Detecting neonatal acute bilirubin encephalopathy based on T1-weighted MRI images and learning-based approaches

Background Neonatal hyperbilirubinemia is a common clinical condition that requires medical attention in newborns, which may develop into acute bilirubin encephalopathy with a significant risk of long-term neurological deficits. The current clinical challenge lies in the separation of acute bilirubin encephalopathy and non-acute bilirubin encephalopathy neonates both with hyperbilirubinemia condition since both of them demonstrated similar T1 hyperintensity and lead to difficulties in clinical diagnosis based on the conventional radiological reading. This study aims to investigate the utility of T1-weighted MRI images for differentiating acute bilirubin encephalopathy and non-acute bilirubin encephalopathy neonates with hyperbilirubinemia. Methods 3 diagnostic approaches, including a visual inspection, a semi-quantitative method based on normalized the T1-weighted intensities of the globus pallidus and subthalamic nuclei, and a deep learning method with ResNet18 framework were applied to classify 47 acute bilirubin encephalopathy neonates and 32 non-acute bilirubin encephalopathy neonates with hyperbilirubinemia based on T1-weighted images. Chi-squared test and t-test were used to test the significant difference of clinical features between the 2 groups. Results The visual inspection got a poor diagnostic accuracy of 53.58 ± 5.71% indicating the difficulty of the challenge in real clinical practice. However, the semi-quantitative approach and ResNet18 achieved a classification accuracy of 62.11 ± 8.03% and 72.15%, respectively, which outperformed visual inspection significantly. Conclusion Our study indicates that it is not sufficient to only use T1-weighted MRI images to detect neonates with acute bilirubin encephalopathy. Other more MRI multimodal images combined with T1-weighted MRI images are expected to use to improve the accuracy in future work. However, this study demonstrates that the semi-quantitative measurement based on T1-weighted MRI images is a simple and compromised way to discriminate acute bilirubin encephalopathy and non-acute bilirubin encephalopathy neonates with hyperbilirubinemia, which may be helpful in improving the current manual diagnosis.

[1]  C. Osuorah,et al.  Clinical evaluation of severe neonatal Hyperbilirubinaemia in a resource-limited setting: a 4-year longitudinal study in south-East Nigeria , 2018, BMC Pediatrics.

[2]  Elizabeth Tipton,et al.  Implications of Small Samples for Generalization: Adjustments and Rules of Thumb , 2017, Evaluation review.

[3]  W. Youden,et al.  Index for rating diagnostic tests , 1950, Cancer.

[4]  Serena Counsell,et al.  Serial brain MRI and ultrasound findings: relation to gestational age, bilirubin level, neonatal neurologic status and neurodevelopmental outcome in infants at risk of kernicterus. , 2008, Early human development.

[5]  Mingwu Lou,et al.  Usefulness of 1H‐MRS in differentiating bilirubin encephalopathy from severe hyperbilirubinemia in neonates , 2013, Journal of magnetic resonance imaging : JMRI.

[6]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[7]  Qinghua Zhao,et al.  Brain tumor classification for MR images using transfer learning and fine-tuning , 2019, Comput. Medical Imaging Graph..

[8]  Ann R Stark,et al.  Early recognition of neonatal hyperbilirubinemia and its emergent management. , 2006, Seminars in fetal & neonatal medicine.

[9]  Abdulhakim Coskun,et al.  Hyperintense globus pallidus on T1-weighted MR imaging in acute kernicterus: is it common or rare? , 2005, European Radiology.

[10]  N. Altuntas.  Is There Any Effect of Hyperbilirubinemia on Breastfeeding? If Any, at Which Level? , 2020, Breastfeeding medicine : the official journal of the Academy of Breastfeeding Medicine.

[11]  Vanathi Gopalakrishnan,et al.  A computational framework for the detection of subcortical brain dysmaturation in neonatal MRI using 3D Convolutional Neural Networks , 2018, NeuroImage.

