A new noninvasive technology, called broadband optical spectroscopy (BOS), has promise to reliably detect necrotizing enterocolitis (NEC) in premature infants at earlier stages, before this devastating intestinal disease progresses enough to be visible on x-ray, according to a first-in-human study from Ann & Robert H. Lurie Children’s Hospital of Chicago published in the Journal of Pediatric Surgery.

NEC strikes suddenly and progresses quickly. At early stages, it can be treated with antibiotics, while more advanced disease often requires surgery and carries a high risk of complications and death. So far, there has been no reliable way to detect it early.

“NEC is one of the most feared diagnoses in any neonatal intensive care unit, and the field has been searching for an early detection tool for decades,” said senior author Seth Goldstein, MD, MPhil, pediatric general and thoracic surgeon at Lurie Children’s and Associate Professor of Surgery at Northwestern University Feinberg School of Medicine. “We are excited that our new handheld device reliably distinguishes NEC from healthy tissue within two minutes, and that it can be used at the bedside without subjecting these fragile babies to any pain or radiation. We hope that additional studies confirm our findings that this tool can screen for early stages of NEC.”

The noninvasive BOS device placed gently on the infant’s abdomen measures infrared light reflecting off the intestines and is able to detect color changes that surgeons recognize as indicative of threatened tissue in NEC.

Dr. Goldstein, working with lead author and research fellow Ashley Dodd, MD, and colleagues, partnered with Vadim Backman’s biomedical engineering lab at Northwestern University to develop BOS.

After achieving prior success using BOS for NEC detection in a mouse model, researchers tested the device in 96 premature infants who were younger than 36 weeks of gestation and did not have congenital cardiac conditions or abdominal wall defects.

“Our results show that BOS is a safe and feasible technology that produced detectable signal changes in premature infants with NEC,” said Dr. Goldstein. “These findings suggest that BOS is a promising and potentially groundbreaking modality for screening and early detection of NEC.”

This research was supported by The Gerber Foundation, Catalyst award from the Center of Physical Genomics at Northwestern University and philanthropic support from Rob and Kristin Goldman and the Christina Carinato Charitable Foundation. Fundings: National Institutes of Health Grants (NIH) grants U54CA268084, R01CA228272, and R01CA225002. Support also came from the Eunice Kennedy Shriver National Institute of Child Health & Human Development of the National Institutes of Health under Award Number T32HD111386. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Ann & Robert H. Lurie Children’s Hospital of Chicago is a nonprofit organization committed to providing access to exceptional care for every child. It is the only independent, research-driven children’s hospital in Illinois and one of less than 35 nationally. This is where the top doctors go to train, practice pediatric medicine, teach, advocate, research and stay up to date on the latest treatments. Exclusively focused on children, all Lurie Children’s resources are devoted to serving their needs. Research at Lurie Children’s is conducted through Stanley Manne Children’s Research Institute, which is focused on improving child health, transforming pediatric medicine and ensuring healthier futures through the relentless pursuit of knowledge. Lurie Children’s is the pediatric training ground for Northwestern University Feinberg School of Medicine. It is ranked as one of the nation’s top children’s hospitals by U.S. News & World Report.

Journal

Journal of Pediatric Surgery

DOI

10.1016/j.jpedsurg.2026.162978

bu içeriği en az 2000 kelime olacak şekilde ve alt başlıklar ve madde içermiyecek şekilde ünlü bir science magazine için İngilizce olarak yeniden yaz. Teknik açıklamalar içersin ve viral olacak şekilde İngilizce yaz. Haber dışında başka bir şey içermesin. Haber içerisinde en az 12 paragraf ve her bir paragrafta da en az 50 kelime olsun. Cevapta sadece haber olsun. Ayrıca haberi yazdıktan sonra içerikten yararlanarak aşağıdaki başlıkların bilgisi var ise haberin altında doldur. Eğer yoksa bilgisi ilgili kısmı yazma.:
Subject of Research:
Article Title:
News Publication Date:
Web References:
References:
Image Credits:

Keywords

Biliary atresia is a severe neonatal hepatic disorder characterized by obstruction or absence of bile ducts, leading to progressive liver damage and eventual liver failure if untreated. The current diagnostic gold standard involves invasive surgical exploration or cholangiography, which carries risks and delays crucial intervention. Early detection is paramount, as timely surgical treatment dramatically enhances survival rates and can obviate the need for liver transplantation. Against this backdrop, the introduction of an AI-empowered, minimally invasive diagnostic system could shift paradigms in clinical management.

