Non-invasive neonatal jaundice detection using machine learning with support vector machine synthetic minority over-sampling technique ensemble voting

Yasir Hussein Shakir¹, Hamzah Mohammed Kadhim², Ammar A. Hasan³, Kawther Kanaan Salih⁴, Wadha Humaid Said Al Mamari⁵, Eshaq Aziz Awadh AL Mandhari⁶, Fahad Taha AL-Dhief^7*, Nurhizam Safie^8*, Noorayisahbe Mohd Yaacob⁹

Show Less

¹ College of Graduate Studies (COGS), Universiti Tenaga Nasional (UNITEN), Kajang, Selangor , Malaysia

² Department of Business Administration, Technical College of Management/Baghdad, Middle Technical University, Baghdad , Iraq

³ Department of Computer Engineering, College of Engineering, University of Baghdad, Baghdad , Iraq

⁴ Department of Pharmaceutics, College of Pharmacy, Al-Zahraa University for Women, Karbala , Iraq

⁵ Department of Nursing, Faculty of Nursing Program, Oman College of Health Sciences, Dhofar , Oman

⁶ Graduate School of Technology, Asia Pacific University of Technology and Innovation (APU), Kuala Lumpur , Malaysia

⁷ Center for Software Technology and Management, Faculty of Information Science and Technology, Universiti Kebangsaan Malaysia (UKM), Bangi, Selangor , Malaysia

⁸ Center for Software Technology and Management, Faculty of Information Science and Technology, Universiti Kebangsaan Malaysia (UKM), Bangi, Selangor , Malaysia

⁹ Center for Software Technology and Management, Faculty of Information Science and Technology, Universiti Kebangsaan Malaysia (UKM), Bangi, Selangor , Malaysia

IJOSI, 026240141 https://doi.org/10.6977/IJoSI.2026011_10(X).026240141

Received: 9 June 2026 | Revised: 28 July 2026 | Accepted: 3 September 2026 | Published online: 13 November 2026

© 2026 by the Author(s). This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution -Noncommercial 4.0 International License (CC-by the license) ( https://creativecommons.org/licenses/by-nc/4.0/ )

Download PDF

XML

Cite

Abstract

Neonatal jaundice is a common condition in newborns that can result in serious consequences such as permanent brain injury. Traditional diagnostic methods often require invasive blood tests that may cause nuisance and health risks for infants. To address this problem, we propose a non-invasive diagnostic method utilizing image processing and machine learning. The study lies in the integration of the support vector machine synthetic minority over-sampling technique (SVMSMOTE) to address the critical but underexplored challenge of class imbalance in neonatal jaundice datasets, combined with a soft-voting ensemble of adaptive boosting (AdaBoost) and extra trees (ET). This study used the Normal and Jaundiced Newborns dataset, containing facial images of newborns aged 2 to 8 days with varied skin tones. The color features were extracted from RGB and YCrCb channels. The primary challenge addressed is data imbalance, which can bias models toward the majority class and reduce sensitivity for jaundiced cases. SVMSMOTE addresses this by generating high-quality synthetic samples near the decision boundary, improving minority-class representation without introducing noise. We evaluated eight machine learning classifiers: support vector machine, multilayer perceptron, logistic regression, AdaBoost, Extra Trees, passive aggressive, and gradient boosting, as well as ensemble methods combining soft voting and hard voting. Our best-performing model achieved 99.35% accuracy, 98.81% precision, 99.55% recall, an F1-score of 99.17%, an AUC of 99.55%, and a Matthews correlation coefficient of 98.36%. This high recall is particularly clinically significant, reducing false negatives, a critical requirement for neonatal screening to prevent missed diagnoses. These findings demonstrate the promising potential of artificial intelligence-based non-invasive approaches for early neonatal jaundice, but further external clinical validation is required before real-world deployment.

Keywords

Neonatal jaundice

Machine learning

Support vector machine synthetic minority over-sampling technique

Normal and Jaundiced Newborns dataset

Voting model

Funding

This research was funded by the Faculty of Information Science and Technology, Universiti Kebangsaan Malaysia.

Conflict of interest

The authors declare that they have no competing interests.