[12]  Lihua Chen,et al.  A clinical prediction rule for acute bilirubin encephalopathy in neonates with extreme hyperbilirubinemia , 2020, Medicine.

[13]  Koji Fujimoto,et al.  Automatic inference model construction for computer-aided diagnosis of lung nodule: Explanation adequacy, inference accuracy, and experts’ knowledge , 2018, PloS one.

[14]  M. J. Maisels,et al.  Managing the jaundiced newborn: a persistent challenge , 2015, Canadian Medical Association Journal.

[15]  Moh. Faturrahman,et al.  Structural MRI classification for Alzheimer's disease detection using deep belief network , 2017, 2017 11th International Conference on Information & Communication Technology and System (ICTS).

[16]  Hilde van der Togt,et al.  Publisher's Note , 2003, J. Netw. Comput. Appl..

[17]  B. Reiser,et al.  Estimation of the Youden Index and its Associated Cutoff Point , 2005, Biometrical journal. Biometrische Zeitschrift.

[18]  R. Scheig.  Neonatal jaundice. , 1974, American family physician.

[19]  Milica M. Badža,et al.  Classification of Brain Tumors from MRI Images Using a Convolutional Neural Network , 2020, Applied Sciences.

[20]  Xiahai Zhuang,et al.  Diagnosis of Alzheimer’s Disease via Multi-Modality 3D Convolutional Neural Network , 2019, Front. Neurosci..

[21]  B. Olusanya,et al.  The burden and management of neonatal jaundice in Nigeria: A scoping review of the literature , 2016, Nigerian journal of clinical practice.

[22]  Ping Zhang,et al.  Studying neonatal bilirubin encephalopathy with conventional MRI, MRS, and DWI , 2008, Neuroradiology.

[23]  Bolei Zhou,et al.  Learning Deep Features for Discriminative Localization , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[24]  Ashok Panigrahy,et al.  Magnetic resonance imaging of bilirubin encephalopathy: current limitations and future promise. , 2014, Seminars in perinatology.

[25]  M. Tatlı.  The role of magnetic resonance imaging in the prediction of the neurodevelopmental outcome of acute bilirubin encephalopathy in newborns , 2009, Turkish Journal of Medical Sciences.

[26]  C. Yin,et al.  Study on automatic detection and classification of breast nodule using deep convolutional neural network system , 2020, Journal of thoracic disease.

[27]  M. Johnston,et al.  Possible Mechanisms in Infants for Selective Basal Ganglia Damage From Asphyxia, Kernicterus, or Mitochondrial Encephalopathies , 2000, Journal of child neurology.

[28]  Emma J Pace,et al.  Neonatal hyperbilirubinemia: An evidence-based approach. , 2019, The Journal of family practice.

[29]  W. Liao,et al.  [Differentiation of hypoxic-ischemic encephalopathy and acute bilirubin encephalopathy with magnetic resonance imaging in neonates]. , 2009, Zhongguo dang dai er ke za zhi = Chinese journal of contemporary pediatrics.

[30]  M. S. Murphy,et al.  Neonatal jaundice: summary of NICE guidance , 2010, BMJ : British Medical Journal.

[31]  Bohyoung Kim,et al.  A computer-aided diagnosis system using artificial intelligence for the diagnosis and characterization of breast masses on ultrasound , 2019, Medicine.

[32]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[33]  U. Rajendra Acharya,et al.  Convolutional neural networks for multi-class brain disease detection using MRI images , 2019, Comput. Medical Imaging Graph..

[34]  Z. Huang,et al.  Effectiveness evaluation of computer-aided diagnosis system for the diagnosis of thyroid nodules on ultrasound , 2019, Medicine.

[35]  Qiang Yang,et al.  A Survey on Transfer Learning , 2010, IEEE Transactions on Knowledge and Data Engineering.

[36]  Enrique F Schisterman,et al.  Youden Index and the optimal threshold for markers with mass at zero , 2008, Statistics in medicine.