Deep Learning Network with Spatial Attention Module for Detecting Acute Bilirubin Encephalopathy in Newborns Based on Multimodal MRI
Abstract
:1. Introduction
2. Materials and Methods
2.1. Study Subjects
2.2. MRI Acquisition
2.3. Image Pre-Processing
2.4. Deep Learning Framework and Spatial Attention Module
2.5. Model Evaluation
3. Results
4. Discussion
5. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Brito, M.A.; Palmela, I.; Cardoso, F.L.; Sá-Pereira, I.; Brites, D. Blood–Brain Barrier and Bilirubin: Clinical Aspects and Experimental Data. Arch. Med. Res. 2014, 45, 660–676. [Google Scholar] [CrossRef] [PubMed]
- Mitra, D.S.; Rennie, D.J. Neonatal jaundice: Aetiology, diagnosis and treatment. Br. J. Hosp. Med. 2017, 78, 699–704. [Google Scholar] [CrossRef]
- Barrington, K.J.; Sankaran, K.; Canadian Paediatric Society; Fetus and Newborn Committee. Guidelines for detection, management and prevention of hyperbilirubinemia in term and late preterm newborn infants (35 or more weeks’ gestation)—Summary. Paediatr. Child Health 2007, 12, 401–418. [Google Scholar] [CrossRef]
- Dai, J.; Parry, D.M.; Krahn, J. Transcutaneous bilirubinometry: Its role in the assessment of neonatal jaundice. Clin. Biochem. 1997, 30, 1–9. [Google Scholar] [CrossRef] [PubMed]
- Wisnowski, J.L.; Panigrahy, A.; Painter, M.J.; Watchko, J.F. Magnetic resonance imaging of bilirubin encephalopathy: Current limitations and future promise. Semin. Perinatol. 2014, 38, 422–428. [Google Scholar] [CrossRef] [PubMed]
- Wang, X.; Wu, W.; Hou, B.L.; Zhang, P.; Chineah, A.; Liu, F.; Liao, W. Studying neonatal bilirubin encephalopathy with conventional MRI, MRS, and DWI. Neuroradiology 2008, 50, 885–893. [Google Scholar] [CrossRef]
- Coskun, A.; Yikilmaz, A.; Kumandas, S.; Karahan, O.I.; Akcakus, M.; Manav, A. Hyperintense globus pallidus on T1-weighted MR imaging in acute kernicterus: Is it common or rare? Eur. Radiol. 2005, 15, 1263–1267. [Google Scholar] [CrossRef]
- Liao, W.-H.; Wang, X.-Y.; Wu, W.-L.; Jiang, X.-Y.; Liu, Y.-H.; Liu, F.; Wang, R.-W. Differentiation of hypoxic-ischemic encephalopathy and acute bilirubin encephalopathy with magnetic resonance imaging in neonates. Zhongguo Dang Dai Er Ke Za Zhi 2009, 11, 181–184. [Google Scholar]
- Cece, H.; Abuhandan, M.; Cakmak, A.; Yildiz, S.; Calik, M.; Karakas, E.; Karakas, O. Diffusion-weighted imaging of patients with neonatal bilirubin encephalopathy. Jpn. J. Radiol. 2013, 31, 179–185. [Google Scholar] [CrossRef]
- Das, S.; van Landeghem, F.K.H. Clinicopathological Spectrum of Bilirubin Encephalopathy/Kernicterus. Diagnostics 2019, 9, 24. [Google Scholar] [CrossRef]
- Joo, S.; Ko, E.S.; Kwon, S.; Jeon, E.; Jung, H.; Kim, J.-Y.; Chung, M.J.; Im, Y.-H. Multimodal deep learning models for the prediction of pathologic response to neoadjuvant chemotherapy in breast cancer. Sci. Rep. 2021, 11, 18800. [Google Scholar] [CrossRef] [PubMed]
- Wu, M.; Shen, X.; Lai, C.; You, Y.; Zhao, Z.; Wu, D. Detecting acute bilirubin encephalopathy in neonates based on multimodal MRI with deep learning. Pediatr. Res. 2021, 91, 1168–1175. [Google Scholar] [CrossRef] [PubMed]
