Machine learning prediction of feeding intolerance in preterm infants: a pre-feeding risk stratification - published 05 2025

Q: What does the document focus on regarding feeding intolerance in preterm infants?

It focuses on using machine learning to predict feeding intolerance and perform risk stratification before feeding.

Q: How is model performance described?

Performance is reported using discrimination metrics such as AUC with confidence intervals, plus statistical comparisons across model approaches.

Q: Why is pre-feeding risk stratification important in the study?

It enables earlier identification of infants at higher risk, supporting closer monitoring and more timely clinical decisions.

Aug 4, 2026

Upload by Evangeline

Machine learning models are applied to predict feeding intolerance in preterm infants and to support pre-feeding risk stratification. The study evaluates model performance using cohort outcomes and reports discrimination metrics including AUC with accompanying confidence intervals, alongside calibration and statistical comparisons between model variants. Findings indicate that the proposed approach can meaningfully separate risk groups for infants with higher likelihood of developing feeding intolerance, providing a data-driven basis for earlier clinical decision-making and targeted monitoring.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages9

Reading Time23 min

AccessFree to download

Introduction

Methods and data

Prediction model and risk stratification

Results and performance metrics

Discussion and conclusions

T􀀚PE O􀀃􀀂g􀀂􀀋􀀊􀀘 R􀀈􀀄􀀈􀀊􀀃 􀀅 PUB􀀜ISHED 05 􀀕􀀈􀀉􀀑􀀈mb􀀈􀀃 2025 DOI 10.3389/􀀔􀀉􀀈􀀌 .2025.1646973 􀀁􀀂􀀃􀀄􀀁􀀂 􀀆􀀇 􀀁􀀂􀀃􀀂􀀄􀀅 􀀇􀀈􀀄􀀅􀀉􀀊􀀋􀀌􀀈􀀍 􀀎􀀅􀀈 􀀏􀀋􀀂􀀐􀀈􀀃􀀄􀀂􀀑􀀒 􀀓􀀔 􀀕􀀒􀀌􀀋􀀈􀀒􀀍 􀀖􀀗􀀄􀀑􀀃􀀊􀀘􀀂􀀊 􀀈􀀁􀀉􀀃􀀁􀀊􀀁􀀂 􀀆􀀇 􀀙􀀗􀀒􀀊􀀓 􀀚􀀂􀀗􀀍 􀀛􀀂􀀃􀀄􀀑 􀀖􀀔􀀜􀀘􀀂􀀊􀀑􀀈􀀌 􀀝􀀓􀀄􀀉􀀂􀀑􀀊􀀘 􀀓􀀔 􀀖􀀋􀀅􀀗􀀂 􀀞􀀈􀀌􀀂 􀀊􀀘􀀏􀀋􀀂􀀐􀀈􀀃􀀄􀀂􀀑􀀒􀀍 C􀀅􀀂􀀋􀀊 􀀞􀀊􀀃􀀂􀀋􀀊 􀀞􀀊􀀒􀀗m􀀂 V􀀈􀀋􀀌􀀃􀀊m􀀈 􀀎􀀊k􀀊􀀓􀀍􀀕􀀑􀀊􀀑􀀈 􀀏􀀋􀀂􀀐􀀈􀀃􀀄􀀂􀀑􀀒 􀀓􀀔 C􀀊m􀀉􀀂􀀋􀀊􀀄􀀍 B􀀃􀀊z􀀂􀀘 *􀀋􀀌􀀈􀀈􀀁􀀍􀀎􀀌􀀏􀀂􀀁􀀏􀀋􀀁 􀀙􀀂􀀋g 􀀚􀀂 j􀀂z􀀅􀀈􀀋􀀘􀀒@163 . 􀀓m 􀀈􀀁􀀋􀀁􀀃􀀉􀀁􀀂 14 J􀀗􀀋􀀈 2025 􀀐􀀋􀀋􀀁􀀎􀀄􀀁􀀂 22 􀀖􀀗g􀀗􀀄􀀑 2025 􀀎􀀑􀀆􀀒􀀃􀀍􀀓􀀁􀀂 05 􀀕􀀈􀀉􀀑􀀈mb􀀈􀀃 2025 􀀋􀀃􀀄􀀐􀀄􀀃􀀌􀀏 􀀞􀀊􀀓 􀀁􀀍 􀀚􀀂 􀀙􀀍 􀀚􀀂 􀀞􀀍 W􀀊􀀋g J 􀀊􀀋􀀌 􀀚􀀂 􀀙 (2025)􀀞􀀊 􀀅􀀂􀀋􀀈 􀀘􀀈􀀊􀀃􀀋􀀂􀀋g 􀀉􀀃􀀈􀀌􀀂 􀀑􀀂􀀓􀀋 􀀓􀀔 􀀔􀀈􀀈􀀌􀀂􀀋g 􀀂􀀋􀀑􀀓􀀘􀀈􀀃􀀊􀀋 􀀈 􀀂􀀋 􀀉􀀃􀀈􀀑􀀈􀀃m 􀀂􀀋􀀔􀀊􀀋􀀑􀀄: 􀀊 􀀉􀀃􀀈-􀀔􀀈􀀈􀀌􀀂􀀋g 􀀃􀀂􀀄k 􀀄􀀑􀀃􀀊􀀑􀀂􀀜 􀀊􀀑􀀂􀀓􀀋 m􀀓􀀌􀀈􀀘 . 􀀛􀀃􀀓􀀋􀀑 . P􀀈􀀌􀀂􀀊􀀑􀀃 . 13:1646973 . 􀀌􀀓􀀂: 10.3389/􀀔􀀉􀀈􀀌 .2025.1646973 􀀋􀀌􀀎􀀇􀀈􀀃􀀔􀀓􀀄 © 2025 􀀞􀀊􀀓􀀍 􀀚􀀂􀀍 􀀚􀀂􀀍 W􀀊􀀋g 􀀊􀀋􀀌 􀀚􀀂 . 