Madhusundar Nelson; Surendran Rajendran

An Intelligent Deep Learning Model for Neonatal Asphyxia Identification Using Linear Graph Neural Networks Optimized with African Vultures Algorithm

PDF (1244KB), PP.1-17

Views: 0 Downloads: 0

Author(s)

Madhusundar Nelson ¹ Surendran Rajendran ^1,*

1. Department of Computer Science and Engineering, Saveetha School of Engineering, Saveetha Institute of Medical and Technical Sciences, Chennai, India

* Corresponding author.

DOI: https://doi.org/10.5815/ijem.2025.03.01

Received: 4 Nov. 2024 / Revised: 3 Jan. 2025 / Accepted: 27 Jan. 2025 / Published: 8 Jun. 2025

Index Terms

Neonatal, Asphyxia, Linear Graph Neural Network (LGNN), Deep Learning, Computer Vision, African Vultures Algorithm

Abstract

Birth marks the evolution from a fluid environment to one in which the baby breathes. Respiration difficulties after birth and in the first hours of life are common. Some babies will experience more severe respiration problems, and some of these will require medical care that only expert physicians can offer. In the first six hours of life, asphyxia is recognized. Neonatal asphyxia is a syndrome categorized by the deprivation of oxygen supply at birth and causes many morbidities and mortalities among newborns. Improved results depend on early detection and action. Using the African Vultures Algorithm that performs hyperparameter optimization expands accuracy and effectiveness in supporting the model. This integration improves the detection and sympathetic of neonatal asphyxia by using African Vultures Algorithm's capability to find fine results in high-dimensional places as well as the capability of LGNN to method structured clinical data. Herein, our model excels over present deep learning models and computer vision techniques for accurate detection of neonatal asphyxia. LGNN model achieved an accuracy of 98% in Dataset 1 and 96.5% in Dataset 2. F1-Score of the proposed model was achieved as 97.6% in Dataset 1 and 96.7% in Dataset 2. In Dataset 1 and Dataset 2, LGNN model as long as 97.9% and 97.1% precision, individually. In Dataset 1 and Dataset 2, the proposed (LGNN) provided 98.8% and 97.7% recall, one-to-one. Efficiently exploit the graph structure for feature aggregation and inference by employing a linear graph neural network architecture. Improved newborn outcomes are made possible by the suggested model, which is a potent diagnostic tool that also helps clinicians comprehend underlying pathophysiological causes.

Cite This Paper

Madhusundar Nelson , Surendran Rajendran, "An Intelligent Deep Learning Model for Neonatal Asphyxia Identification Using Linear Graph Neural Networks Optimized with African Vultures Algorithm", International Journal of Engineering and Manufacturing (IJEM), Vol.15, No.3, pp. 1-17, 2025. DOI:10.5815/ijem.2025.03.01