- Liu, Z.; Ji, B.; Zhang, Y.; Cui, G.; Liu, L.; Man, S.; Ding, L.; Yang, X.; Mao, H.; Wang, L. Machine Learning Assisted MRI Characterization for Diagnosis of Neonatal Acute Bilirubin Encephalopathy. Front. Neurol. 2019, 10, 1018. [Google Scholar] [CrossRef] [PubMed]
- Woo, S.; Park, J.; Lee, J.-Y.; Kweon, I.S. CBAM: Convolutional Block Attention Module. In Proceedings of the Computer Vision–ECCV, Munich, Germany, 8–14 September 2018; Springer International Publishing: Cham, Switzerland, 2018; pp. 3–19. [Google Scholar]
- Rashid, T.; Liu, H.; Ware, J.B.; Li, K.; Romero, J.R.; Fadaee, E.; Nasrallah, I.M.; Hilal, S.; Bryan, R.N.; Hughes, T.M.; et al. Deep learning based detection of enlarged perivascular spaces on brain MRI. Neuroimage Rep. 2023, 3, 100162. [Google Scholar] [CrossRef] [PubMed]
- Guo, S.; Wang, L.; Chen, Q.; Wang, L.; Zhang, J.; Zhu, Y. Multimodal MRI Image Decision Fusion-Based Network for Glioma Classification. Front. Oncol. 2022, 12, 819673. [Google Scholar] [CrossRef]
- Shapiro, S.M. Definition of the Clinical Spectrum of Kernicterus and Bilirubin-Induced Neurologic Dysfunction (BIND). J. Perinatol. 2005, 25, 54–59. [Google Scholar] [CrossRef]
- Smith, S.M.; Jenkinson, M.; Woolrich, M.W.; Beckmann, C.F.; Behrens, T.E.; Johansen-Berg, H.; Bannister, P.R.; De Luca, M.; Drobnjak, I.; Flitney, D.E.; et al. Advances in functional and structural MR image analysis and implementation as FSL. Neuroimage 2004, 23 (Suppl. S1), S208–S219. [Google Scholar] [CrossRef]
- Hoopes, A.; Mora, J.S.; Dalca, A.V.; Fischl, B.; Hoffmann, M. SynthStrip: Skull-stripping for any brain image. NeuroImage 2022, 260, 119474. [Google Scholar] [CrossRef]
- He, K.; Zhang, X.; Ren, S.; Sun, J. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA, 27–30 June 2016; pp. 770–778. [Google Scholar]
- Deng, J.; Dong, W.; Socher, R.; Li, L.J.; Kai, L.; Li, F.-F. ImageNet: A large-scale hierarchical image database. In Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami Beach, FL, USA, 20–25 June 2009; pp. 248–255. [Google Scholar]
- Kingma, D.P.; Ba, J. Adam: A method for stochastic optimization. arXiv 2014, arXiv:1412.6980. [Google Scholar]
- Govaert, P.; Lequin, M.; Swarte, R.; Robben, S.; De Coo, R.; Weisglas-Kuperus, N.; De Rijke, Y.; Sinaasappel, M.; Barkovich, J. Changes in globus pallidus with (pre)term kernicterus. Pediatrics 2003, 112 Pt 1, 1256–1263. [Google Scholar] [CrossRef]
- Ribeiro, B.N.; Lima, G.A.; Ventura, N.; Gasparetto, E.L.; Marchiori, E. Chronic kernicterus: Magnetic resonance imaging findings. Radiol. Bras. 2016, 49, 407–408. [Google Scholar] [CrossRef] [PubMed]
- Assefa Neknek, G.; Woldemichael, K.; Moges, A.; Zewdneh Solomon, D. MRI of bilirubin encephalopathy (kernicterus): A case series of 4 patients from Sub-Saharan Africa, May 2017. Radiol. Case Rep. 2018, 13, 676–679. [Google Scholar] [CrossRef] [PubMed]
- Vaswani, A.; Shazeer, N.; Parmar, N.; Uszkoreit, J.; Jones, L.; Gomez, A.N.; Kaiser, Ł.; Polosukhin, I.J. Ainips: Attention is all you need. Proc. Adv. Neural Inf. Process. Syst. 2017, 30, 6000–6010. [Google Scholar]
- Dosovitskiy, A.; Beyer, L.; Kolesnikov, A.; Weissenborn, D.; Zhai, X.; Unterthiner, T.; Dehghani, M.; Minderer, M.; Heigold, G.; Gelly, S.J. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv 2020, arXiv:2010.11929. [Google Scholar]