􀀎􀀅􀀂􀀄 􀀂􀀄 􀀊􀀋􀀓􀀉􀀈􀀋-􀀊 􀀈􀀄􀀄 􀀊􀀃􀀑􀀂 􀀘􀀈 􀀌􀀂􀀄􀀑􀀃􀀂b􀀗􀀑􀀈􀀌 􀀗􀀋􀀌􀀈􀀃 􀀑􀀅􀀈􀀑􀀈􀀃m􀀄 􀀓􀀔 􀀑􀀅􀀈 C􀀃􀀈􀀊􀀑􀀂􀀐􀀈 C􀀓mm􀀓􀀋􀀄 􀀖􀀑􀀑􀀃􀀂b􀀗􀀑􀀂􀀓􀀋􀀚􀀂 􀀈􀀋􀀄􀀈 (CC B􀀙) . 􀀎􀀅􀀈 􀀗􀀄􀀈􀀍 􀀌􀀂􀀄􀀑􀀃􀀂b􀀗􀀑􀀂􀀓􀀋 􀀓􀀃􀀃􀀈􀀉􀀃􀀓􀀌􀀗 􀀑􀀂􀀓􀀋 􀀂􀀋 􀀓􀀑􀀅􀀈􀀃 􀀔􀀓􀀃􀀗m􀀄 􀀂􀀄 􀀉􀀈􀀃m􀀂􀀑􀀑􀀈􀀌􀀍􀀉􀀃􀀓􀀐􀀂􀀌􀀈􀀌 􀀑􀀅􀀈 􀀓􀀃􀀂g􀀂􀀋􀀊􀀘 􀀊􀀗􀀑􀀅􀀓􀀃(􀀄) 􀀊􀀋􀀌 􀀑􀀅􀀈 􀀓􀀉􀀒􀀃􀀂g􀀅􀀑 􀀓w􀀋􀀈􀀃(􀀄) 􀀊􀀃􀀈 􀀃􀀈􀀌􀀂􀀑􀀈􀀌 􀀊􀀋􀀌 􀀑􀀅􀀊􀀑 􀀑􀀅􀀈􀀓􀀃􀀂g􀀂􀀋􀀊􀀘 􀀉􀀗b􀀘􀀂 􀀊􀀑􀀂􀀓􀀋 􀀂􀀋 􀀑􀀅􀀂􀀄 j􀀓􀀗􀀃􀀋􀀊􀀘 􀀂􀀄 􀀂􀀑􀀈􀀌􀀍 􀀂􀀋􀀊 􀀓􀀃􀀌􀀊􀀋 􀀈 w􀀂􀀑􀀅 􀀊 􀀈􀀉􀀑􀀈􀀌 􀀊 􀀊􀀌􀀈m􀀂 􀀉􀀃􀀊 􀀑􀀂 􀀈 . N􀀓 􀀗􀀄􀀈􀀍 􀀌􀀂􀀄􀀑􀀃􀀂b􀀗􀀑􀀂􀀓􀀋 􀀓􀀃 􀀃􀀈􀀉􀀃􀀓􀀌􀀗 􀀑􀀂􀀓􀀋 􀀂􀀄􀀉􀀈􀀃m􀀂􀀑􀀑􀀈􀀌 w􀀅􀀂 􀀅 􀀌􀀓􀀈􀀄 􀀋􀀓􀀑 􀀓m􀀉􀀘􀀒 w􀀂􀀑􀀅􀀑􀀅􀀈􀀄􀀈 􀀑􀀈􀀃m􀀄 . 􀀁􀀂􀀃􀀄􀀅􀀆􀀇 􀀉􀀇􀀂􀀊􀀆􀀅􀀆􀀋 􀀌􀀊􀀇􀀍􀀅􀀃􀀎􀀅􀀏􀀆 􀀏􀀐􀀐􀀇􀀇􀀍􀀅􀀆􀀋 􀀅􀀆􀀎􀀏􀀉􀀇􀀊􀀂􀀆􀀃􀀇 􀀅􀀆 􀀌􀀊􀀇􀀎􀀇􀀊􀀑􀀅􀀆􀀐􀀂􀀆􀀎􀀒􀀓 􀀂 􀀌􀀊􀀇􀀔􀀐􀀇􀀇􀀍􀀅􀀆􀀋 􀀊􀀅􀀒􀀕􀀒􀀎􀀊􀀂􀀎􀀅􀀖􀀃􀀂􀀎􀀅􀀏􀀆 􀀑􀀏􀀍􀀇􀀉 􀀗􀀂􀀅 􀀁􀀂􀀏􀀘􀀙 􀀚􀀛􀀇 􀀜􀀅 􀀘􀀙 􀀁􀀅􀀆 􀀜􀀅 􀀘􀀙 􀀝􀀅􀀆 􀀞􀀂􀀆􀀋 􀀂􀀆􀀍 􀀚􀀅􀀆􀀋 􀀜􀀅 * 1 􀀇􀀈􀀉􀀊􀀃􀀑m􀀈􀀋􀀑 􀀓􀀔 􀀎􀀃􀀊􀀌􀀂􀀑􀀂􀀓􀀋􀀊􀀘 C􀀅􀀂􀀋􀀈􀀄􀀈 􀀞􀀈􀀌􀀂 􀀂􀀋􀀈􀀍 C􀀊􀀉􀀂􀀑􀀊􀀘 C􀀈􀀋􀀑􀀈􀀃 􀀔􀀓􀀃 C􀀅􀀂􀀘􀀌􀀃􀀈􀀋’􀀄 􀀝􀀈􀀊􀀘􀀑􀀅􀀍 C􀀊􀀉􀀂􀀑􀀊􀀘 􀀞􀀈􀀌􀀂 􀀊􀀘􀀏􀀋􀀂􀀐􀀈􀀃􀀄􀀂􀀑􀀒􀀍 C􀀊􀀉􀀂􀀑􀀊􀀘 I􀀋􀀄􀀑􀀂􀀑􀀗􀀑􀀈 􀀓􀀔 P􀀈􀀌􀀂􀀊􀀑􀀃􀀂 􀀄􀀍 B􀀈􀀂j􀀂􀀋g􀀍 C􀀅􀀂􀀋􀀊􀀍 2 􀀇􀀈􀀉􀀊􀀃􀀑m􀀈􀀋􀀑 􀀓􀀔 N􀀈􀀓􀀋􀀊􀀑􀀓􀀘􀀓g􀀒􀀍 C􀀊􀀉􀀂􀀑􀀊􀀘 C􀀈􀀋􀀑􀀈􀀃􀀔􀀓􀀃 