Application of Machine Vision in Classifying Gait Frailty Among Older Adults

Background: Frail older adults have an increased risk of adverse health outcomes and premature death. They also exhibit altered gait characteristics in comparison with healthy individuals.Methods: In this study, we created a Fried’s frailty phenotype (FFP) labelled casual walking video set of older adults based on the West China Health and Aging Trend study. A series of hyperparameters in machine vision models were evaluated for body key point extraction (AlphaPose), silhouette segmentation (Pose2Seg, DPose2Seg, and Mask R-CNN), gait feature extraction (Gaitset, LGaitset, and DGaitset), and feature classification (AlexNet and VGG16), and were highly optimised during analysis of gait sequences of the current dataset.Results: The area under the curve (AUC) of the receiver operating characteristic (ROC) at the physical frailty state identification task for AlexNet was 0.851 (0.827–0.8747) and 0.901 (0.878–0.920) in macro and micro, respectively, and was 0.855 (0.834–0.877) and 0.905 (0.886–0.925) for VGG16 in macro and micro, respectively. Furthermore, this study presents the machine vision method equipped with better predictive performance globally than age and grip strength, as well as than 4-m-walking-time in healthy and pre-frailty classifying.Conclusion: The gait analysis method in this article is unreported and provides promising original tool for frailty and pre-frailty screening with the characteristics of convenience, objectivity, rapidity, and non-contact. These methods can be extended to any gait-related disease identification processes, as well as in-home health monitoring.

Download Full-text

Linking Persistent Pain and Frailty in Older Adults

Pain Medicine ◽

10.1093/pm/pnz174 ◽

2019 ◽

Cited By ~ 1

Author(s):

Fabio Guerriero ◽

M C Reid

Keyword(s):

Older Adults ◽

Persistent Pain ◽

Geriatric Syndrome ◽

Frailty Phenotype ◽

Adverse Health Outcomes ◽

Increased Risk ◽

Phenotype Model ◽

Increasing Risk ◽

The Impact ◽

The Relationship

Abstract Objective In older adults, the impact of persistent pain goes beyond simple discomfort, often contributing to worsening functional outcomes and ultimately frailty. Frailty is a geriatric syndrome that, like persistent pain, increases in prevalence with age and is characterized by a decreased ability to adapt to common stressors such as acute illness, thereby increasing risk for multiple adverse health outcomes. Evidence supports a relationship between persistent pain and both the incidence and progression of frailty, independent of health, social, and lifestyle confounders. Design and Setting In this article, we synthesize recent evidence linking persistent pain and frailty in an effort to clarify the nature of the relationship between these two commonly occurring geriatric syndromes. Setting We propose an integration of the frailty phenotype model by considering the impact of persistent pain on vulnerability toward external stressors, which can ultimately contribute to frailty in older adults. Results and Conclusions Incorporating persistent pain into the frailty construct can help us better understand frailty and ultimately improve care for patients with, as well as those at increased risk for, pain and frailty.

Download Full-text

The Framingham risk score is associated with incident frailty, or is it?

BMC Geriatrics ◽

10.1186/s12877-021-02387-4 ◽

2021 ◽

Vol 21 (1) ◽

Author(s):

Hui Shi ◽

Mei-Ling Ge ◽

Birong Dong ◽

Qian-Li Xue

Keyword(s):

Older Adults ◽

Risk Score ◽

Cox Regression ◽

Activity Level ◽

Cvd Risk ◽

Frailty Phenotype ◽

Frailty Status ◽

Adverse Health Outcomes ◽

Comorbidity Burden

Abstract Backgrounds Cardiovascular disease (CVD) risk factors are individually associated with frailty. This study examined whether Framingham CVD risk score (FRS) as an aggregate measure of CVD risk is associated with incident frailty among Chinese older adults. Methods This study used data from the China Health and Retirement Longitudinal Study. A sample of 3,618 participants aged 60 to 95 years and without CVD at baseline were followed for four years. FRS was calculated at baseline. Frailty status was defined as not-frail (0–2 criteria) or frail (3–5 criteria) based on the physical frailty phenotype consisting of five binary criteria (weakness, slowness, exhaustion, low activity level, and weight loss). After excluding subjects who were frail (n = 248) at baseline, discrete-time Cox regression was used to evaluate the relationship between FRS and incident frailty. Results During a median follow-up of 4.0 years, 323 (8 %) participants developed CVD and 318 (11 %) subjects had frailty onset. Higher FRS was associated with greater risk of incident frailty (HR: 1.03, 95 % CI: 1.00 to 1.06) after adjusting for education, marital status, obesity, comorbidity burden, and cognitive function. This association however was no longer significant (HR: 1.00, 95 % CI: 0.97 to 1.03) after additionally adjusting for age. These findings remained essentially unchanged after excluding subjects with depression (n = 590) at baseline or incident CVD (n = 323) during the 4-year follow-up. Conclusions The FRS was not independently associated with incident frailty after adjusting for chronological age. More research is needed to assess the clinical utility of the FRS in predicting adverse health outcomes other than CVD in older adults.

