Neonatal Jaundice Severity Detection from Skin Images using Deep Transfer Learning Techniques

Banan ALdabbagh; Mazin H. Aziz

doi:10.35882/jeeemi.v7i1.576

Banan ALdabbagh Department of Computer Engineering, University of Mosul, Mosul, IRAQ https://orcid.org/0009-0000-3334-5150
Mazin H. Aziz Department of Computer Engineering, University of Mosul, Mosul, IRAQ https://orcid.org/0000-0003-0124-7841

DOI: https://doi.org/10.35882/jeeemi.v7i1.576

Keywords: Neonatal jaundice, Machine Learning, bilirubin, Total Serum Bilirubin, deep-learning, VGG16, ResNet50.

Abstract

Neonates in the initial weeks postpartum frequently experience jaundice, a prevalent medical condition characterized by the yellow discoloration of the sclera and integumentary surfaces. This phenomenon transpires as a result of the elevation of bilirubin concentrations within the circulatory system. When bilirubin levels reach critical thresholds, they present a considerable risk for severe complications, including neurological impairment, which represents one of the gravest outcomes that may ensue if the condition is not addressed with due diligence. This study investigates a non-invasive method for assessing jaundice severity in full-term infants from 1 to 29 days, focusing on infants in Mosul city. A dataset of 344 images was collected using an iPhone 12 Pro Max (9MP camera) at Ibn Al-Atheer Hospital, capturing various skin tones and lighting conditions to ensure accurate analysis. Advanced computer vision techniques were used to classify jaundice severity into three and four categories based on skin images. Pre-trained deep transfer learning models, namely VGG16 and ResNet50, were utilized for training, with the fully connected layer removed and a suitable classifier designed for each model. VGG16 achieved 91.71% accuracy for the three-category classification, while ResNet50 reached 95.98%. For the four-category classification, accuracies of 94.92% and 94.66% were achieved, respectively. These high accuracy levels suggest that non-invasive, image-based assessments can reduce the need for repeated blood tests. This research highlights the potential of using smartphone-based methods for jaundice screening in neonatal care, providing a reliable, accessible tool to reduce strain on medical facilities and improve early detection.

Downloads

Download data is not yet available.

References

G. G. Asefa et al., “Determinants of Neonatal Jaundice among Neonates Admitted to Neonatal Intensive Care Unit in Public General Hospitals of Central Zone, Tigray, Northern Ethiopia, 2019: A Case-Control Study,” Biomed Res Int, vol. 2020, 2020, https://doi.org/10.1155/2020/4743974

H. Shirzadfar, K. Sheikhi, and Z. Meschian, “The epidemiologic study of neonatal jaundice, relation between jaundice and liver and alternative methods to cure jaundice,” Clinical Practice, vol. 16, no. 3, 2019, doi: 10.37532/fmcp.2019.16(3).1117-1125.

H. Brits et al., “The prevalence of neonatal jaundice and risk factors in healthy term neonates at National District Hospital in Bloemfontein,” Afr J Prim Health Care Fam Med, vol. 10, no. 1, 2018, doi: 10.4102/phcfm.v10i1.1582.

M. M. Tiwari and H. N. Pise, “Comparative study between serum and transcutaneous bilirubin measurements in new-borns,” Int J Contemp Pediatrics, vol. 6, no. 2, p. 817, Feb. 2019, doi: 10.18203/2349-3291.ijcp20190735.

K. Darka and A. Gül, “Correlation of Capillary Tube and Transcutaneous Methods with Serum Biochemistry in Bilirubin Levels in Neonates with Jaundice,” Journal of Contemporary Medicine, vol. 14, no. 2, pp. 67–71, Mar. 2024, doi: 10.16899/jcm.1434289.

H. S. Khalid and N. A. Fakhre, “SMARTPHONES APP AS PORTABLE COLORIMETRIC SENSOR FOR LOWCOST QUANTIFICATION OF SAFRANIN DYE IN AQUEOUS SOLUTIONS,” Bull Chem Soc Ethiop, vol. 37, no. 5, pp. 1065–1080, 2023, doi: 10.4314/bcse.v37i5.1.

K. Hasanah, L. Ode, and A. Rahman, “Efektifitas Skrining Jaundice Berbasis Smartphone Mengurangi Prosedur Invasive Pada Neonates : Literature Review,” JKEP (Jurnal Keperawatan), vol. 8, no. 2, 2023.

M. S. Jarjees, S. S. M. Sheet, and B. T. Ahmed, “Leukocytes identification using augmentation and transfer learning based convolution neural network,” Telkomnika (Telecommunication Computing Electronics and Control), vol. 20, no. 2, pp. 314–320, 2022, doi: 10.12928/TELKOMNIKA.v20i2.23163.

A. Vuckovic, V. J. F. Gallardo, M. Jarjees, M. Fraser, and M. Purcell, “Prediction of central neuropathic pain in spinal cord injury based on EEG classifier,” Clinical Neurophysiology, vol. 129, no. 8, pp. 1605–1617, Aug. 2018, doi: 10.1016/j.clinph.2018.04.750.

M. J. Mohammed, E. A. Mohammed, and M. S. Jarjees, “Recognition of multifont English electronic prescribing based on convolution neural network algorithm,” Bio-Algorithms and Med-Systems, vol. 16, no. 3, 2020, doi: 10.1515/bams-2020-0021.