C􀀅􀀂􀀘􀀌􀀃􀀈􀀋’􀀄 􀀝􀀈􀀊􀀘􀀑􀀅􀀍 C􀀊􀀉􀀂􀀑􀀊􀀘 􀀞􀀈􀀌􀀂 􀀊􀀘 􀀏􀀋􀀂􀀐􀀈􀀃􀀄􀀂􀀑􀀒􀀍 C􀀊􀀉􀀂􀀑􀀊􀀘 I􀀋􀀄􀀑􀀂􀀑􀀗􀀑􀀈 􀀓􀀔 P􀀈􀀌􀀂􀀊􀀑􀀃􀀂 􀀄􀀍 B􀀈􀀂j􀀂􀀋g􀀍 C􀀅􀀂􀀋􀀊 􀀆􀀕􀀖􀀗􀀘􀀙􀀚􀀛􀀜􀀝􀀞 􀀛􀀈􀀈􀀌􀀂􀀋g 􀀂􀀋􀀑􀀓􀀘􀀈􀀃􀀊􀀋 􀀈 (􀀛I) 􀀃􀀈􀀉􀀃􀀈􀀄􀀈􀀋􀀑􀀄 􀀊 􀀉􀀃􀀈􀀐􀀊􀀘􀀈􀀋􀀑 􀀊􀀋􀀌 􀀄􀀈􀀃􀀂􀀓􀀗􀀄 􀀓m􀀉􀀘􀀂 􀀊􀀑􀀂􀀓􀀋 􀀂􀀋 􀀉􀀃􀀈􀀑􀀈􀀃m 􀀂􀀋􀀔􀀊􀀋􀀑􀀄􀀍 􀀓􀀋􀀑􀀃􀀂b􀀗􀀑􀀂􀀋g 􀀑􀀓 􀀌􀀈􀀘􀀊􀀒􀀈􀀌 􀀈􀀋􀀑􀀈􀀃􀀊􀀘 􀀋􀀗􀀑􀀃􀀂􀀑􀀂􀀓􀀋􀀍􀀉􀀃􀀓􀀘􀀓􀀋g􀀈􀀌 􀀅􀀓􀀄􀀉􀀂􀀑􀀊􀀘􀀂z􀀊􀀑􀀂􀀓􀀋􀀍 􀀊􀀋􀀌 􀀂􀀋 􀀃􀀈􀀊􀀄􀀈􀀌 m􀀓􀀃b􀀂􀀌􀀂􀀑􀀒 . E􀀊􀀃􀀘􀀒 􀀂􀀌􀀈􀀋􀀑􀀂􀀜 􀀊􀀑􀀂􀀓􀀋 􀀓􀀔 􀀅􀀂g􀀅 -􀀃􀀂􀀄k 􀀂􀀋􀀔􀀊􀀋􀀑􀀄 􀀃􀀈m􀀊􀀂􀀋􀀄 􀀅􀀊􀀘􀀘􀀈􀀋g􀀂􀀋g 􀀌􀀗􀀈 􀀑􀀓 􀀘􀀂m􀀂􀀑􀀈􀀌 􀀉􀀃􀀈􀀌􀀂 􀀑􀀂􀀐􀀈 􀀑􀀓􀀓􀀘􀀄 􀀊􀀐􀀊􀀂􀀘􀀊b􀀘􀀈 b􀀈􀀔􀀓􀀃􀀈􀀔􀀈􀀈􀀌􀀂􀀋g 􀀂􀀋􀀂􀀑􀀂􀀊􀀑􀀂􀀓􀀋 . eth􀀚􀀝s􀀞 W􀀈 􀀓􀀋􀀌􀀗 􀀑􀀈􀀌 􀀊 􀀃􀀈􀀑􀀃􀀓􀀄􀀉􀀈 􀀑􀀂􀀐􀀈 􀀓􀀅􀀓􀀃􀀑 􀀄􀀑􀀗􀀌􀀒 􀀓􀀔 402 􀀉􀀃􀀈􀀑􀀈􀀃m 􀀂􀀋􀀔􀀊􀀋􀀑􀀄(<37 w􀀈􀀈k􀀄 g􀀈􀀄􀀑􀀊􀀑􀀂􀀓􀀋􀀊􀀘 􀀊g􀀈) 􀀊􀀌m􀀂􀀑􀀑􀀈􀀌 b􀀈􀀑w􀀈􀀈􀀋 J􀀊􀀋􀀗􀀊􀀃􀀒 2023 􀀊􀀋􀀌 􀀞􀀊􀀒 2024 . C􀀘􀀂􀀋􀀂 􀀊􀀘 􀀌􀀊􀀑􀀊 􀀓􀀘􀀘􀀈 􀀑􀀈􀀌 􀀊􀀑 􀀊􀀌m􀀂􀀄􀀄􀀂􀀓􀀋 􀀗􀀋􀀌􀀈􀀃w􀀈􀀋􀀑 􀀔􀀈􀀊􀀑􀀗􀀃􀀈 􀀄􀀈􀀘􀀈 􀀑􀀂􀀓􀀋 􀀗􀀄􀀂􀀋g 􀀃􀀓􀀄􀀄 -􀀐􀀊􀀘􀀂􀀌􀀊􀀑􀀈􀀌 􀀚􀀖􀀕􀀕O 􀀃􀀈g􀀃􀀈􀀄􀀄􀀂􀀓􀀋 . E􀀘􀀈􀀐􀀈􀀋 m􀀊 􀀅􀀂􀀋􀀈 􀀘􀀈􀀊􀀃􀀋􀀂􀀋g 􀀊􀀘g􀀓􀀃􀀂􀀑􀀅m􀀄 w􀀈􀀃􀀈􀀄􀀒􀀄􀀑􀀈m􀀊􀀑􀀂 􀀊􀀘􀀘􀀒 􀀓m􀀉􀀊􀀃􀀈􀀌 􀀗􀀄􀀂􀀋g 􀀊 􀀗􀀃􀀊 􀀒􀀍 􀀊􀀃􀀈􀀊 􀀗􀀋􀀌􀀈􀀃 􀀑􀀅􀀈 􀀃􀀈 