia by the European reference network on rare inherited and congenital anomalies (ERNICA). Prenatal Diagnosis. 2018;38(9):629–37. pmid:29924391
    • View Article
    • PubMed/NCBI
    • Google Scholar
42. 42. Benachi A, Cordier A-G, Cannie Yard, Jani J. Advances in prenatal diagnosis of congenital diaphragmatic hernia. Seminars in Fetal and Neonatal Medicine. 2014;xix(6):331–vii. pmid:25306469
    • View Article
    • PubMed/NCBI
    • Google Scholar
43. 43. Wong G, Reyes J, Lapidus-Krol Eastward, Chiang M, Humpl T, Al-Faraj M, et al. Pulmonary hypertension in congenital diaphragmatic hernia patients: prognostic markers and long-term outcomes. J Pediatr Surg. 2018;53(5):918–24. pmid:29502796
    • View Commodity
    • PubMed/NCBI
    • Google Scholar
44. 44. Basurto D, Russo FM, Van der Veeken Fifty, Van der Merwe J, Hooper S, Benachi A, et al. Prenatal diagnosis and management of congenital diaphragmatic hernia. All-time Practice & Inquiry Clinical Obstetrics & Gynaecology. 2019;58:93–106. pmid:30772144
    • View Article
    • PubMed/NCBI
    • Google Scholar
45. 45. Fuke South, Kanzaki T, Mu J, Wasada G, Takemura M, Mitsuda N, et al. Antenatal prediction of pulmonary hypoplasia by dispatch time/ejection time ratio of fetal pulmonary arteries by Doppler blood flow velocimetry. American journal of obstetrics and gynecology. 2003;188(1):228–33. pmid:12548222
    • View Article
    • PubMed/NCBI
    • Google Scholar
46. 46. Rizzo G, Capponi A, Chaoui R, Taddei F, Arduini D, Romanini C. Blood flow velocity waveforms from peripheral pulmonary arteries in unremarkably grown and growth‐retarded fetuses. Ultrasound in Obstetrics and Gynecology: The Official Periodical of the International Club of Ultrasound in Obstetrics and Gynecology. 1996;eight(2):87–92. pmid:8883309
    • View Article
    • PubMed/NCBI
    • Google Scholar
47. 47. Sokol J, Shimizu Due north, Bohn D, Doherty D, Ryan G, Hornberger LK. Fetal pulmonary avenue diameter measurements as a predictor of morbidity in antenatally diagnosed built diaphragmatic hernia: a prospective study. American periodical of obstetrics and gynecology. 2006;195(2):470–7. pmid:16777047
    • View Article
    • PubMed/NCBI
    • Google Scholar
48. 48. Ruano R, Aubry K-C, Barthe B, Mitanchez D, Dumez Y, Benachi A. Quantitative analysis of fetal pulmonary vasculature by 3-dimensional power Doppler ultrasonography in isolated congenital diaphragmatic hernia. American journal of obstetrics and gynecology. 2006;195(6):1720–8. pmid:16769020
    • View Article
    • PubMed/NCBI
    • Google Scholar
49. 49. Hernandez‐Andrade East, Thuring‐Jönsson A, Jansson T, Lingman G, Maršál K. Fractional moving claret volume estimation in the fetal lung using power Doppler ultrasound: a reproducibility study. Ultrasound in Obstetrics and Gynecology: The Official Journal of the International Society of Ultrasound in Obstetrics and Gynecology. 2004;23(4):369–73. pmid:15065187
    • View Article
    • PubMed/NCBI
    • Google Scholar
50. 50. Cruz‐Martinez R, Castanon M, Moreno‐Alvarez O, Acosta‐Rojas R, Martinez J, Gratacos E. Usefulness of lung‐to‐head ratio and intrapulmonary arterial Doppler in predicting neonatal morbidity in fetuses with congenital diaphragmatic hernia treated with fetoscopic tracheal occlusion. Ultrasound Obstet Gynecol. 2013;41(1):59–65. pmid:22689226
    • View Article
    • PubMed/NCBI
    • Google Scholar
51. 51. DeKoninck P, Jimenez J, Russo FM, Hodges R, Gratacós Eastward, Deprest J. Cess of pulmonary vascular reactivity to oxygen using partial moving blood volume in fetuses with normal lung development and pulmonary hypoplasia in congenital diaphragmatic hernia. Prenatal diagnosis. 2014;34(10):977–81. pmid:24863699
    • View Article
    • PubMed/NCBI
    • Google Scholar
52. 52. Martin-Sanchez F, Verspoor K. Big data in medicine is driving large changes. Yearb Med Inform. 2014;9(ane):14.
