Three-dimensional facial-image analysis to predict heterogeneity of the human ageing rate and the impact of lifestyle

Xia, Xian; Chen, Xingwei; Wu, Gang; Fang, Li; Wang, Yiyang; Yang, Chen; Chen, Mingxu; Wang, Xinyu; Chen, Weiyang; Xian, Bo; Chen, Weizhong; Yun, Cao; Xu, Chi; Gong, W. X.; Chen, Guoyu; Cai, Donghong; Wei, Wenxin; Yan, Yizhen; Liu, Kangping; Qiao, Nan; Zhao, Xiaohui; Jia, Jin; Wang, Wei; Kennedy, Brian K.; Zhang, Kang; Cannistraci, Carlo Vittorio; Zhou, Yingling; Han, Jing‐Dong J.

doi:10.1038/s42255-020-00270-x

Cited by 52 publications

(47 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previous studies have demonstrated MAEs of 3·3-5·2 years for DNA methylation clock, 18,19 5·5-5·9 years MAEs for blood profiles, 20,21 and 6·2-7·8 years MAEs for the transcriptome ageing clock. 22,23 Neuroimaging and 3D facial imaging have achieved accurate performances in age prediction with MAEs between 4·3 and 7·3, 7,24 and 2·8 and 6·4 years, 6,25 respectively. Despite these reasonable accuracies, the invasiveness of cellular and molecular ageing biomarkers, high cost and time-consuming nature of neuroimaging and 3D facial ages, and ethical and privacy concerns of facial imaging, have limited their utilities.…”

Section: Discussionmentioning

confidence: 98%

“…Recent studies have demonstrated successful applications of DL models in the prediction of age using clinical images. 5,6,13 Taken together, this raises the potential that biological age can be predicted by applying DL to retinal images. For optimal utility, viable biomarkers of ageing must also relate to the risk of age-related morbidity and mortality.…”

Section: Introductionmentioning

confidence: 88%

See 1 more Smart Citation

Retinal Age as a Predictive Biomarker for Mortality Risk

Zhu

Shi

Peng

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

SummaryBackgroundAgeing varies substantially, thus an accurate quantification of ageing is important. We developed a deep learning (DL) model that predicted age from fundus images (retinal age). We investigated the association between retinal age gap (retinal age-chronological age) and mortality risk in a population-based sample of middle-aged and elderly adults.MethodsThe DL model was trained, validated and tested on 46,834, 15,612 and 8,212 fundus images respectively from participants of the UK Biobank study alive on 28th February 2018. Retinal age gap was calculated for participants in the test (n=8,212) and death (n=1,117) datasets. Cox regression models were used to assess association between retinal age gap and mortality risk. A restricted cubic spline analyses was conducted to investigate possible non-linear association between retinal age gap and mortality risk.FindingsThe DL model achieved a strong correlation of 0·83 (P<0·001) between retinal age and chronological age, and an overall mean absolute error of 3·50 years. Cox regression models showed that each one-year increase in the retinal age gap was associated with a 2% increase in mortality risk (hazard ratio=1·02, 95% confidence interval:1·00-1·04, P=0·021). Restricted cubic spline analyses showed a non-linear relationship between retinal age gap and mortality (Pnon-linear=0·001). Higher retinal age gaps were associated with substantially increased risks of mortality, but only if the gap exceeded 3·71 years.InterpretationOur findings indicate that retinal age gap is a robust biomarker of ageing that is closely related to risk of mortality.FundingNational Health and Medical Research Council Investigator Grant, Science and Technology Program of Guangzhou.Research in contextEvidence before this studyAgeing at an individual level is heterogeneous. An accurate quantification of the biological ageing process is significant for risk stratification and delivery of tailored interventions. To date, cell-, molecular-, and imaging-based biomarkers have been developed, such as epigenetic clock, brain age and facial age. While the invasiveness of cellular and molecular ageing biomarkers, high cost and time-consuming nature of neuroimaging and facial ages, as well as ethical and privacy concerns of facial imaging, have limited their utilities. The retina is considered a window to the whole body, implying that the retina could provide clues for ageing.Added value of this studyWe developed a deep learning (DL) model that can detect footprints of aging in fundus images and predict age with high accuracy for the UK population between 40 and 69 years old. Further, we have been the first to demonstrate that each one-year increase in retinal age gap (retinal age-chronological age) was significantly associated with a 2% increase in mortality risk. Evidence of a non-linear association between retinal age gap and mortality risk was observed. Higher retinal age gaps were associated with substantially increased risks of mortality, but only if the retinal age gap exceeded 3·71 years.Implications of all the available evidenceThis is the first study to link the retinal age gap and mortality risk, implying that retinal age is a clinically significant biomarker of ageing. Our findings show the potential of retinal images as a screening tool for risk stratification and delivery of tailored interventions. Further, the capability to use fundus imaging in predicting ageing may improve the potential health benefits of eye disease screening, beyond the detection of sight-threatening eye diseases.