􀀈􀀂􀀐􀀈􀀃 􀀓􀀉􀀈􀀃􀀊􀀑􀀂􀀋g 􀀅􀀊􀀃􀀊 􀀑􀀈􀀃􀀂􀀄􀀑􀀂 􀀗􀀃􀀐􀀈 (􀀖􀀏C)􀀍 􀀄􀀈􀀋􀀄􀀂􀀑􀀂􀀐􀀂􀀑􀀒􀀍 􀀊􀀋􀀌 􀀄􀀉􀀈 􀀂􀀜 􀀂􀀑􀀒 . C􀀘􀀂􀀋􀀂 􀀊􀀘 􀀗􀀑􀀂􀀘􀀂􀀑􀀒 w􀀊􀀄 􀀊􀀄􀀄􀀈􀀄􀀄􀀈􀀌 􀀑􀀅􀀃􀀓􀀗g􀀅 􀀌􀀈 􀀂􀀄􀀂􀀓􀀋 􀀗􀀃􀀐􀀈 􀀊􀀋􀀊􀀘􀀒􀀄􀀂􀀄 (􀀇C􀀖) . 􀀈es􀀛lts􀀞 􀀛I 􀀌􀀈􀀐􀀈􀀘􀀓􀀉􀀈􀀌 􀀂􀀋 199 (49 . 5%) 􀀂􀀋􀀔􀀊􀀋􀀑􀀄 . 􀀕􀀂g􀀋􀀂􀀜 􀀊􀀋􀀑 b􀀈􀀑w􀀈􀀈􀀋-g􀀃􀀓􀀗􀀉􀀌􀀂􀀔􀀔􀀈􀀃􀀈􀀋 􀀈􀀄 w􀀈􀀃􀀈 􀀓b􀀄􀀈􀀃􀀐􀀈􀀌 􀀔􀀓􀀃 b􀀂􀀃􀀑􀀅 w􀀈􀀂g􀀅􀀑􀀍 g􀀈􀀄􀀑􀀊􀀑􀀂􀀓􀀋􀀊􀀘 􀀊g􀀈􀀍 􀀑􀀂m􀀈 􀀑􀀓 􀀜􀀃􀀄􀀑􀀔􀀈􀀈􀀌􀀂􀀋g􀀍 􀀔􀀈􀀑􀀊􀀘 􀀌􀀂􀀄􀀑􀀃􀀈􀀄􀀄􀀍 m􀀗􀀘􀀑􀀂􀀉􀀘􀀈 g􀀈􀀄􀀑􀀊􀀑􀀂􀀓􀀋􀀍 􀀉􀀃􀀈􀀋􀀊􀀑􀀊􀀘 􀀌􀀈x􀀊m􀀈􀀑􀀅􀀊􀀄􀀓􀀋􀀈 􀀈x􀀉􀀓􀀄􀀗􀀃􀀈􀀍􀀋􀀈􀀓􀀋􀀊􀀑􀀊􀀘 􀀂􀀋􀀔􀀈 􀀑􀀂􀀓􀀋􀀍 􀀃􀀈􀀄􀀉􀀂􀀃􀀊􀀑􀀓􀀃􀀒 􀀌􀀂􀀄􀀑􀀃􀀈􀀄􀀄􀀍 􀀊􀀋􀀌 􀀂􀀋􀀐􀀊􀀄􀀂􀀐􀀈 m􀀈 􀀅􀀊􀀋􀀂 􀀊􀀘 􀀐􀀈􀀋􀀑􀀂􀀘􀀊􀀑􀀂􀀓􀀋 (􀀊􀀘􀀘􀀁 < 0 . 01) . 􀀚􀀖􀀕􀀕O 􀀃􀀈g􀀃􀀈􀀄􀀄􀀂􀀓􀀋 􀀂􀀌􀀈􀀋􀀑􀀂􀀜􀀈􀀌 14 􀀓􀀉􀀑􀀂m􀀊􀀘 􀀉􀀃􀀈􀀌􀀂 􀀑􀀂􀀐􀀈 􀀐􀀊􀀃􀀂􀀊b􀀘􀀈􀀄 . 􀀖m􀀓􀀋g 􀀑􀀈􀀄􀀑􀀈􀀌 􀀊􀀘g􀀓􀀃􀀂􀀑􀀅m􀀄􀀍 􀀖􀀌􀀊B􀀓􀀓􀀄􀀑 􀀌􀀈m􀀓􀀋􀀄􀀑􀀃􀀊􀀑􀀈􀀌 􀀄􀀗􀀉􀀈􀀃􀀂􀀓􀀃 􀀉􀀈􀀃􀀔􀀓􀀃m􀀊􀀋 􀀈 [􀀊 􀀗􀀃􀀊 􀀒: 0. 957; 􀀖􀀏C: 0 . 964 (95% CI: 0 .929–1. 000); 􀀄􀀈􀀋􀀄􀀂􀀑􀀂􀀐􀀂􀀑􀀒: 0 . 957; 􀀄􀀉􀀈 􀀂􀀜 􀀂􀀑􀀒: 0. 958] . 􀀇C􀀖 􀀓􀀋􀀜􀀃m􀀈􀀌 g􀀃􀀈􀀊􀀑􀀈􀀃 􀀋􀀈􀀑 􀀘􀀂􀀋􀀂 􀀊􀀘 b􀀈􀀋􀀈􀀜􀀑 􀀓m􀀉􀀊􀀃􀀈􀀌 􀀑􀀓 “􀀑􀀃􀀈􀀊􀀑 􀀊􀀘􀀘” 􀀓􀀃“􀀑􀀃􀀈􀀊􀀑 􀀋􀀓􀀋􀀈” 􀀄􀀑􀀃􀀊􀀑􀀈g􀀂􀀈􀀄 . 