    • View Commodity
    • Google Scholar
53. 53. Ayers B, Wood K, Gosev I, Prasad S. Predicting Survival later on Extracorporeal Membrane Oxygenation using Auto Learning. The Register of Thoracic Surgery. 2020;110(4):1193–200. pmid:32454016
    • View Commodity
    • PubMed/NCBI
    • Google Scholar
54. 54. Masino AJ, Harris MC, Forsyth D, Ostapenko S, Srinivasan 50, Bonafide CP, et al. Car learning models for early sepsis recognition in the neonatal intensive intendance unit using readily available electronic health record data. PLoS One. 2019;14(2):e0212665. pmid:30794638
    • View Article
    • PubMed/NCBI
    • Google Scholar
55. 55. Hamilton EF, Dyachenko A, Ciampi A, Maurel Thousand, Warrick PA, Garite TJ. Estimating risk of severe neonatal morbidity in preterm births nether 32 weeks of gestation. The Journal of Maternal-Fetal & Neonatal Medicine. 2020;33(1):73–lxxx. pmid:29886760
    • View Article
    • PubMed/NCBI
    • Google Scholar
56. 56. Mani S, Ozdas A, Aliferis C, Varol HA, Chen Q, Carnevale R, et al. Medical decision support using automobile learning for early detection of tardily-onset neonatal sepsis. Journal of the American Medical Informatics Association. 2014;21(2):326–36. pmid:24043317
    • View Commodity
    • PubMed/NCBI
    • Google Scholar
57. 57. Brooks B, Olm MR, Firek BA, Baker R, Geller-McGrath D, Reimer SR, et al. The developing premature infant gut microbiome is a major factor shaping the microbiome of neonatal intensive care unit rooms. Microbiome. 2018;six(1):1–12. pmid:29291746
    • View Article
    • PubMed/NCBI
    • Google Scholar
58. 58. Podda Yard, Bacciu D, Micheli A, Bellù R, Placidi K, Gagliardi L. A machine learning arroyo to estimating preterm infants survival: evolution of the Preterm Infants Survival Assessment (PISA) predictor. Sci Rep. 2018;8(1):i–ix. pmid:29311619
    • View Article
    • PubMed/NCBI
    • Google Scholar
59. 59. Cooper JN, Minneci PC, Deans KJ. Postoperative neonatal bloodshed prediction using superlearning. J Surg Res. 2018;221:311–9. pmid:29229144
    • View Article
    • PubMed/NCBI
    • Google Scholar
60. 60. Irles C, González-Pérez G, Carrera Muinos Due south, Michel Macias C, Sanchez Gomez C, Martínez-Zepeda A, et al. Interpretation of neonatal intestinal perforation associated with necrotizing enterocolitis by machine learning reveals new fundamental factors. International journal of environmental research and public wellness. 2018;15(xi):2509. pmid:30423965
    • View Article
    • PubMed/NCBI
    • Google Scholar
61. 61. Callaway DA, Campbell IM, Stover SR, Hernandez-Garcia A, Jhangiani SN, Punetha J, et al. Prioritization of candidate genes for congenital diaphragmatic hernia in a critical region on chromosome 4p16 using a machine-learning algorithm. Journal of pediatric genetics. 2018;7(4):164. pmid:30430034
    • View Commodity
    • PubMed/NCBI
    • Google Scholar
62. 62. DeKoninck P, Gratacos E, Van Mieghem T, Richter J, Lewi P, Ancel AM, et al. Results of fetal endoscopic tracheal occlusion for built diaphragmatic hernia and the set up of the randomized controlled Full trial. Early Hum Dev. 2011;87(9):619–24. pmid:21907109
    • View Article
    • PubMed/NCBI
    • Google Scholar
63. 63. Deprest JA, Flemmer AW, Gratacos E, Nicolaides Thousand. Antenatal prediction of lung volume and in-utero treatment past fetal endoscopic tracheal occlusion in severe isolated built diaphragmatic hernia. Seminars in Fetal and Neonatal Medicine. 2009;14(ane):8–13. pmid:18845492
    • View Commodity
    • PubMed/NCBI
    • Google Scholar
64. 64. Cruz‐Martinez R, Martínez‐Rodríguez K, Nieto‐Castro B, Gámez‐Varela A, Cruz‐Lemini One thousand, Luna‐García J, et al. Longitudinal changes in lung size and intrapulmonary‐artery Doppler during the second half of pregnancy in fetuses with built diaphragmatic hernia. Prenatal Diagnosis. 2019;39(ane):45–51. pmid:30511776
    • View Article
    • PubMed/NCBI
    • Google Scholar
65. 65. Cordier AG, Jani J, Cannie M, Rodó C, Fabietti I, Persico N, et al. Tum position in prediction of survival in left‐sided congenital diaphragmatic hernia with or without fetoscopic endoluminal tracheal occlusion. Ultrasound Obstet Gynecol. 2015;46(two):155–61. pmid:25487417
    • View Article
    • PubMed/NCBI
    • Google Scholar
66. 66. Ferguson DM, Gupta VS, Lally PA, Luco One thousand, Tsao K, Lally KP, et al. Early on, Postnatal Pulmonary Hypertension Severity Predicts Inpatient Outcomes in Built Diaphragmatic Hernia. Neonatology. 2021;118(two):147–54. Epub 20210413. pmid:33849011.
