Pulmonary hypoplasia, or underdeveloped lungs, is a final common pathway shared by many fetal diseases. Congenital diaphragmatic hernia, in which abdominal organs push into the chest, is the best-known example, but kidney malformations with low amniotic fluid, bladder outlet obstruction, bronchopulmonary malformations, and skeletal dysplasias can all leave a fetus with lungs too small to sustain life after birth. Fetal magnetic resonance imaging provides a way to quantify the problem: by tracing the lungs slice by slice, radiologists can calculate total lung volume and compare it with expected values for gestational age, producing numbers that guide decisions about delivery planning and even fetal surgery.

The bottleneck has always been the tracing itself. Manual segmentation of the lungs on fetal MRI typically takes more than thirty minutes, and longer when the anatomy is distorted, the fetus moves during the scan, or amniotic fluid levels are abnormal. Tertiary fetal imaging centers often perform only two to five fetal MRI examinations per day, partly because of this labor requirement. Existing automated models offered little help beyond the most common diagnosis, because most were trained exclusively on congenital diaphragmatic hernia cases or structurally normal fetuses, leaving the many other causes of pulmonary hypoplasia without a reliable tool.

To build a more broadly applicable system, the researchers assembled a retrospective dataset of fetal MRI examinations performed between January 2016 and March 2025. Of 1,203 examinations identified, 906 had complete, recoverable expert-adjudicated lung segmentations that had been created during routine clinical care and used for actual patient counseling. The cohort spanned gestational ages from 16 to 38 weeks, with a mean of 27.1 weeks, and included the full spectrum of pulmonary hypoplasia etiologies. The data were split into 724 training and 182 independent test examinations for total lung volume, with a smaller subset of 196 examinations, including thoracic cavity and mediastinal labels, used for a second model predicting percent predicted lung volume.

Technically, the team used nnU-Net v2, a self-configuring segmentation framework that automatically determines preprocessing steps such as image resampling, intensity normalization, and patch size. Models were trained on an NVIDIA A100 graphics processing unit using five-fold cross-validation with both two-dimensional and three-dimensional full-resolution configurations, and final predictions combined the two model types through an ensemble. Notably, the researchers deliberately worked with the clinically acquired coronal T2-weighted single-shot fast spin-echo image stacks rather than reconstructed three-dimensional volumes. These source stacks are available at the time of interpretation, require no extra reconstruction step, and are less dependent on slice-to-volume reconstruction, a design choice intended to support rapid, pragmatic deployment in real clinics.

The results were striking for total lung volume. The automated model achieved a mean Dice similarity coefficient of 0.81, indicating strong spatial overlap with expert contours, and an intraclass correlation coefficient of 0.955 with manual volumetry, a level conventionally classified as excellent. Bland-Altman analysis showed minimal systematic bias of about one milliliter, with 95 percent limits of agreement between roughly minus 15 and plus 13 milliliters. Median absolute percentage error was 11.1 percent, close to the ten percent benchmark the authors selected as approximating inter-reader variability in fetal lung volumetry. Average inference time was 3.8 seconds per examination, roughly a 450-fold speedup over manual tracing.

The second model, which estimates percent predicted lung volume by additionally segmenting the mediastinum and thoracic cavity, performed less well. Its Dice coefficient averaged 0.76, its intraclass correlation coefficient was 0.727, and median absolute percentage error rose to 14.5 percent. The authors attribute this to the compounded difficulty of segmenting multiple structures that are variably distorted by herniated abdominal contents, a thymus that resembles hypoplastic lungs on T2-weighted images, and mass effects that shift the chest anatomy. At the clinically meaningful level of risk-category assignment, automated observed-to-expected total lung volume matched manual assignments in 86.8 percent of test examinations, with most disagreements occurring between adjacent severity categories, whereas percent predicted lung volume agreed in only 54.1 percent of cases.

Performance was remarkably stable across the disease spectrum. The total lung volume model performed best on skeletal dysplasia, left-sided congenital diaphragmatic hernia, and renal disease, and remained reliable in the presence of severe motion artifact, oligohydramnios, and polyhydramnios. It struggled most with very large bronchopulmonary malformations whose lesion volume exceeded that of the lungs themselves, where the model sometimes mistook the lesion for lung tissue and overestimated lung volume. Accuracy also dipped in fetuses scanned before 20 weeks of gestation, though it improved with advancing gestational age, precisely when lung volume metrics matter most for predicting neonatal outcomes and planning interventions.