􀀖􀀋 􀀂􀀋􀀑􀀈􀀃􀀊 􀀑􀀂􀀐􀀈 􀀘􀀂􀀋􀀂 􀀊􀀘 􀀌􀀈 􀀂􀀄􀀂􀀓􀀋 􀀄􀀗􀀉􀀉􀀓􀀃􀀑 􀀑􀀓􀀓􀀘 w􀀊􀀄􀀌􀀈􀀐􀀈􀀘􀀓􀀉􀀈􀀌 􀀔􀀓􀀃 􀀉􀀃􀀊 􀀑􀀂 􀀊􀀘 􀀂m􀀉􀀘􀀈m􀀈􀀋􀀑􀀊􀀑􀀂􀀓􀀋 . 􀀋􀀚􀀜􀀖l􀀛si􀀚􀀜s􀀞 􀀎􀀅􀀈 􀀉􀀃􀀓􀀉􀀓􀀄􀀈􀀌 m􀀊 􀀅􀀂􀀋􀀈 􀀘􀀈􀀊􀀃􀀋􀀂􀀋g m􀀓􀀌􀀈􀀘 􀀊 􀀗􀀃􀀊􀀑􀀈􀀘􀀒 􀀉􀀃􀀈􀀌􀀂 􀀑􀀄􀀔􀀈􀀈􀀌􀀂􀀋g 􀀂􀀋􀀑􀀓􀀘􀀈􀀃􀀊􀀋 􀀈 b􀀈􀀔􀀓􀀃􀀈 􀀜􀀃􀀄􀀑 􀀔􀀈􀀈􀀌􀀂􀀋g 􀀗􀀄􀀂􀀋g 14 􀀃􀀓􀀗􀀑􀀂􀀋􀀈􀀘􀀒 􀀓􀀘􀀘􀀈 􀀑􀀈􀀌 􀀘􀀂􀀋􀀂 􀀊􀀘􀀐􀀊􀀃􀀂􀀊b􀀘􀀈􀀄 . 􀀎􀀅􀀂􀀄 􀀊􀀉􀀉􀀃􀀓􀀊 􀀅 􀀈􀀋􀀊b􀀘􀀈􀀄 􀀈􀀊􀀃􀀘􀀒 􀀃􀀂􀀄k 􀀄􀀑􀀃􀀊􀀑􀀂􀀜 􀀊􀀑􀀂􀀓􀀋 􀀊􀀋􀀌 m􀀊􀀒 􀀂m􀀉􀀃􀀓􀀐􀀈 􀀘􀀂􀀋􀀂 􀀊􀀘 􀀓􀀗􀀑 􀀓m􀀈􀀄 􀀑􀀅􀀃􀀓􀀗g􀀅 􀀑􀀂m􀀈􀀘􀀒 􀀂􀀋􀀑􀀈􀀃􀀐􀀈􀀋􀀑􀀂􀀓􀀋 . Ex􀀑􀀈􀀃􀀋􀀊􀀘 􀀐􀀊􀀘􀀂􀀌􀀊􀀑􀀂􀀓􀀋 􀀂􀀋 m􀀗􀀘􀀑􀀂 􀀈􀀋􀀑􀀈􀀃 􀀓􀀅􀀓􀀃􀀑􀀄 􀀂􀀄 w􀀊􀀃􀀃􀀊􀀋􀀑􀀈􀀌 􀀑􀀓 􀀓􀀋􀀜􀀃m g􀀈􀀋􀀈􀀃􀀊􀀘􀀂z􀀊b􀀂􀀘􀀂􀀑􀀒 . K􀀁􀀇􀀊􀀌􀀈􀀂􀀍 fee􀀝i􀀜􀀘 i􀀜t􀀚le􀀙􀀕􀀜􀀖e, p􀀙ete􀀙m i􀀜f􀀕􀀜ts, m􀀕􀀖hi􀀜e le􀀕􀀙􀀜i􀀜􀀘, 􀀙is􀀗 p􀀙e􀀝i􀀖ti􀀚􀀜, 􀀕􀀝􀀕b􀀚􀀚st, e􀀕􀀙lyi􀀜te􀀙ve􀀜ti􀀚􀀜 I􀀆􀀎􀀊􀀏􀀍􀀛􀀃􀀎􀀅􀀏􀀆 􀀁􀀂􀀂􀀃􀀄􀀅􀀆 􀀄􀀅􀀈􀀉􀀊􀀂􀀋􀀌􀀅􀀍􀀂 􀀎􀀁􀀏􀀐 􀀄􀀑 􀀌 􀀍􀀉􀀒􀀒􀀉􀀅 􀀌􀀅􀀃 􀀍􀀊􀀄􀀅􀀄􀀍􀀌􀀊􀀊􀀓 􀀑􀀄􀀆􀀅􀀄􀀔􀀍􀀌􀀅􀀈 􀀍􀀉􀀒􀀕􀀊􀀄􀀍􀀌􀀈􀀄􀀉􀀅 􀀌􀀒􀀉􀀅􀀆􀀕􀀋􀀂􀀈􀀂􀀋􀀒 􀀄􀀅􀀖􀀌􀀅􀀈􀀑􀀗 􀀘􀀄􀀈􀀙 􀀆􀀊􀀉􀀚􀀌􀀊 􀀄􀀅􀀍􀀄􀀃􀀂􀀅􀀍􀀂 􀀂􀀑􀀈􀀄􀀒􀀌􀀈􀀂􀀑 􀀋􀀌􀀅􀀆􀀄􀀅􀀆 􀀖􀀋􀀉􀀒 􀀛􀀜􀀝􀀞 9􀀝 􀀎􀀛􀀐 . 