2026-01-21 スタンフォード大学
Stanford Universityの研究チームは、人工知能(AI)を用いて早産児の健康転帰や合併症リスクを高精度に予測できる新しいアルゴリズムを開発した。早産は新生児死亡や長期的な健康障害の主要因であり、個々のリスクを事前に把握することが重要とされている。本研究では、妊娠中および出生直後の臨床データを統合し、呼吸障害、神経発達障害、感染症などの発生確率を予測するAIモデルを構築した。その結果、従来手法よりも高い精度で予後を推定でき、医師が治療や介入の優先順位を判断する支援が可能となった。研究は、AIが周産期医療における意思決定を補完し、早産児の長期的健康改善につながる可能性を示している。
Getty Images
<関連情報>
- https://news.stanford.edu/stories/2026/01/ai-algorithm-premature-birth-health-outcomes-complications-research
- https://www.science.org/doi/10.1126/scitranslmed.adv4942
乾燥血液スポット代謝プロファイルとディープラーニングを用いた新生児の健康の定量的評価 Quantitative assessment of neonatal health using dried blood spot metabolite profiles and deep learning
Alan L. Chang, Jonathan D. Reiss, Anthony Culos, Martin Becker, […] , and Nima Aghaeepour
Science Translational Medicine Published:21 Jan 2026
DOI:https://doi.org/10.1126/scitranslmed.adv4942
Editor’s summary
Infants born prematurely can experience severe associated conditions such as necrotizing enterocolitis, but risk-stratifying newborns for the development of these conditions is challenging. Here, Chang and colleagues applied deep learning methods to routine metabolic information in the form of dried blood spots collected within the first 48 hours after birth from very premature infants in California with linked outcomes data to develop a metabolic health index, which risk-stratified the infants for four severe conditions of prematurity better than other machine learning approaches or clinical factors. Their approach developed a biological risk metric of prematurity that, because newborn dried blood spot screening for inborn errors of metabolism is routine in many countries, could be broadly generalizable. —Melissa L. Norton
Abstract
Neonatal prematurity leads to considerable morbidity and mortality, partly because of acquired conditions such as bronchopulmonary dysplasia (BPD), intraventricular hemorrhage (IVH), necrotizing enterocolitis (NEC), and retinopathy of prematurity (ROP). Standard gestational age and birthweight-based classifications of prematurity inadequately capture the variation in newborns’ health outcomes, creating an urgent need to develop risk stratification tools for vulnerable newborn infants to initiate the most appropriate care pathways as early as possible. We hypothesized that the metabolic profiles of newborn infants capture additional risk information beyond current measures. A total of 13,536 newborn screening (NBS) blood spot tests from preterm infants in California with linked clinical outcomes of prematurity were used to develop an NBS-based metabolic health index to stratify preterm infants at risk for BPD, IVH, NEC, and ROP (12,096 cases with one or more conditions and 1440 controls) through a deep learning model that provides a single index score in tandem with subgroup discovery to identify individuals with the strongest metabolite biomarker signals for adverse outcomes of prematurity. This metabolic health index captured risk signals that were distinct from gestational age and birthweight and outperformed other machine learning algorithms and clinical risk variable-based models in stratifying at-risk individuals for adverse outcomes of prematurity. The metabolic health index was externally validated in an independent retrospective cohort of 3299 very premature newborns from Ontario, Canada (2117 cases and 1182 controls), which recapitulated common metabolic risk subgroups. In summary, combining widespread metabolite screening with deep learning established a generalizable biological risk metric of prematurity.