    • View Article
    • PubMed/NCBI
    • Google Scholar
67. 67. Amodeo I, Pesenti Due north, Raffaeli One thousand, Macchini F, Condo V, Borzani I, et al. NeoAPACHE Ii. Relationship Between Radiographic Pulmonary Area and Pulmonary Hypertension, Mortality, and Hernia Recurrence in Newborns With CDH. Front Pediatr. 2021;9:692210. Epub 2021/07/30. pmid:34322463; PubMed Central PMCID: PMC8311172.
    • View Article
    • PubMed/NCBI
    • Google Scholar
68. 68. Gaies MG, Gurney JG, Yen AH, Napoli ML, Gajarski RJ, Ohye RG, et al. Vasoactive-inotropic score every bit a predictor of morbidity and mortality in infants later cardiopulmonary bypass. Pediatr Crit Intendance Med. 2010;xi(2):234–8. pmid:19794327.
    • View Article
    • PubMed/NCBI
    • Google Scholar
69. 69. Amodeo I, Borzani I, Corsani G, Pesenti N, Raffaeli Grand, Macchini F, et al. Fetal MRI mediastinal shift angle and respiratory and cardiovascular pharmacological support in newborns with congenital diaphragmatic hernia. Eur J Pediatr. 2021. pmid:34302257
    • View Article
    • PubMed/NCBI
    • Google Scholar
70. lxx. Yamoto Thou, Iwazaki T, Takeuchi K, Sano Yard, Fukumoto K, Takahashi T, et al. The fetal lung-to-liver signal intensity ratio on magnetic resonance imaging as a predictor of outcomes from isolated congenital diaphragmatic hernia. Pediatr Surg Int. 2018;34(2):161–8. pmid:29018962
    • View Article
    • PubMed/NCBI
    • Google Scholar
71. 71. Romiti A, Viggiano Grand, Savelli South, Salvi S, Vicario R, Vassallo C, et al. Comparing of mediastinal shift angles obtained with ultrasound and magnetic resonance imaging in fetuses with isolated left sided built diaphragmatic hernia. The Journal of Maternal-Fetal & Neonatal Medicine. 2020:1–half dozen. pmid:31973612
    • View Commodity
    • PubMed/NCBI
    • Google Scholar
72. 72. Cannie K, Jani J, De Keyzer F, Roebben I, Dymarkowski Southward, Deprest J. Diffusion‐weighted MRI in lungs of normal fetuses and those with congenital diaphragmatic hernia. Ultrasound in Obstetrics and Gynecology. 2009;34(6):678–86. pmid:19866446
    • View Article
    • PubMed/NCBI
    • Google Scholar
73. 73. Rypens F, Metens T, Rocourt N, Sonigo P, Brunelle F, Quere MP, et al. Fetal lung volume: estimation at MR imaging—initial results. Radiology. 2001;219(i):236–41. pmid:11274563
    • View Article
    • PubMed/NCBI
    • Google Scholar
74. 74. Victoria T, Bebbington MW, Danzer E, Fleck AW, Johnson MP, Dinan D, et al. Use of magnetic resonance imaging in prenatal prognosis of the fetus with isolated left congenital diaphragmatic hernia. Prenatal diagnosis. 2012;32(8):715–23. pmid:22674674
    • View Commodity
    • PubMed/NCBI
    • Google Scholar
75. 75. Ruano R, Lazar D, Cass D, Zamora I, Lee T, Cassady C, et al. Fetal lung volume and quantification of liver herniation past magnetic resonance imaging in isolated congenital diaphragmatic hernia. Ultrasound Obstet Gynecol. 2014;43(6):662–9. pmid:24127326
    • View Article
    • PubMed/NCBI
    • Google Scholar
76. 76. Savelli South, Bascetta S, Carducci C, Carnevale Due east, Caforio L, Romiti A, et al. Fetal MRI cess of mediastinal shift angle in isolated left congenital diaphragmatic hernia: A new postnatal survival predictive tool? Prenatal Diagnosis. 2020;40(1):136–41. pmid:31883153
    • View Article
    • PubMed/NCBI
    • Google Scholar
77. 77. Romiti A, Viggiano M, Conforti A, Valfré L, Ravà L, Ciofi Degli Atti M, et al. Ultrasonographic assessment of mediastinal shift angle (MSA) in isolated left built diaphragmatic hernia for the prediction of postnatal survival. The Journal of Maternal-Fetal & Neonatal Medicine. 2020;33(8):1330–5. pmid:30153757
    • View Article
    • PubMed/NCBI
    • Google Scholar
78. 78. Dassios T, Ali One thousand, Makin East, Bhat R, Krokidis G, Greenough A. Prediction of mortality in newborn infants with astringent congenital diaphragmatic hernia using the chest radiographic thoracic area. Pediatric Disquisitional Care Medicine| Lodge of Critical Care Medicine. 2019;20(6):534–9. pmid:30925570
    • View Article
    • PubMed/NCBI
    • Google Scholar
79. 79. Amodeo I, Raffaeli K, Pesenti Due north, Macchini F, Condò V, Borzani I, et al. The NeoAPACHE Report Protocol I: Assessment of the Radiographic Pulmonary Surface area and Long-Term Respiratory Function in Newborns With Congenital Diaphragmatic Hernia. Frontiers in Pediatrics. 2020;8. pmid:33194913
    • View Article
    • PubMed/NCBI
    • Google Scholar
80. 80. Duda RO, Hart PE, Stork DG. Pattern Classification 2d ed. 605 3rd Avenue New York, NY, Usa: Wiley-Interscience; 2000.