The research team, comprising experts in hepatology, pediatric surgery, and data science, meticulously developed a diagnostic algorithm calibrated on a diverse cohort of infants suspected of having biliary atresia. Employing advanced machine learning techniques, the model integrates multidimensional clinical data, laboratory parameters, and imaging characteristics to discern BA from other causes of neonatal cholestasis. This integrative approach leverages the nuanced patterns and subtle biomarkers often imperceptible to human observers but computable by AI.

Technically, the model is rooted in deep learning architectures, likely convolutional neural networks, optimized for pattern recognition across ultrasound images and serological data. The diagnostic pipeline was rigorously trained and validated against a robust dataset, ensuring high sensitivity and specificity. Emphasizing reproducibility and generalizability, the dataset included multi-center inputs, reflecting diverse patient demographics and clinical presentations essential for real-world applicability.

The model’s performance metrics are particularly impressive. It achieved diagnostic accuracy surpassing that of conventional clinical algorithms and competitor AI models tested on similar datasets. Sensitivity and specificity metrics indicate that the algorithm minimizes false negatives—a critical aspect to ensure no cases pending urgent treatment are overlooked—and also reduces false positives, preventing unnecessary invasive procedures. These balanced trade-offs underscore the AI system’s clinical reliability.

From a minimization of invasiveness standpoint, the model relies primarily on non-invasive imaging and accessible laboratory tests rather than surgical or endoscopic techniques. This dramatically reduces patient discomfort, procedural risks, and healthcare costs. Moreover, the AI-based method shortens the diagnostic timeline, permitting faster clinical decisions and potential initiation of treatment within the narrow therapeutic window that defines biliary atresia management success.

The interdisciplinary collaboration highlighted in this work showcases the fusion of cutting-edge computational science with pediatric hepatology, underscoring how AI is no longer a futuristic concept but an immediate translational tool within clinical workflows. By harnessing AI’s ability to process vast, complex datasets rapidly and accurately, clinicians are empowered to diagnose challenging cases with unprecedented precision and timeliness.

One of the study’s critical contributions is its emphasis on explainability and clinician integration. The AI model is designed with transparency features that provide interpretable outputs, ensuring that healthcare providers can understand and trust the diagnostic suggestions. This addresses a common barrier in AI adoption—the ‘black box’ problem—thereby facilitating acceptance and smoother implementation in medical settings.

While the research demonstrates groundbreaking potential, the authors prudently acknowledge the necessity for further extensive clinical trials and cross-population validations. Such future studies are crucial to fine-tune the model’s predictive capabilities, address any biases inherent in initial training data, and verify efficacy across varied healthcare infrastructures globally. The promising preliminary results, however, signal a paradigm shift in pediatric diagnostics.

Additionally, this AI application aligns with broader movements in precision medicine, where individualized diagnostic and therapeutic strategies are increasingly emphasized. The model’s capacity to analyze granular patient-specific data and guide personalized clinical decisions exemplifies this trend. It illustrates how digital innovations can systematically transform not only diagnosis but also overall patient management strategies.

The societal impact of this development may also extend beyond immediate clinical practice. By potentially reducing the need for liver transplants and long hospitalizations, healthcare systems could see significant economic benefits. Families endure less anxiety and trauma associated with invasive procedures and delayed diagnoses, improving overall psychological and quality-of-life outcomes for affected infants and their caregivers.

A critical dimension is the ethical framework surrounding AI deployment in pediatric care. The researchers have underscored rigorous data privacy standards and ensured algorithmic fairness in design to mitigate disparities in healthcare access or outcomes. This proactive approach is vital for maintaining public trust and ethical integrity as AI tools become increasingly embedded in sensitive and high-stakes medical environments.

This study not only opens new frontiers for biliary atresia but also exemplifies how AI can be strategically leveraged in rare disease diagnostics where clinical uncertainty and procedural risks are high. Its success could inspire similar innovations across other pediatric conditions characterized by diagnostic challenges, setting a new benchmark in child health technology integration.

In summary, the development and validation of this minimally invasive AI diagnostic model herald a new chapter for biliary atresia management. Marrying accuracy with safety and efficiency, it offers hope for dramatically improving patient outcomes. As AI technology continues to evolve and embed itself deeper within clinical settings, such transformative approaches underscore the immense potential residing at the intersection of artificial intelligence and pediatric healthcare.

Subject of Research: Development and validation of a minimally invasive diagnostic model for biliary atresia using artificial intelligence

Article Title: Development and validation of a minimally invasive diagnostic model for biliary atresia using artificial intelligence

Article References:
Jiang, JY., Dong, R., Sun, YH. et al. Development and validation of a minimally invasive diagnostic model for biliary atresia using artificial intelligence. World J Pediatr (2025). https://doi.org/10.1007/s12519-025-00988-2

Image Credits: AI Generated

DOI: 10.1007/s12519-025-00988-2