Crucially, the system is designed for human oversight rather than autonomous use. In a simulated human-in-the-loop workflow, an attending fetal radiologist reviewed and corrected the 25 lowest-performing outputs, those with a Dice coefficient below 0.70, in an average of just 3.1 minutes per case. Most corrections involved removing labels erroneously assigned to bronchopulmonary malformations or adjusting boundaries along the diaphragm and herniated contents. Combining four-second automated segmentation with a few minutes of expert adjudication still cuts the total workflow time roughly tenfold, potentially allowing fetal lung volumetry to be delivered on the same day as imaging, when counseling and multidisciplinary decisions are made.

The implications extend beyond the clinic. Scalable, reproducible lung segmentations are a prerequisite for radiomics-based research, in which small boundary inaccuracies can propagate into unreliable quantitative features across large heterogeneous datasets. The authors caution that their study was retrospective and single-institution, that risk-category thresholds were derived from left-sided congenital diaphragmatic hernia outcome data and may not generalize to other etiologies, and that external validation across institutions, scanners, and vendors remains necessary. Still, by collapsing a thirty-minute expert task into seconds while preserving expert review for the hardest cases, the work represents a concrete step toward making quantitative fetal lung assessment a routine, same-day part of prenatal care.

Subject of Research: Automated deep learning segmentation of fetal MRI lungs for rapid lung volume estimation in pulmonary hypoplasia

Article Title: Automated fetal magnetic resonance imaging lung segmentation for rapid lung volume estimation across diverse pulmonary hypoplasia phenotypes

Article References: Choi, J. J., Hegde, S., Dudley, J., Galganski, L. A., Kline-Fath, B. M., Somasundaram, E., & Dillman, J. R. (2026). Automated fetal magnetic resonance imaging lung segmentation for rapid lung volume estimation across diverse pulmonary hypoplasia phenotypes. Pediatric Radiology. https://doi.org/10.1007/s00247-026-06765-1

Image Credits: AI Generated

DOI: 10.1007/s00247-026-06765-1

Keywords: fetal MRI, lung segmentation, deep learning, pulmonary hypoplasia, congenital diaphragmatic hernia, nnU-Net, lung volume, prenatal diagnosis, artificial intelligence, medical imaging, pediatric radiology, human-in-the-loop

The immediate consequence of congenital diaphragmatic hernia is structural. A hole in the diaphragm, the muscular partition separating the chest from the abdomen, allows abdominal organs such as the stomach, liver, and intestines to migrate into the chest cavity during fetal development. The crowding effect is severe: the developing lungs are compressed at precisely the window of gestation when their airways and vascular trees should be branching and expanding. The result is pulmonary hypoplasia, lungs that are smaller, simpler in architecture, and less able to support gas exchange after birth. But the mechanical compression is only part of the story. Newborns with this condition frequently develop persistent pulmonary hypertension, in which the resistance of the pulmonary circulation is so high that the heart struggles to pump blood through the lungs, leading to critical oxygen deprivation that can be fatal even with aggressive intervention.

The pulmonary hypertension associated with the condition has always hinted at a vascular problem, and vascular problems invite scrutiny of the endothelium. Endothelial cells form the inner lining of every blood vessel, and in the pulmonary circulation they do far more than serve as a passive barrier. They regulate vascular tone by producing nitric oxide, a potent vasodilator; they orchestrate angiogenesis, the growth of new vessels; they modulate inflammation and blood clotting; and they communicate constantly with the smooth muscle cells that wrap around vessels and control their caliber. In pulmonary hypertension of any origin, endothelial dysfunction is a recurring theme, and congenital diaphragmatic hernia is no exception. Previous work has described abnormal vascular remodeling in the lungs of affected infants, including medial thickening of pulmonary arteries and aberrant extension of muscle into distal, normally non-muscularized vessels. What the new study adds is a specific mechanistic suspect: disturbed energy metabolism within the endothelial cells themselves.

Bioenergetics refers to the set of biochemical pathways by which a cell converts nutrients into usable chemical energy, chiefly in the form of adenosine triphosphate, or ATP. For most cells in the body, the mitochondria are the power plants, and their efficiency determines whether a cell thrives or struggles under stress. Endothelial cells are unusual in this respect. Unlike many other cell types, healthy endothelial cells generate the majority of their ATP through glycolysis, the anaerobic breakdown of glucose, even when oxygen is abundantly available. This glycolytic preference is not a quirk; it is functionally important. It spares oxygen for the surrounding tissue, positions the endothelium to survive in low-oxygen environments during vessel sprouting, and produces biosynthetic intermediates needed for the rapid proliferation of new vessel lining during angiogenesis. Any disruption to this finely tuned metabolic balance can therefore ripple outward, impairing nitric oxide production, promoting oxidative stress, and destabilizing the vessel wall.