A. Deshmukh, “Artificial Intelligence in Medical Imaging: Applications of Deep Learning for Disease Detection and Diagnosis,” Universal Research Reports, vol. 11, no. 3, pp. 31–36, Jun. 2024, doi: 10.36676/urr.v11.i3.1284.

A. Ritahani Ismail, Syed Qamrun Nisa, S. A. Shaharuddin, S. I. Masni, and S. A. Suharudin Amin, “Utilising VGG-16 of Convolutional Neural Network for Medical Image Classification,” International Journal on Perceptive and Cognitive Computing, vol. 10, no. 1, pp. 113–118, Jan. 2024, doi: 10.31436/ijpcc.v10i1.460.

F. Prinzi, T. Currieri, S. Gaglio, and S. Vitabile, “Shallow and deep learning classifiers in medical image analysis,” Dec. 01, 2024, Springer Science and Business Media Deutschland GmbH. doi: 10.1186/s41747-024-00428-2.

M. N. Mansor et al., “Jaundice in newborn monitoring using color detection method,” in Procedia Engineering, 2012, pp. 1631–1635. doi: 10.1016/j.proeng.2012.01.185.

B. O. Olusanya, T. M. Slusher, D. O. Imosemi, and A. A. Emokpae, “Maternal detection of neonatal jaundice during birth hospitalization using a novel two-color icterometer,” PLoS One, vol. 12, no. 8, Aug. 2017, doi: 10.1371/journal.pone.0183882.

R. Karim, M. Zaman, and W. H. Yong, “A Non-invasive Methods for Neonatal Jaundice Detection and Monitoring to Assess Bilirubin Level: A Review,” Jan. 01, 2023, International Association for Educators and Researchers (IAER). doi: 10.33166/AETiC.2023.01.002.

A. Gupta, A. Kumar, and P. Khera, “Method and Model for Jaundice Prediction Through Non-Invasive Bilirubin Detection Technique.” [Online]. Available: www.ijert.org

A. Althnian, N. Almanea, and N. Aloboud, “Neonatal jaundice diagnosis using a smartphone camera based on eye, skin, and fused features with transfer learning,” Sensors, vol. 21, no. 21, Nov. 2021, doi: 10.3390/s21217038.

J. A. Taylor et al., “Use of a Smartphone App to Assess Neonatal Jaundice.”

S. Swarna, S. Pasupathy, B. Chinnasami, N. M. D., and B. Ramraj, “The smart phone study: assessing the reliability and accuracy of neonatal jaundice measurement using smart phone application,” Int J Contemp Pediatrics, vol. 5, no. 2, p. 285, Feb. 2018, doi: 10.18203/2349-3291.ijcp20175928.

J. Castro-Ramos, C. Toxqui-Quitl, F. Villa Manriquez, E. Orozco-Guillen, A. Padilla-Vivanco, and JJ. Sánchez-Escobar, “Detecting jaundice by using digital image processing,” in Three-Dimensional and Multidimensional Microscopy: Image Acquisition and Processing XXI, SPIE, Mar. 2014, p. 89491U. doi: 10.1117/12.2041354.

W. Y. Hsu and H. C. Cheng, “A fast and effective system for detection of neonatal jaundice with a dynamic threshold white balance algorithm,” Healthcare (Switzerland), vol. 9, no. 8, Aug. 2021, doi: 10.3390/healthcare9081052.

M. Aydın, F. Hardalaç, B. Ural, and S. Karap, “Neonatal Jaundice Detection System,” J Med Syst, vol. 40, no. 7, Jul. 2016, doi: 10.1007/s10916-016-0523-4.

R. Angelico et al., “A novel mobile phone application for infant stool color recognition: An easy and effective tool to identify acholic stools in newborns,” J Med Screen, vol. 28, no. 3, pp. 230–237, Sep. 2021, doi: 10.1177/0969141320974413.

A. Yaseen Abdulrazzak, S. Latif Mohammed, A. Al-Naji, and J. Chahl, “Computer-Aid System for Automated JaDetectionundice,” Journal of Techniques, vol. 5, no. 1, pp. 8–15, Mar. 2023, doi: 10.51173/jt.v5i1.1128.

M. S. Jarjees, N. Abdulkadir, and K. Kandla, “Neonatal Jaundice Detection System Using Convolutional Neural Network Algorithm,” 2023, doi: 10.14704/nq.2022.20.8.NQ44777.

E. P. Rahayu, M. N. Widyawati, and S. Suryono, “Euclidean distance digital image processing for jaundice detect,” IOP Conf Ser Mater Sci Eng, vol. 1108, no. 1, p. 012022, Mar. 2021, doi: 10.1088/1757-899x/1108/1/012022.

S. Dissaneevate et al., “A Mobile Computer-Aided Diagnosis of Neonatal Hyperbilirubinemia using Digital Image Processing and Machine Learning Techniques,” International Journal of Innovative Research and Scientific Studies, vol. 5, no. 1, pp. 10–17, 2022, doi: 10.53894/ijirss.v5i1.334.

L. De Greef et al., “BiliCam: Using mobile phones to monitor newborn jaundice,” in UbiComp 2014 - Proceedings of the 2014 ACM International Joint Conference on Pervasive and Ubiquitous Computing, Association for Computing Machinery, Inc, 2014, pp. 331–342. doi: 10.1145/2632048.2632076.

B. J. Lauer and N. D. Spector, “Hyperbilirubinemia in the newborn,” Aug. 2011. doi: 10.1542/pir.32-8-341.