Reference

[1]Boskabadi H, Zakerihamidi M, Ghayour Mobarhan M, Bagheri F, Moradi A, Beiraghi Toosi M. Comparison of new biomarkers in the diagnosis of perinatal asphyxia. Iran J Child Neurol. 2023 Winter;17(1):99-110. doi: 10.22037/ijcn.v17i2.38561. Epub 2023 Jan 1. PMID: 36721830; PMCID: PMC9881829.1
[2]Keles E, Bagci U. The past, current, and future of neonatal intensive care units with artificial intelligence: a systematic review. NPJ Digit Med. 2023 Nov 27;6(1):220. doi: 10.1038/s41746-023-00941-5. PMID: 38012349; PMCID: PMC10682088.
[3]Chioma, R.; Sbordone, A.; Patti, M.L.; Perri, A.; Vento, G.; Nobile, S. Applications of Artificial Intelligence in Neonatology. Appl. Sci. 2023, 13, 3211. https://doi.org/10.3390/app13053211.
[4]Beck J, Loron G, Ancel PY, Alison M, Hertz Pannier L, Vo Van P, Debillon T, Bednarek N. An Updated Overview of MRI Injuries in Neonatal Encephalopathy: LyTONEPAL Cohort. Children (Basel). 2022 Apr 14;9(4):561. doi: 10.3390/children9040561. PMID: 35455605; PMCID: PMC9032533.
[5]Parmentier CEJ, de Vries LS, Groenendaal F. Magnetic Resonance Imaging in (Near-)Term Infants with Hypoxic-Ischemic Encephalopathy. Diagnostics (Basel). 2022 Mar 6;12(3):645. doi: 10.3390/diagnostics12030645. PMID: 35328199; PMCID: PMC8947468.
[6]Kieu, S.T.H.; Bade, A.; Hijazi, M.H.A.; Kolivand, H. A Survey of Deep Learning for Lung Disease Detection on Medical Images: State-of-the-Art, Taxonomy, Issues and Future Directions. J. Imaging 2020, 6, 131. https://doi.org/10.3390/jimaging6120131.
[7]Cao Shujuan, Wang Xiaoming, Ye Zhonglin, Li Mingyuan, Zhao Haixing, LGNN: a novel linear graph neural network algorithm, Frontiers in Computational Neuroscience.
[8]Subrato Bharati, Prajoy Podder, M. Rubaiyat Hossain Mondal, Hybrid deep learning for detecting lung diseases from X-ray images, Informatics in Medicine Unlocked, Volume 20,2020,100391,ISSN 2352-9148,https://doi.org/10.1016/j.imu.2020.100391.
[9]Sarnat, H.B.; Sarnat, M.S. Neonatal Encephalopathy Following Fetal Distress. A Clinical and Electroencephalographic Study. Arch. Neurol. 1976, 33, 696–705.
[10]Barkovich, A.; Miller, S.; Bartha, A.; Newton, N.; Hamrick, S.; Mukherjee, P.; Glenn, O.; Xu, D.; Partridge, J.; Ferriero, D.; et al. MR imaging, MR spectroscopy, and diffusion tensor imaging of sequential studies in neonates with encephalopathy. AJNR Am. J. Neuroradiol. 2006, 27, 533–547.
[11]Barkovich, A.J.; Baranski, K.; Vigneron, D.; Partridge, J.C.; Hallam, D.K.; Hajnal, B.L.; Ferriero, D.M. Proton MR spectroscopy for the evaluation of brain injury in asphyxiated, term neonates. AJNR Am J Neuroradiol. 1999, 20, 1399–1405.
[12]Penrice, J.; Lorek, A.; Cady, E.B.; Amess, P.N.; Wylezinska, M.; Cooper, C.E.; D’Souza, P.; Brown, G.C.; Kirkbride, V.; Edwards, A.D.; et al. Proton magnetic resonance spectroscopy of the brain during acute hypoxia-ischemia and delayed cerebral energy failure in the newborn piglet. Pediatr. Res. 1997, 41, 795–802.
[13]Zhang, Y.; Xu, Y.; Zhang, Y. A Graph Neural Network Node Classification Application Model with Enhanced Node Association. Appl. Sci. 2023, 13, 7150. https://doi.org/10.3390/app13127150.
[14]Li, Y., Wang, Y., Yang, X. et al. Speech emotion recognition based on Graph-LSTM neural network. J AUDIO SPEECH MUSIC PROC. 2023, 40 (2023). https://doi.org/10.1186/s13636-023-00303-9.
[15]M. Nelson, S. Rajendran, Y. Alotaibi, Vision graph neural network-based neonatal identification to avoid swapping and abduction, AIMS Mathematics, 8 (2023), 21554–21571. https://doi.org/10.3934/math.20231098
[16]T. Tamilvizhi, R. Surendran, K. Anbazhagan, K. Rajkumar, Quantum behaved particle swarm optimization-based deep transfer learning model for sugarcane leaf disease detection and classification, Math. Probl. Eng., 2022 (2022), 3452413. https://doi.org/10.1155/2022/3452413
[17]https://www.kaggle.com/datasets/bhavkaur/synthetic-infant-health-dataset.
[18]https://www.mrineonatalbrain.com/

International Journal of Engineering and Manufacturing (IJEM)