A Predictive Model for Perinatal Brain Injury Using Machine Learning Based on Early Birth Data

Overview

Journal Children (Basel)

Specialty Health Services

Date 2024 Nov 27

PMID 39594888

Authors

Ga Won Jeon

Yeong Seok Lee

Won-Ho Hahn

Yong Hoon Jun

Affiliations

Soon will be listed here.

Abstract

Background/objective: It is difficult to predict perinatal brain injury, and performing brain magnetic resonance imaging (MRI) based on suspected injury remains a clinical challenge. Therefore, we aimed to develop a reliable method for predicting perinatal brain injury using a machine learning model with early birth data.

Methods: Neonates admitted to our institution from January 2017 to June 2024 with a gestational age of ≥36 weeks, a birth weight of ≥1800 g, admission within 6 h of birth, and who underwent brain MRI to confirm perinatal brain injury were included. Various machine learning models, including gradient boosting, were trained using early birth data to predict perinatal brain injury. Synthetic minority over-sampling and adaptive synthetic sampling (ADASYN) were applied to address class imbalance. Model performance was evaluated using accuracy, F1 score, and ROC curves. Feature importance scores and Shapley additive explanations (SHAP) values were also calculated.

Results: Among 179 neonates, 39 had perinatal brain injury. There were significant differences between the injury and non-injury groups in mode of delivery, Apgar scores, capillary pH, lactate dehydrogenase (LDH) levels, and whether therapeutic hypothermia was performed. The gradient boosting model with the ADASYN method achieved the best performance. In terms of feature importance scores, the 1 min Apgar score was the most influential predictor. Additionally, SHAP analysis showed that LDH levels had the highest SHAP values.

Conclusion: the gradient boosting model with ADASYN oversampling effectively predicts perinatal brain injury, potentially improving early detection for predicting long-term outcomes, reducing unnecessary MRI scans, and lowering healthcare costs.

References

Kim S, Oh Y, Son J, Jung D, Kim D, Ryu S . Machine learning-based analysis for prediction of surgical necrotizing enterocolitis in very low birth weight infants using perinatal factors: a nationwide cohort study. Eur J Pediatr. 2024; 183(6):2743-2751. PMC: 11098869. DOI: 10.1007/s00431-024-05505-7. View

Dunbar M, Mineyko A, Hill M, Hodge J, Floer A, Kirton A . Population Based Birth Prevalence of Disease-Specific Perinatal Stroke. Pediatrics. 2020; 146(5). DOI: 10.1542/peds.2020-013201. View

Vasudevan C, Levene M . Epidemiology and aetiology of neonatal seizures. Semin Fetal Neonatal Med. 2013; 18(4):185-91. DOI: 10.1016/j.siny.2013.05.008. View

Roy B, Webb A, Walker K, Morgan C, Badawi N, Nunez C . Prevalence & Risk Factors for Perinatal Stroke: A Population-Based Study. Child Neurol Open. 2023; 10:2329048X231217691. PMC: 10729630. DOI: 10.1177/2329048X231217691. View

Bogavac I, Jelicic L, Marisavljevic M, Boskovic Matic T, Subotic M . Arterial Presumed Perinatal Ischemic Stroke: A Mini Review and Case Report of Cognitive and Speech-Language Profiles in a 5-Year-Old Girl. Children (Basel). 2024; 11(1). PMC: 10814911. DOI: 10.3390/children11010033. View

Robi Y, Sitote T . Neonatal Disease Prediction Using Machine Learning Techniques. J Healthc Eng. 2023; 2023:3567194. PMC: 9981287. DOI: 10.1155/2023/3567194. View

Molloy E, Branagan A, Hurley T, Quirke F, Devane D, Taneri P . Neonatal encephalopathy and hypoxic-ischemic encephalopathy: moving from controversy to consensus definitions and subclassification. Pediatr Res. 2023; 94(6):1860-1863. DOI: 10.1038/s41390-023-02775-z. View

Kim H, Kim J, Park S . Predicting severe intraventricular hemorrhage or early death using machine learning algorithms in VLBWI of the Korean Neonatal Network Database. Sci Rep. 2024; 14(1):11113. PMC: 11096174. DOI: 10.1038/s41598-024-62033-y. View

Reinboth B, Koster C, Abberger H, Prager S, Bendix I, Felderhoff-Muser U . Endogenous hypothermic response to hypoxia reduces brain injury: Implications for modeling hypoxic-ischemic encephalopathy and therapeutic hypothermia in neonatal mice. Exp Neurol. 2016; 283(Pt A):264-75. DOI: 10.1016/j.expneurol.2016.06.024. View

10.

