Construction and SHAP interpretability analysis of a risk prediction model for feeding intolerance in preterm newborns based on machine learning

Table 2 Spearman correlation analysis of FI in preterm newborns

Risk factors	Correlation coefficient	Risk factors	Correlation coefficient
Gender	-0.071	HIE	-0.021
Gestational age	-0.472^**	Neonatal asphyxia	0.280^**
Birth weight	-0.513^**	Neonatal hyperbilirubinemia	0.032
Apgar score in one minute	-0.295^**	Infection	0.147^**
Mother's age	-0.008	PDA	0.171^**
Polyembryony	0.018	Apnea	0.309^**
Assisted reproductive	-0.024	High temperature	0.309^**
Pregnancy complications	-0.005	Mechanical ventilation	0.417^**
Mode of delivery	-0.089	History of resuscitation	0.556^**
Abnormal fetal position	0.088	Milk opening time	0.398^**
Amniotic fluid abnormality	-0.013	Interval between two stools	0.481^**
Umbilical cord abnormality	0.032	Blood transfusion	0.323^**
Placenta abnormality	0.016	Antibiotic	0.105
Premature rupture of membranes	0.034	PS	0.321^**
NRDS	0.359^**	Probiotics	-0.371^**

FI feeding intolerance, NRDS neonatal respiratory distress syndrome, HIE hypoxic ischemic encephalopathy, PDA patent ductus arteriosus, PS pulmonary surfactant
^*P < 0.05; ^**P < 0.01

ISSN: 1472-6947