Download Full-text

Oxidative stress, telomere length, and frailty in an old age population

10.1101/414680 ◽

2018 ◽

Author(s):

José Darío Martínez-Ezquerro ◽

Aleida Rodriguez-Castañeda ◽

Mauricio Ortiz-Ramirez ◽

Sergio Sanchez-Garcia ◽

Haydee Rosas-Vargas ◽

...

Keyword(s):

Oxidative Stress ◽

Older Adults ◽

Telomere Length ◽

Old Age ◽

Cross Sectional Study ◽

Frailty Phenotype ◽

Cross Sectional ◽

Adverse Health Outcomes ◽

Biomarkers Of Aging ◽

Dichlorofluorescin Diacetate

AbstractBackgroundA global aging population requires focusing on the risk factors for unhealthy aging, preventive medicine, and chronic disease management. The identification of adverse health outcomes in older adults has been addressed by the characterization of frailty as a biological syndrome. On the other hand, oxidative stress and telomere length have been suggested as biomarkers of aging.ObjectiveTo study the association of oxidative stress, telomere length, and frailty in an old age population.MethodsThis was a cross-sectional study based on 2015 data from 202 members from a cohort of older adults (n=202; gender F/M ratio: 133/69; mean age: 69.89 ± 7.39 years). Reactive oxygen species (ROS) were measured by dichlorofluorescin diacetate, and lipid peroxidation by malondialdehyde. Telomere length was determined using qPCR with SYBR Green Master Mix.ResultsStatistical analysis showed an association between telomere length and frailty but no association between oxidative stress on telomere length or frailty.ConclusionsTelomere length could eventually be used as a marker to discriminate between healthy and unhealthy aging as expressed by frailty phenotype. However, oxidative stress seems as just a biological process of aging.

Download Full-text

Using Dietary Macronutrient Patterns to Predict Sarcopenic Obesity in Older Adults: A Representative Korean Nationwide Population-Based Study

Nutrients ◽

10.3390/nu13114031 ◽

2021 ◽

Vol 13 (11) ◽

pp. 4031

Author(s):

Jun-Hyuk Lee ◽

Hye-Min Park ◽

Yong-Jae Lee

Keyword(s):

Older Adults ◽

Population Based ◽

Sarcopenic Obesity ◽

Calorie Intake ◽

Macronutrient Intake ◽

Dietary Recall ◽

Adverse Health Outcomes ◽

Increased Risk ◽

Total Calorie ◽

Total Calorie Intake

Older adults with sarcopenic obesity (SO) are at increased risk of adverse health outcomes. It has not been identified which pattern of macronutrient intake is appropriate in relation to SO. We aimed to compare the patterns of macronutrient intake for predicting SO in older adults. Data from a total of 3828 older adults who participated in the 2008–2011 Korea National Health and Nutrition Examination Survey were analyzed. The one-day 24 h dietary recall method was used to assess macronutrient intake. SO was defined by a combination of body mass index (BMI) ≥ 25 kg/m2 and BMI adjusted-appendicular skeletal muscle mass <0.789 for men and <0.512 for women. Weighted logistic regression analysis revealed the odds ratio (95% confidence interval) for SO of total calorie intake per 100 increments and carbohydrate (CHO) intake (g/kg/day) per 1 increment to be 0.95 (0.91–0.99) and 0.83 (0.74–0.94), respectively, after adjusting for confounding variables in women. The predictive power for SO of CHO intake (g/kg/day) was higher compared with the other patterns of macronutrient intake both in men and women. In conclusion, total calorie intake and CHO intake (g/kg/day) are inversely related to SO in women. CHO intake (g/kg/day) could be the best index for determining SO.

Download Full-text

Predicting tuberculosis relapse in patients treated with the standard 6-month regimen: an individual patient data meta-analysis

Thorax ◽

10.1136/thoraxjnl-2017-211120 ◽

2018 ◽

Vol 74 (3) ◽

pp. 291-297 ◽

Cited By ~ 7

Author(s):

Kamila Romanowski ◽

Robert F Balshaw ◽

Andrea Benedetti ◽

Jonathon R Campbell ◽

Dick Menzies ◽

...