Barrois M, Patkai J, Delorme P, Chollat C, Goffinet F, LE Ray C . Factors associated with neonatal hypoxic ischemic encephalopathy in infants with an umbilical artery pH less than 7.00. Eur J Obstet Gynecol Reprod Biol. 2019; 236:69-74. DOI: 10.1016/j.ejogrb.2019.02.009. View

11.

Ferriero D, Fullerton H, Bernard T, Billinghurst L, Daniels S, DeBaun M . Management of Stroke in Neonates and Children: A Scientific Statement From the American Heart Association/American Stroke Association. Stroke. 2019; 50(3):e51-e96. DOI: 10.1161/STR.0000000000000183. View

12.

Srivastava R, Kirton A . Perinatal Stroke: A Practical Approach to Diagnosis and Management. Neoreviews. 2021; 22(3):e163-e176. DOI: 10.1542/neo.22-3-e163. View

13.

Cainelli E, Vedovelli L, Mastretta E, Gregori D, Suppiej A, Bisiacchi P . Long-Term Outcomes after Neonatal Hypoxic-Ischemic Encephalopathy in the Era of Therapeutic Hypothermia: A Longitudinal, Prospective, Multicenter Case-Control Study in Children without Overt Brain Damage. Children (Basel). 2021; 8(11). PMC: 8625352. DOI: 10.3390/children8111076. View

14.

Jeon G, Sheldon R, Ferriero D . Hypoxia-inducible factor: role in cell survival in superoxide dismutase overexpressing mice after neonatal hypoxia-ischemia. Korean J Pediatr. 2019; 62(12):444-449. PMC: 6933302. DOI: 10.3345/kjp.2019.00850. View

15.

Munoz D, Hidalgo M, Balut F, Troncoso M, Lara S, Barrios A . Risk Factors for Perinatal Arterial Ischemic Stroke: A Case-Control Study. Cell Med. 2020; 10:2155179018785341. PMC: 6172995. DOI: 10.1177/2155179018785341. View

16.

Sheikhtaheri A, Zarkesh M, Moradi R, Kermani F . Prediction of neonatal deaths in NICUs: development and validation of machine learning models. BMC Med Inform Decis Mak. 2021; 21(1):131. PMC: 8056638. DOI: 10.1186/s12911-021-01497-8. View

17.

Sheldon R, Windsor C, Lu F, Stewart N, Jiang X, Ferriero D . Hypothermia Treatment after Hypoxia-Ischemia in Glutathione Peroxidase-1 Overexpressing Mice. Dev Neurosci. 2023; 46(2):98-111. PMC: 10667569. DOI: 10.1159/000531204. View

18.

Ahn S, Chang Y, Sung D, Sung S, Park W . Hypothermia broadens the therapeutic time window of mesenchymal stem cell transplantation for severe neonatal hypoxic ischemic encephalopathy. Sci Rep. 2018; 8(1):7665. PMC: 5955959. DOI: 10.1038/s41598-018-25902-x. View

19.

Srivastava R, Mailo J, Dunbar M . Perinatal Stroke in Fetuses, Preterm and Term Infants. Semin Pediatr Neurol. 2022; 43:100988. DOI: 10.1016/j.spen.2022.100988. View

20.

Leavy A, Mateos E . Perinatal Brain Injury and Inflammation: Lessons from Experimental Murine Models. Cells. 2020; 9(12). PMC: 7764185. DOI: 10.3390/cells9122640. View