References

Abbas, M. W., Shamshad, T., Ashraf, M. A., & Javaid, R. (2016). Jaundice: a basic review. International Journal of Research in Medical Sciences, 4(5), 1313–1319. https://doi.org/10.18203/2320-6012.IJRMS20161196

Abd Karim, S. A., Mohamad, U. H., & Nohuddin, P. N. E. (2025). Interaction-based feature selection technique using fuzzy discretization and class association rule mining for breast cancer classification. Malaysian Journal of Fundamental and Applied Sciences, 21(2), 1832–1859. https://doi.org/10.11113/mjfas.v21n2.3787

Abdulrazzak, A. Y., Mohammed, S. L., & Al-Naji, A. (2023). NJN: A Dataset for the Normal and Jaundiced Newborns. BioMedInformatics, 3(3), 543–552. https://doi.org/10.3390/BIOMEDINFORMATICS3030037

Abdulrazzak, A. Y., Mohammed, S. L., Al-Naji, A., & Chahl, J. (2023). Computer-Aid System for Automated Jaundice Detection. Journal of Techniques, 5(1), 8–15. https://doi.org/10.51173/JT.V5I1.1128

Abdulrazzak, A. Y., Mohammed, S. L., Al-Naji, A., & Chahl, J. (2024a). Neonatal Jaundice detection using machine-learning algorithms: A comparative study. AIP Conference Proceedings, 3232(1), 040001. https://doi.org/10.1063/5.0236954

Abdulrazzak, A. Y., Mohammed, S. L., Al-Naji, A., & Chahl, J. (2024b). Real-Time Jaundice Detection in Neonates Based on Machine Learning Models. BioMedInformatics, 4(1), 623–637. https://doi.org/10.3390/BIOMEDINFORMATICS4010034

Abu-Ain, W. A. K., Abdullah, S. N. H. S., Omar, K., & Abd. Rahman, S. Z. (2018). Automatic multi-lingual script recognition application. GEMA Online® Journal of Language Studies, 18(3), 203–221. https://doi.org/10.17576/gema-2018-1803-12

Albadr,Al-Dhief, F. T., Latiff, N. M. A., Malik, N. N. N. A., Baki, M. M., Sabri, N., Albadr, M. A. A., & Md Sazihan, N. F. S. (2023). Voice pathology detection using decision tree classifier. In 2023 14th International Conference on Information and Communication Technology Convergence (ICTC) (pp. 36–41). IEEE. https://doi.org/10.1109/ICTC58733.2023.10392786

Alipoor, G., & Samadi, E. (2018). Robust Gender Identification using EMD-Based Cepstral Features. Asia-Pacific Journal of Information Technology & Multimedia, 07(01), 71–81. https://doi.org/10.17576/apjitm-2018-0701-06

Ali, Y. A. A., Kamani, T., Patel, S. K., Saidani, T., Almawgani, A. H. M., & Alabsi, B. A. (2025). Design and Analysis of Surface Plasmon Resonance-Based Refractive Index Biosensor for Detection of Colorectal Cancer with Machine Learning Approach. Plasmonics, 20(9), 7597–7609. https://doi.org/10.1007/s11468-024-02720-8

Althnian, A., Almanea, N., & Aloboud, N. (2021). Neonatal Jaundice Diagnosis Using a Smartphone Camera Based on Eye, Skin, and Fused Features with Transfer Learning. Sensors. 21(21), 7038. https://doi.org/10.3390/S21217038

Alzawali, M. I. H., Yusoff, Y., Alwee, R., Yunos, Z. M., Talib, M. S., Hassan, H., AL-Dhief, F. T., Albadr, M. A. A., Alsemawi, M. R. M., & Ahmad, S. Z. R. S. (2024). Facial emotion images recognition based on binarized genetic algorithm-random forest. Baghdad Science Journal, 21(2), 31. https://doi.org/10.21123/bsj.2024.9698

Anggraini, D., Widyawati, M. N., & Suryono, S. (2021). Information system for diagnosing Neonatal Jaundice using rule-based algorithm. Journal of Physics: Conference Series, 1943(1), 012037. https://doi.org/10.1088/1742-6596/1943/1/012037

Armishty, F., Mohammed, L. Y., Ibrahim, R., Haji, I., Omar, A., Jameel, D., Khattab, O., Rajab, B. R., Saleem, A., Ahmed, M. M., Hassan, S., Nezar, D. P., & Ahmad, H. A. (2023). The Prevalence of Anemia and Associated Factors among Children under 5 years in Zakho City, Kurdistan Region, Iraq. Passer Journal of Basic and Applied Sciences, 5(2), 391–397. https://doi.org/10.24271/PSR.2023.390463.1287

