Uncertainty-Based Biological Age Estimation of Brain MRI Scans

Age is an essential factor in modern diagnostic procedures. However, assessment of the true biological age (BA) remains a daunting task due to the lack of reference ground-truth labels. Current BA estimation approaches are either restricted to skeletal images or rely on non-imaging modalities that yield a whole-body BA assessment. However, various organ systems may exhibit different aging characteristics due to lifestyle and genetic factors. In this initial study, we propose a new framework for organ-specific BA estimation utilizing 3D magnetic resonance image (MRI) scans. As a first step, this framework predicts the chronological age (CA) together with the corresponding patient-dependent aleatoric uncertainty. An iterative training algorithm is then utilized to segregate atypical aging patients from the given population based on the predicted uncertainty scores. In this manner, we hypothesize that training a new model on the remaining population should approximate the true BA behavior. We apply the proposed methodology on a brain MRI dataset containing healthy individuals as well as Alzheimer’s patients. We demonstrate the correlation between the predicted BAs and the expected cognitive deterioration in Alzheimer’s patients.

[1]  Stuart J. Ritchie,et al.  Brain age predicts mortality , 2017, Molecular Psychiatry.

[2]  Evgeny Putin,et al.  Deep biomarkers of human aging: Application of deep neural networks to biomarker development , 2016, Aging.

[3]  Kailash D. Kharat,et al.  Brain Tumor Classification Using Neural Network Based Methods , 2012 .

[4]  Richie Poulton,et al.  Eleven Telomere, Epigenetic Clock, and Biomarker-Composite Quantifications of Biological Aging: Do They Measure the Same Thing? , 2017, American journal of epidemiology.

[5]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[6]  Vera Ignjatovic,et al.  Age-Related Differences in Plasma Proteins: How Plasma Proteins Change from Neonates to Adults , 2011, PloS one.

[7]  Luigi Ferrucci,et al.  A new aging measure captures morbidity and mortality risk across diverse subpopulations from NHANES IV: A cohort study , 2018, PLoS medicine.

[8]  Bin Yang,et al.  Towards Adversarial Denoising of Radar Micro-Doppler Signatures , 2018, 2019 International Radar Conference (RADAR).

[9]  Syed Ashiqur Rahman,et al.  Deep Learning using Convolutional LSTM estimates Biological Age from Physical Activity , 2019, Scientific Reports.

[10]  Kenneth Rockwood,et al.  Heterogeneity of Human Aging and Its Assessment , 2016, The journals of gerontology. Series A, Biological sciences and medical sciences.

[11]  Weiguang Zhang,et al.  Common methods of biological age estimation , 2017, Clinical interventions in aging.

[12]  James H. Cole,et al.  Neuroimaging-derived brain-age: an ageing biomarker? , 2017, Aging.

[13]  Giovanni Montana,et al.  Predicting brain age with deep learning from raw imaging data results in a reliable and heritable biomarker , 2016, NeuroImage.

[14]  Bin Yang,et al.  ipA-MedGAN: Inpainting of Arbitrary Regions in Medical Imaging , 2019, 2020 IEEE International Conference on Image Processing (ICIP).

[15]  Daniel S. Marcus,et al.  OASIS-3: Longitudinal Neuroimaging, Clinical, and Cognitive Dataset for Normal Aging and Alzheimer Disease , 2019 .

[16]  Forrest N. Iandola,et al.  SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <1MB model size , 2016, ArXiv.

[17]  M. Levine,et al.  DNA methylation-based measures of biological age: meta-analysis predicting time to death , 2016, Aging.

[18]  Andrew Zisserman,et al.  Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[19]  Syed Ashiqur Rahman,et al.  Centroid of Age Neighborhoods: A New Approach to Estimate Biological Age , 2020, IEEE Journal of Biomedical and Health Informatics.

[20]  Daniel C. Alexander,et al.  Uncertainty modelling in deep learning for safer neuroimage enhancement: Demonstration in diffusion MRI , 2020, NeuroImage.

[21]  Beate Sick,et al.  Integrating uncertainty in deep neural networks for MRI based stroke analysis , 2020, Medical Image Anal..