Keyword(s):

Individual Patient Data ◽

Reference Model ◽

Meta Analysis ◽

Area Under The Curve ◽

Predictive Performance ◽

Patient Data ◽

Standard Regimen ◽

Increased Risk ◽

Cavitary Disease ◽

Positive Smear

BackgroundRelapse continues to place significant burden on patients and tuberculosis (TB) programmes worldwide. We aimed to determine clinical and microbiological factors associated with relapse in patients treated with the WHO standard 6-month regimen and then evaluate the accuracy of each factor at predicting an outcome of relapse.MethodsA systematic review was performed to identify randomised controlled trials reporting treatment outcomes on patients receiving the standard regimen. Authors were contacted and invited to share patient-level data (IPD). A one-step IPD meta-analysis, using random intercept logistic regression models and receiver operating characteristic curves, was performed to evaluate the predictive performance of variables of interest.ResultsIndividual patient data were obtained from 3 of the 12 identified studies. Of the 1189 patients with confirmed pulmonary TB who completed therapy, 67 (5.6%) relapsed. In multipredictor analysis, the presence of baseline cavitary disease with positive smear at 2 months was associated with an increased odds of relapse (OR 2.3(95% CI 1.3 to 4.2)) and a relapse risk of 10%. When area under the curve for each multipredictor model was compared, discrimination between low-risk and higher-risk patients was modest and similar to that of the reference model which accounted for age, sex and HIV status.ConclusionDespite its poor predictive value, our results indicate that the combined presence of cavitary disease and 2-month positive smear status may be the best currently available marker for identifying individuals at an increased risk of relapse, particularly in resource-limited setting. Further investigation is required to assess whether this combined factor can be used to indicate different treatment requirements in clinical practice.

Download Full-text

Perceived Neighborhood Environmental Factors Related to Sarcopenia in Urban-Dwelling Older Adults

Innovation in Aging ◽

10.1093/geroni/igaa057.1418 ◽

2020 ◽

Vol 4 (Supplement_1) ◽

pp. 439-439

Author(s):

Yuri Seo ◽

Miji Kim ◽

Hayoung Shim ◽

Heeeun Jung ◽

Seoyoon Jane Lee ◽

...

Keyword(s):

Older Adults ◽

Environmental Factors ◽

Public Transportation ◽

Urban Areas ◽

Cross Sectional Study ◽

Community Dwelling ◽

Older Individuals ◽

Cross Sectional ◽

Adverse Health Outcomes ◽

Increased Risk

Abstract Sarcopenia is common among older individuals and has adverse health outcomes. However, little is known about its association with neighborhood environmental factors. We explored the relationship between sarcopenia and neighborhood environmental factors among community-dwelling older adults aged 70–84 years in urban areas in the Korean Frailty and Aging Cohort Study. There were 1,776 participants in this cross-sectional study (mean age 75.9±3.8 years, 54.1% women). Sarcopenia was defined using the Asian Working Group for Sarcopenia guidelines. The neighborhood environmental factors were assessed using the 17-item Environmental Module of the International Physical Activity Questionnaire (IPAQ-E). The prevalence of sarcopenia was 22.5%. In the multivariate analysis adjusted for potential confounders, compared to the 5th quintile of the IPAQ-E score, the odds ratio (OR and 95% confidence interval [CI]) for sarcopenia in the 1st, 2nd, 3rd and 4th quintile were 2.14 (1.41-3.26), 1.70 (1.11-2.61), 1.76 (1.16-2.68) and 1.62 (1.07-2.47), respectively. Sarcopenia was associated with environmental factors including access to destinations (β = -0.015) and neighborhood safety (β = -0.008) (all p<0.05). Furthermore, no access to public transportation (OR 2.05, 95% CI 1.20-3.50), poor access to recreational facilities (OR 1.40, 95% CI 1.02-1.92), no presence of destination (OR 1.53, 95% CI 1.07-2.21), hill hazard (OR 1.34, 95% CI 1.02-1.77), and lack of safety from traffic (OR 1.35, 95% CI 1.02-1.79) was associated with an increased risk of sarcopenia. Our study suggests that neighborhood environmental characteristics are associated with sarcopenia and better neighborhood environmental strategies can help prevent sarcopenia among older adults.

Download Full-text

World Health Organization growth standards: How do Canadian children measure up?

Paediatrics & Child Health ◽

10.1093/pch/pxaa053 ◽

2020 ◽

Author(s):

Celia Rodd ◽

Allison Feely ◽

Allan B Becker ◽

Theo J Moraes ◽

Padmaja Subbarao ◽

...