􀀁􀀏􀀗 􀀃􀀂􀀔􀀅􀀂􀀃􀀌􀀑 􀀈􀀙􀀂 􀀄􀀅􀀌􀀚􀀄􀀊􀀄􀀈􀀓 􀀈􀀉 􀀈􀀉􀀊􀀂􀀋􀀌􀀈􀀂 􀀂􀀅􀀈􀀂􀀋􀀌􀀊 􀀅u􀀈􀀋􀀄􀀈􀀄􀀉􀀅􀀗 􀀄􀀑 􀀍􀀙􀀌􀀋􀀌􀀍􀀈􀀂􀀋􀀄z􀀂􀀃 􀀚􀀓 􀀄􀀅􀀍􀀋􀀂􀀌􀀑􀀂􀀃 􀀆􀀌􀀑􀀈􀀋􀀄􀀍 􀀋􀀂􀀑􀀄􀀃u􀀌􀀊􀀑􀀗 v􀀉􀀒􀀄􀀈􀀄􀀅􀀆􀀗 􀀌􀀚􀀃􀀉􀀒􀀄􀀅􀀌􀀊 􀀃􀀄􀀑􀀈􀀂􀀅􀀑􀀄􀀉􀀅􀀗 􀀌􀀅􀀃 􀀈􀀙􀀂 􀀅􀀂􀀂􀀃 􀀖􀀉􀀋 􀀖􀀂􀀂􀀃􀀄􀀅􀀆 􀀄􀀅􀀈􀀂􀀋􀀋u􀀕􀀈􀀄􀀉􀀅 􀀉􀀋 􀀃􀀂􀀊􀀌􀀓. T􀀙􀀂

FAQs

What does the document focus on regarding feeding intolerance in preterm infants?

It focuses on using machine learning to predict feeding intolerance and perform risk stratification before feeding.

How is model performance described?

Performance is reported using discrimination metrics such as AUC with confidence intervals, plus statistical comparisons across model approaches.

Why is pre-feeding risk stratification important in the study?

It enables earlier identification of infants at higher risk, supporting closer monitoring and more timely clinical decisions.

Aug 4, 2026

Upload by Evangeline