[22]  K. Nikolaou,et al.  Independent brain F-FDG PET attenuation correction using a deep learning approach with Generative Adversarial Networks , 2019 .

[23]  Morteza Mardani,et al.  Uncertainty Quantification in Deep MRI Reconstruction , 2021, IEEE Transactions on Medical Imaging.

[24]  Martin Urschler,et al.  What automated age estimation of hand and wrist MRI data tells us about skeletal maturation in male adolescents , 2015, Annals of human biology.

[25]  Nancy L. Pedersen,et al.  Biological Age Predictors , 2017, EBioMedicine.

[26]  Julien Cornebise,et al.  Weight Uncertainty in Neural Networks , 2015, ArXiv.

[27]  M. Levine.  Modeling the rate of senescence: can estimated biological age predict mortality more accurately than chronological age? , 2013, The journals of gerontology. Series A, Biological sciences and medical sciences.

[28]  Alex Kendall,et al.  What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision? , 2017, NIPS.

[29]  Nuzhat Hassan,et al.  Bone Age Assessment Methods: A Critical Review , 1969, Pakistan journal of medical sciences.

[30]  Artem Sevastopolsky,et al.  PhotoAgeClock: deep learning algorithms for development of non-invasive visual biomarkers of aging , 2018, Aging.

[31]  Petr Klemera,et al.  A new approach to the concept and computation of biological age , 2006, Mechanisms of Ageing and Development.

[32]  E. Nakamura,et al.  A method for identifying biomarkers of aging and constructing an index of biological age in humans. , 2007, The journals of gerontology. Series A, Biological sciences and medical sciences.

[33]  K. Nikolaou,et al.  Independent attenuation correction of whole body [18F]FDG-PET using a deep learning approach with Generative Adversarial Networks , 2020, EJNMMI Research.

[34]  Ben Glocker,et al.  Deep Learning Methods for Estimating "Brain Age" from Structural MRI Scans , 2018 .

[35]  Urs Schneider,et al.  An Adversarial Super-Resolution Remedy for Radar Design Trade-offs , 2019, 2019 27th European Signal Processing Conference (EUSIPCO).

[36]  Bin Yang,et al.  Organ-Based Chronological Age Estimation Based on 3D MRI Scans , 2019, 2020 28th European Signal Processing Conference (EUSIPCO).

[37]  Daniel Weiskopf,et al.  Age-Net: An MRI-Based Iterative Framework for Brain Biological Age Estimation , 2021, IEEE Transactions on Medical Imaging.

[38]  Gianfranco Doretto,et al.  Deep learning for biological age estimation , 2020, Briefings Bioinform..

[39]  Alexander J. Smola,et al.  Heteroscedastic Gaussian process regression , 2005, ICML.

[40]  Christian Gaser,et al.  Longitudinal Changes in Individual BrainAGE in Healthy Aging, Mild Cognitive Impairment, and Alzheimer’s Disease , 2012 .

[41]  Kazunori Sato,et al.  Age estimation from brain MRI images using deep learning , 2017, 2017 IEEE 14th International Symposium on Biomedical Imaging (ISBI 2017).

[42]  Ersilia Barbato,et al.  Value of MRI of the hand and the wrist in evaluation of bone age: Preliminary results , 2014, Journal of magnetic resonance imaging : JMRI.

[43]  Yaakov Stern,et al.  The Clinical Dementia Rating scale , 1996, Neurology.

[44]  Timothy V Pyrkov,et al.  Extracting biological age from biomedical data via deep learning: too much of a good thing? , 2017, Scientific Reports.

[45]  Christian Payer,et al.  Reducing acquisition time for MRI-based forensic age estimation , 2018, Scientific Reports.

[46]  Christian Payer,et al.  Automatic Age Estimation and Majority Age Classification From Multi-Factorial MRI Data , 2019, IEEE Journal of Biomedical and Health Informatics.

[47]  Kyung S. Park,et al.  An empirical comparative study on biological age estimation algorithms with an application of Work Ability Index (WAI) , 2010, Mechanisms of Ageing and Development.