Bharani Nayagi, S., & Shiny Angel, T. S. (2023). Diagnosis of neonatal hyperbilirubinemia using CNN model along with color card techniques. Biomedical Signal Processing and Control, 85, 104746. https://doi.org/10.1016/J.BSPC.2023.104746

Chang, W. L., Mohd Noh, L., Abdul Latiff, A. H., Woo, K. C. K., Ismail, I. H., Abd Hamid, I. J., Siniah, S., Zainal Abidin, M. A., Sham, M., Mat Ripen, A., Baharin, M. F., Abdul Wahab, A., Zainudeen, Z. T., Hashim, I. F., Wong, Y. M., Ahmad Shawaludin, M. Q., & Ali, A. (2024). Newborn screening for severe combined immunodeficiency in Malaysia: current status, challenges and progress. Frontiers in Immunology, 15. https://doi.org/10.3389/fimmu.2024.1456769

Choi, Y., Park, S., & Lee, H. (2025). Neonatal Jaundice Requiring Phototherapy Risk Factors in a Newborn Nursery: Machine Learning Approach. Children, 12(8), 1020. https://doi.org/10.3390/CHILDREN12081020

Dissaneevate, S., Wongsirichot, T., Siriwat, P., Jintanapanya, N., Boonyakarn, U., Janjindamai, W., Thatrimontrichai, A., & Maneenil, G. (2022). A Mobile Computer-Aided Diagnosis of Neonatal Hyperbilirubinemia using Digital Image Processing and Machine Learning Techniques. International Journal of Innovative Research and Scientific Studies, 5(1), 10–17. https://doi.org/10.53894/IJIRSS.V5I1.334

Gupta, K., Sharma, V., & Singh Kathait, S. (2024). Smart Screening: Non-Invasive Detection of Severe Neonatal Jaundice using Computer Vision and Deep Learning. International Journal of Computer Applications, 186(35), 35–43.

Hardalaç, F., Aydin, M., Kutbay, U., Ayturan, K., Akyel, A., Çağlar, A., HAi, B., & Mert, F. (2021). Classification of neonatal jaundice in mobile application with noninvasive imageprocessing methods. Turkish Journal of Electrical Engineering and Computer Sciences, 29(4), 2116–2126. https://doi.org/10.3906/elk-2008-76

Herari, N., Assegaff, S. B., Utomo, A. G., & Friscillia, A. P. (2024). Using SEM Model for Fighting Diabetes: Strategic Communication Plan in Jakarta City, Indonesia. Jurnal Komunikasi: Malaysian Journal of Communication, 40(4), 539–560. https://doi.org/10.17576/JKMJC-2024-4004-30

Hulzebos, C. V., Vitek, L., Coda Zabetta, C. D., Dvořák, A., Schenk, P., van der Hagen, E. A. E., Cobbaert, C., & Tiribelli, C. (2021). Screening methods for neonatal hyperbilirubinemia: benefits, limitations, requirements, and novel developments. Pediatric Research, 90(2), 272–276. https://doi.org/10.1038/s41390-021-01543-1

Jafari, T., Nasiri, S., Sayadi, M., Emami, H., & Mohammadpour, S. (2023). A Neonatal jaundice prediction system based on the support vector machine algorithm. Journal of Health Administration, 25(4), 28–44. https://doi.org/10.22034/25.4.28

May, Z., Alam, M. K., & Nayan, N. A. (2022). Recent advances in nondestructive method and assessment of corrosion undercoating in carbon–steel pipelines. Sensors, 22(17), 6654. https://doi.org/10.3390/s22176654

Madhusundar, N., & Surendran, R. (2023). Neonatal Jaundice Identification Over the Face and Sclera using Graph Neural Networks. In: Proceedings of the 5th International Conference on Smart Systems and Inventive Technology, ICSSIT 2023, January 23–25, 2023, Tamilnadu, India. 1243–1249. https://doi.org/10.1109/ICSSIT55814.2023.10060877

Maulood, P. (2024). DRIS formulas and PCA approach for evaluation of the nutritional status of walnut trees in the Hawraman area. Passer Journal of Basic and Applied Sciences, 6(1), 74–79. https://doi.org/10.24271/PSR.2023.415676.1388

Menon, L. A., Joseph, M. A., A, N. N., S, A. R., & Surya, D. (2025). Non-Invasive Neonatal Jaundice Detection Using Image Processing and Machine Learning. AIJR Proceedings, 7(5), 53–61. https://doi.org/10.21467/PROCEEDINGS.7.5.8