- Ranaldi, L.; Pucci, G. Knowing Knowledge: Epistemological Study of Knowledge in Transformers. Appl. Sci. 2023, 13, 677. [Google Scholar] [CrossRef]
Clinical Features | ABE (n = 97) | Non-ABE (n = 80) | p-Value |
---|---|---|---|
Gender (male/female) | 49/48 | 49/31 | 0.15 |
Weight (kg) | 3.80 ± 0.70 | 4.06 ± 1.20 | 0.19 |
Age (days) | 9.86 ± 5.76 | 16.04 ± 12.57 | 0.03 |
MRI Modality | Accuracy | AUC | Sensitivity | Specificity | Precision Score | F1 Score |
---|---|---|---|---|---|---|
T1 | 0.666 ± 0.107 | 0.706 ± 0.141 | 0.669 ± 0.166 | 0.662 ± 0.084 | 0.701 ± 0.088 | 0.681 ± 0.121 |
T2 | 0.745 ± 0.062 | 0.804 ± 0.071 | 0.708 ± 0.155 | 0.787 ± 0.071 | 0.805 ± 0.030 | 0.745 ± 0.088 |
ADC | 0.583 ± 0.079 | 0.633 ± 0.011 | 0.527 ± 0.096 | 0.650 ± 0.011 | 0.648 ± 0.081 | 0.579 ± 0.080 |
T1 + ADC | 0.656 ± 0.109 | 0.721 ± 0.088 | 0.662 ± 0.197 | 0.650 ± 0.130 | 0.695 ± 0.089 | 0.670 ± 0.124 |
T2 + ADC | 0.690 ± 0.066 | 0.779 ± 0.043 | 0.672 ± 0.194 | 0.712 ± 0.144 | 0.747 ± 0.054 | 0.693 ± 0.100 |
T1 + T2 | 0.763 ± 0.029 | 0.816 ± 0.021 | 0.836 ± 0.096 | 0.675 ± 0.103 | 0.761 ± 0.039 | 0.793 ± 0.034 |
T1 + T2 + ADC | 0.673 ± 0.052 | 0.674 ± 0.069 | 0.690 ± 0.118 | 0.650 ± 0.130 | 0.712 ± 0.061 | 0.695 ± 0.062 |
MRI Modality | Accuracy | AUC | Sensitivity | Specificity | Precision Score | F1 Score |
---|---|---|---|---|---|---|
T1 | 0.674 ± 0.155 | 0.736 ± 0.143 | 0.58 ± 0.164 | 0.788 ± 0.18 | 0.777 ± 0.179 | 0.659 ± 0.156 |
T2 | 0.768 ± 0.029 | 0.796 ± 0.039 | 0.763 ± 0.149 | 0.775 ± 0.144 | 0.82 ± 0.075 | 0.778 ± 0.048 |
ADC | 0.576 ± 0.016 | 0.638 ± 0.034 | 0.525 ± 0.081 | 0.637 ± 0.120 | 0.646 ± 0.050 | 0.573 ± 0.034 |
T1 + ADC | 0.696 ± 0.115 | 0.713 ± 0.122 | 0.652 ± 0.174 | 0.750 ± 0.088 | 0.755 ± 0.087 | 0.694 ± 0.130 |
T2 + ADC | 0.644 ± 0.123 | 0.735 ± 0.053 | 0.525 ± 0.211 | 0.787 ± 0.114 | 0.726 ± 0.158 | 0.598 ± 0.209 |
T1 + T2 | 0.808 ± 0.069 | 0.808 ± 0.057 | 0.856 ± 0.083 | 0.750 ± 0.076 | 0.806 ± 0.057 | 0.830 ± 0.064 |
T1 + T2 + ADC | 0.678 ± 0.086 | 0.764 ± 0.090 | 0.650 ± 0.120 | 0.713 ± 0.169 | 0.744 ± 0.102 | 0.686 ± 0.085 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Zhang, H.; Zhuang, Y.; Xia, S.; Jiang, H. Deep Learning Network with Spatial Attention Module for Detecting Acute Bilirubin Encephalopathy in Newborns Based on Multimodal MRI. Diagnostics 2023, 13, 1577. https://doi.org/10.3390/diagnostics13091577
Zhang H, Zhuang Y, Xia S, Jiang H. Deep Learning Network with Spatial Attention Module for Detecting Acute Bilirubin Encephalopathy in Newborns Based on Multimodal MRI. Diagnostics. 2023; 13(9):1577. https://doi.org/10.3390/diagnostics13091577
Chicago/Turabian StyleZhang, Huan, Yi Zhuang, Shunren Xia, and Haoxiang Jiang. 2023. "Deep Learning Network with Spatial Attention Module for Detecting Acute Bilirubin Encephalopathy in Newborns Based on Multimodal MRI" Diagnostics 13, no. 9: 1577. https://doi.org/10.3390/diagnostics13091577
APA StyleZhang, H., Zhuang, Y., Xia, S., & Jiang, H. (2023). Deep Learning Network with Spatial Attention Module for Detecting Acute Bilirubin Encephalopathy in Newborns Based on Multimodal MRI. Diagnostics, 13(9), 1577. https://doi.org/10.3390/diagnostics13091577