show abstract

Section: Discussionmentioning

confidence: 98%

Section: Introductionmentioning

confidence: 88%

Retinal Age as a Predictive Biomarker for Mortality Risk

Zhu

Shi

Peng

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…In contrast to these previous studies, AnthropoAge and S-AnthropoAge attempt to capture the contribution of anthropometric measurements to 10-year all-cause mortality risk independently of CA, which may clarify the relationship between body composition and aging beyond BMI, given that this latter metric may not completely capture the complexity of this phenomenon 22,23 . Recent analyses have also shown that a richer diversity of biomarkers may lead to more precise assessments of BA and that aging phenotypes may have influence on body composition and even facial expressions [24][25][26] .…”

Section: Discussionmentioning

confidence: 99%

AnthropoAge, a novel approach to integrate body composition into the estimation of biological age

Márquez‐Salinas¹,

Guerra²,

Zavala-Romero³

et al. 2021

Preprint

View full text Add to dashboard Cite

Methods to estimate biological age (BA) capture different aspects of aging. Here, we consider the role of changes in body composition related to aging as a starting point to incorporate anthropometry into the estimation of BA. To that end, we developed AnthropoAge, a metric to estimate 10-year mortality risk as a proxy of BA using anthropometric and linked mortality data from NHANES-III (n=11,865) and validated it in NHANES-IV (n=7,065). We identified that thigh circumference, arm circumference, body-mass index (BMI), waist-to-height ratio (WHtR) and arm length were useful to predict BA in men, whilst weight, thigh circumference, subscapular and tricipital skinfolds and WHtR in women. We also developed a simplified version of AnthropoAge (S-AnthropoAge) which used only BMI and WHtR, with strong concordance with the complete metric. Both AnthropoAge and S-AnthropoAge were useful to predict 10-year mortality independent of ethnicity, sex, and comorbidities. In comparison to PhenoAge, AnthropoAge/S-AnthropoAge were superior for prediction of cardiovascular, cerebrovascular, cancer-related and nephritis/nephrosis related mortality risk in contrast with other causes. Accelerated aging metrics AnthropoAgeAccel/S-AnthropoAgeAccel identified males with phenotypes of decreased lean and fat mass and females with phenotypes of increased fat mass and increased abdominal adiposity, which likely reflected sexual dimorphisms related to accelerated body composition aging. When jointly assessing PhenoAge and AnthropoAge/S-AnthropoAge, we identified unique aging trajectories with differential mortality risk and comorbidity clustering. AnthropoAge is a useful proxy of BA, which captures cause-specific mortality risk; assessing aging using different BA measures may be useful to better characterize the heterogeneity of the aging process.

show abstract

“…This has generated a large number of objective data of tongue and pulse diagnoses, which are also microscopic indices in a sense. In recent years, studies based on data of tongue and pulse diagnoses have been increasing, with many researchers applying machine learning and data mining methods to the fields of image recognition, target detection, natural language processing, and others [15][16][17][18]. In addition, studies have demonstrated that accurate detection, identification, and multidimensional quantitative analysis based on tongue data and pulse data have been gradually applied to disease diagnosis.…”

Section: Introductionmentioning

confidence: 99%

A New Method for Syndrome Classification of Non-Small-Cell Lung Cancer Based on Data of Tongue and Pulse with Machine Learning

Shi

Liu

et al. 2021

BioMed Research International

View full text Add to dashboard Cite

Objective. To explore the data characteristics of tongue and pulse of non-small-cell lung cancer with Qi deficiency syndrome and Yin deficiency syndrome, establish syndrome classification model based on data of tongue and pulse by using machine learning methods, and evaluate the feasibility of syndrome classification based on data of tongue and pulse. Methods. We collected tongue and pulse of non-small-cell lung cancer patients with Qi deficiency syndrome ( n = 163 ), patients with Yin deficiency syndrome ( n = 174 ), and healthy controls ( n = 185 ) using intelligent tongue diagnosis analysis instrument and pulse diagnosis analysis instrument, respectively. We described the characteristics and examined the correlation of data of tongue and pulse. Four machine learning methods, namely, random forest, logistic regression, support vector machine, and neural network, were used to establish the classification models based on symptom, tongue and pulse, and symptom and tongue and pulse, respectively. Results. Significant difference indices of tongue diagnosis between Qi deficiency syndrome and Yin deficiency syndrome were TB-a, TB-S, TB-Cr, TC-a, TC-S, TC-Cr, perAll, and the tongue coating texture indices including TC-CON, TC-ASM, TC-MEAN, and TC-ENT. Significant difference indices of pulse diagnosis were t4 and t5. The classification performance of each model based on different datasets was as follows: tongue and pulse < symptom < symptom and tongue and pulse. The neural network model had a better classification performance for symptom and tongue and pulse datasets, with an area under the ROC curves and accuracy rate which were 0.9401 and 0.8806. Conclusions. It was feasible to use tongue data and pulse data as one of the objective diagnostic basis in Qi deficiency syndrome and Yin deficiency syndrome of non-small-cell lung cancer.

show abstract

Three-dimensional facial-image analysis to predict heterogeneity of the human ageing rate and the impact of lifestyle

Cited by 52 publications

References 39 publications

Retinal Age as a Predictive Biomarker for Mortality Risk

Retinal Age as a Predictive Biomarker for Mortality Risk

AnthropoAge, a novel approach to integrate body composition into the estimation of biological age

A New Method for Syndrome Classification of Non-Small-Cell Lung Cancer Based on Data of Tongue and Pulse with Machine Learning

Contact Info

Product

Resources

About