It is against this background that the reported alterations in endothelial bioenergetics take on significance. The study investigated how endothelial cells in the setting of congenital diaphragmatic hernia differ in their energy-generating machinery, examining markers of mitochondrial function and respiratory activity. Cells under metabolic stress typically exhibit a recognizable signature: diminished mitochondrial respiratory capacity, altered balance between oxidative phosphorylation and glycolysis, elevated production of reactive oxygen species, and reduced ability to adapt when energy demands spike. Each of these changes can feed a vicious cycle relevant to pulmonary hypertension. Mitochondrial dysfunction impairs the enzymes that synthesize nitric oxide, and when the endothelial nitric oxide synthase enzyme becomes uncoupled, it can actually generate superoxide instead of the protective vasodilator molecule. The resulting oxidative burden damages proteins, lipids, and DNA inside the vessel wall, encouraging the very remodeling and vasoconstriction that define hypertensive lung circulation.

The clinical stakes of this line of investigation are considerable. Infants with congenital diaphragmatic hernia are among the most intensively supported patients in neonatal intensive care, often requiring high-frequency ventilation, inhaled nitric oxide therapy, and in severe cases extracorporeal membrane oxygenation, a heart-lung bypass machine that takes over gas exchange while the infant’s own circulation stabilizes. Despite these measures, mortality remains substantial, particularly in cases diagnosed early with severe lung involvement. One of the most difficult problems clinicians face is predicting which infants will deteriorate and tailoring the intensity of support accordingly. If endothelial bioenergetic status proves to correlate with disease severity, metabolic markers could eventually supplement current predictors of outcome, giving intensive care teams a biochemical window into the state of the pulmonary vasculature that imaging and blood gas measurements cannot fully provide.

There is also a therapeutic dimension to consider. Metabolism has moved to the center of vascular biology research over the past two decades, and drugs that modulate mitochondrial function or cellular energy pathways already exist for other conditions. The concept of a metabolically targeted therapy for pulmonary vascular disease is no longer speculative in principle; several metabolic modulators, including agents that influence fatty acid oxidation and mitochondrial dynamics, are under investigation for pulmonary hypertension more broadly. If the endothelial energy deficit identified in congenital diaphragmatic hernia can be corrected, or even partially compensated, it could complement existing treatments that work through entirely different mechanisms. Inhaled nitric oxide, for example, acts downstream to relax vessels, but it does nothing to repair the underlying endothelial dysfunction that limits natural nitric oxide production. A therapy aimed at restoring mitochondrial health would attack the problem at its cellular source.

Careful interpretation remains essential. The connection between altered endothelial bioenergetics and clinical outcomes in this disease is an association that must now be interrogated further. Key questions include whether the metabolic changes observed are a cause of the pulmonary vascular pathology or a consequence of it, whether they are present before birth or emerge postnatally under the stress of intensive care, and whether they can be detected reliably in accessible clinical samples such as blood or in cells grown from patient-derived material. Animal models of the condition, including the nitrofen-induced rodent model widely used in the field, will likely play an important role in establishing causality, since they allow researchers to examine lung vessels at defined developmental stages that are difficult or impossible to access in human fetuses. The heterogeneity of the disease, which ranges from mild cases detected incidentally to lethal ones diagnosed prenatally, adds another layer of complexity to any attempt at generalization.

For the broader research community, the study sits at the intersection of two fields that have been converging rapidly: developmental vascular biology and cellular metabolism. The same bioenergetic principles that govern tumor angiogenesis and wound healing apply to the construction of the fetal pulmonary circulation, and perturbations of those principles during gestation may leave lasting functional fingerprints. Congenital diaphragmatic hernia, precisely because its vascular pathology is so pronounced and its clinical course so well documented, offers a natural setting in which to test whether endothelial energetics shape developmental outcomes. As follow-up work refines these findings, the hope is that a condition whose treatment has long been defined by supportive care will begin to yield to targeted, mechanism-based interventions, giving newborns with this devastating diagnosis a better chance at healthy, unrestricted breathing from their very first days.

Subject of Research: Endothelial cell bioenergetic alterations in congenital diaphragmatic hernia

Article Title: Alterations of endothelial cell bioenergetics in congenital diaphragmatic hernia

Article References: Emrick, B. F., Zhevlakova, I., Novotny, M., Mavrakis, L., Mulya, A., Cass, D. L., Miyasaka, E., Byzova, T. V., Asosingh, K., Erzurum, S. C., & Robertson, J. O. (2026). Alterations of endothelial cell bioenergetics in congenital diaphragmatic hernia. Pediatric Research. https://doi.org/10.1038/s41390-026-05447-w

Image Credits: AI Generated

DOI: 10.1038/s41390-026-05447-w

Keywords: congenital diaphragmatic hernia, endothelial cells, bioenergetics, mitochondria, pulmonary hypertension, pulmonary hypoplasia, nitric oxide, neonatal care, vascular remodeling, pediatric research, Alterations, endothelial