81. 81. Çiçek Ö, Abdulkadir A, Lienkamp SS, Brox T, Ronneberger O. 3D U-Internet: learning dense volumetric sectionalization from sparse annotation. In: Ourselin S., Joskowicz L., Sabuncu Grand., Unal K., Due west. W, editors. Medical Image Computing and Computer-Assisted Intervention–MICCAI 2016 MICCAI 2016 Lecture Notes in Computer Science, vol 9901. 9901: Springer; 2016. p. 424–32.
82. 82. Ronneberger O, Fischer P, Brox T. U-Cyberspace: Convolutional Networks for Biomedical Prototype Sectionalization. Springer International Publishing; 2015. p. 234–41.
83. 83. Huang C, Han H, Yao Q, Zhu Due south, Zhou SK. 3D U2-Net: A 3D Universal U-Net for Multi-Domain Medical Image Segmentation. Lecture Notes in Computer Science: Springer International Publishing; 2019.
84. 84. Shorten C, Khoshgoftaar TM. A survey on Prototype Information Augmentation for Deep Learning. Journal of Big Data. 2019;half-dozen(1). pmid:31700766
    • View Article
    • PubMed/NCBI
    • Google Scholar
85. 85. Lambin P, Rios-Velazquez E, Leijenaar R, Carvalho S, Van Stiphout RGPM, Granton P, et al. Radiomics: Extracting more information from medical images using avant-garde feature analysis. Eur J Cancer. 2012;48(4):441–6. pmid:22257792
    • View Article
    • PubMed/NCBI
    • Google Scholar
86. 86. Bellman R. Adaptive command processes: a guided bout princeton university press. Princeton, New Jersey, United states. 1961:96.
87. 87. Liu H, Motoda H. Feature extraction, structure and selection: A data mining perspective: Springer Science & Concern Media; 1998.
88. 88. Wanner J, Herm L-V, Janiesch C. How much is the black box? The value of explainability in machine learning models. 2020.
    • View Article
    • Google Scholar
89. 89. Taormina 5, Cascio D, Abbene Fifty, Raso G. Performance of Fine-Tuning Convolutional Neural Networks for HEp-2 Epitome Classification. Applied Sciences. 2020;10(nineteen):6940.
    • View Article
    • Google Scholar
90. 90. Au-Yong-Oliveira M, Pesqueira A, Sousa MJ, Dal Mas F, Soliman M. The Potential of Big Data Enquiry in HealthCare for Medical Doctors' Learning. J Med Syst. 2021;45(1):i–14. pmid:33409620
    • View Article
    • PubMed/NCBI
    • Google Scholar
91. 91. Chung C, Kalpathy-Cramer J, Knopp MV, Jaffray DA. In the Era of Deep Learning, Why Reconstruct an Image at All? J Am Coll Radiol. 2021;18(one):170–three. pmid:33413895
    • View Article
    • PubMed/NCBI
    • Google Scholar
92. 92. Lu L, Zheng Y, Carneiro One thousand, Yang L. Deep learning and convolutional neural networks for medical epitome computing. Precision Medicine, High Performance and Big-Calibration Datasets. Part of the Advances in Computer Vision and Pattern Recognition book series (ACVPR) 102017. p. 978–3.
    • View Commodity
    • Google Scholar

lambertonhinume.blogspot.com

Source: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0259724