Milad, A., Hussein, S. H., Khekan, A. R., Rashid, M., Al-Msari, H., & Tran, T. H. (2022). Development of ensemble machine learning approaches for designing fiber-reinforced polymer composite strain prediction model. Engineering with Computers, 38(4), 3625–3637. https://doi.org/10.1007/S00366-021-01398-4

Mu’azzah Abdul Latiff, N., Al-Dhief, F. T., Sazihan, N. F. S. M., Baki, M. M., Malik, N. N. N. Abd., Albadr, M. A. A., & Abbas, A. H. (2025). Voice pathology detection using machine learning algorithms based on different voice databases. Results in Engineering, 25, 103937. https://doi.org/10.1016/j.rineng.2025.103937

Nayan, N. A., Yi, C. J., Suboh, M. Z., Mazlan, N. F., Periyasamy, P., Abdul Rahim, M. Y. Z., & Shah, S. A. (2022). Covid-19 prediction with machine learning technique from extracted features of photoplethysmogram morphology. Frontiers in Public Health, 10, 920849. https://doi.org/10.3389/fpubh.2022.920849

Outlaw, F., Nixon, M., Odeyemi, O., MacDonald, L. W., Meek, J., & Leung, T. S. (2020). Smartphone screening for neonatal jaundice via ambient-subtracted sclera chromaticity. PLoS ONE, 15(3), e0216970. https://doi.org/10.1371/JOURNAL.PONE.0216970

Padidar, P., Shaker, M., Amoozgar, H., Khorraminejad-Shirazi, M., Hemmati, F., Najib, K. S., Pourarian, S., Padidar, P., Shaker, M., Amoozgar, H., Khorraminejad-Shirazi, M., Hemmati, F., Najib, K. S., & Pourarian, S. (2019). Detection of Neonatal Jaundice by Using an Android OS-Based Smartphone Application. Iranian Journal of Pediatrics, 29(2), e84397. https://doi.org/10.5812/IJP.84397

Rabbani, S. A., El-Tanani, M., Sharma, S., Rabbani, S. S., El-Tanani, Y., Kumar, R., & Saini, M. (2025). Generative Artificial Intelligence in Healthcare: Applications, Implementation Challenges, and Future Directions. BioMedInformatics, 5(3), 37. https://doi.org/10.3390/BIOMEDINFORMATICS5030037

Shamsuddin, S. N., Ismail, N., & Nur-Firyal, R. (2023). Life insurance prediction and its sustainability using machine learning approach. Sustainability, 15(13), 10737. https://doi.org/10.3390/su151310737

Singla, R., & Singh, S. (2016). A framework for detection of jaundice in new born babies using homomorphic filtering based image processing. In: Proceedings of the International Conference on Inventive Computation Technologies, August 26–27, 2016, Coimbatore, India. 1-5. https://doi.org/10.1109/INVENTIVE.2016.7830209

Srinivasula Reddy, D., Gopi Krishna P, B., Pallikonda Rajasekaran, M., Vardhan Reddy M, V., Kowshik, B., & Muneeswaran, V. (2024). Precise and Prompt Identification of Jaundice in Infants using AI. In: Proceedings of the 2024 International Conference on Cognitive Robotics and Intelligent Systems, April 17–19, 2024, Coimbatore, India. 108–114. https://doi.org/10.1109/ICC-ROBINS60238.2024.10534016

Ullah, S., Rahman, K., & Hedayati, M. (2016). Hyperbilirubinemia in Neonates: Types, Causes, Clinical Examinations, Preventive Measures and Treatments: A Narrative Review Article. Iranian Journal of Public Health, 45(5), 558–568.

Xie, D., Wei, J., Xiong, L., Zhu, H., Peng, X., Li, B., & Zou, K. (2025). Association of birth defects during the perinatal period with child mortality under 5 years. Frontiers in Pediatrics, 13, 1485176. https://doi.org/10.3389/FPED.2025.1485176

Za'im, N. A. N., Al-Dhief, F. T., Azman, M., Alsemawi, M. R. M., Latiff, N. M. A., & Baki, M. M. (2023). The accuracy of an online sequential extreme learning machine in detecting voice pathology using the Malaysian voice pathology database. Journal of Otolaryngology-Head & Neck Surgery, 52(1). https://doi.org/10.1186/s40463-023-00661-6

Previous article in this issue

Next article in this issue

International Journal of Systematic Innovation, Electronic ISSN: 2077-8767 Print ISSN: 2077-7973, Published by AccScience Publishing