Keyword(s):

World Health Organization ◽

Growth Parameters ◽

Early Recognition ◽

Area Under The Curve ◽

Predictive Performance ◽

World Health ◽

Growth Standards ◽

Increased Risk ◽

Z Scores ◽

Health Organization

Abstract Background World Health Organization (WHO) growth standards for children aged 0 to 5 years describe growth under optimal conditions and were adopted for use in Canada in 2012. We are seeking to validate these charts in a well-characterized, longitudinal cohort of healthy, Canadian youngsters, assess tracking over time, and evaluate the prognostic implications of early growth. Methods Data from 2,795 mother–infant dyads from the CHILD birth cohort were classified by feeding modality at 6 months as exclusively breastfed, partially breastfed, or formula-fed. WHO z-scores (z) were calculated at birth, 3 months, 1 year, and 3 years. Receiver operator characteristics (ROC) assessed the predictive performance of early weight (WT), weight-for-length (WfL), or body mass index (BMI) z-scores for overweight/obesity at 3 years. Results Compared to WHO standards, Canadian children at birth had lower median WfLz (−0.73) and BMIz (−0.29), with more positive scores by 3 years (WfLz=BMIz=0.58). At both 1 and 3 years, formula feeding was associated with higher scores than breastfeeding, even after regression adjustment for covariates. Head circumference z-score was typically positive at all times and regardless of feeding modality. At 1 year, ROC area under the curve was 0.79 for WTz, WfLz, and BMIz, and BMIz>0.88 identified children with increased risk of overweight/obesity (BMIz >2) at age 3 years (20.3% versus 3.0%, P<0.001). Conclusions Compared to WHO growth charts, Canadian children at 3 years show an upward shift in BMIz and WfLz, particularly when formula-fed. Infant growth parameters may identify infants with increased risk of overweight/obesity at age 3 years; early recognition may allow targeting infants at higher risk.

Download Full-text

Self-Efficacy Proxy Predicts Physical Frailty Incidence Over 8 Years in Non-Institutionalized Older Adults

Innovation in Aging ◽

10.1093/geroni/igaa057.1263 ◽

2020 ◽

Vol 4 (Supplement_1) ◽

pp. 392-392

Author(s):

Melissa Hladek ◽

Jiafeng Zhu ◽

Brian Buta ◽

Sarah Szanton ◽

Karen Bandeen-Roche ◽

...

Keyword(s):

Older Adults ◽

Activities Of Daily Living ◽

Sex Education ◽

Self Efficacy ◽

Daily Living ◽

Functional Decline ◽

Adverse Outcomes ◽

Physical Frailty ◽

Frailty Phenotype ◽

Increased Risk

Abstract Physical frailty is defined as a syndrome of decreased physiologic reserve conferring vulnerability to functional decline, mortality and other adverse outcomes in response to a stressor. One potential modifiable risk factor of frailty is self-efficacy, which is confidence in one’s ability to perform well at a task or domain in life. Self-efficacy is associated with improved health behavior and decreased chronic disease burden but has not been studied extensively in frailty research. Therefore, the purpose of this study was to evaluate a general self-efficacy proxy measure’s ability to predict frailty in a nationally representative sample of older adults using data from the National Health and Aging Trends Study (NHATS) collected from 2011-2018. 4,835 older adults (65+) were dichotomized into low and high self-efficacy groups using the one-item self-efficacy proxy measure in NHATS. The Physical Frailty Phenotype was used to assess frailty. A discrete time hazard model was used to obtain incident hazard ratios of frailty in two models. Model 1 was adjusted for age, race, sex, education and income. Model 2 contained Model 1 covariates and activities of daily living and co-morbidities. We found that low self-efficacy predicted a 41% increased risk of developing frailty over 8 years after adjustment for socio-demographics (P<0.0001) and a 27% risk of incident frailty after further adjustment for activities of daily living and co-morbidities (P=0.004). This study provides preliminary evidence that self-efficacy may be a key modifiable element to incorporate into multi-modal frailty interventions.

Download Full-text

THE ASSOCIATION BETWEEN LOWER EXTREMITY FUNCTION, FRAILTY, AND LOW-MILEAGE DRIVER STATUS AMONG OLDER ADULTS

Innovation in Aging ◽

10.1093/geroni/igz038.2528 ◽

2019 ◽

Vol 3 (Supplement_1) ◽

pp. S685-S685

Author(s):

Christopher L Crowe ◽

Howard Andrews ◽

Lisa J Molnar ◽

David W Eby ◽

Carolyn DiGuiseppi ◽

...

Keyword(s):

Older Adults ◽

Age Groups ◽

Physical Capacity ◽

Older Drivers ◽

Lower Extremity Function ◽

Frailty Phenotype ◽

Crash Rates ◽

Increased Risk ◽

Binomial Regression ◽

Elevated Rate

Abstract The crash rate per mile driven among older adults is higher than that of most age groups and comparable to that of the youngest, most inexperienced drivers. The low-mileage bias posits that the elevated rate among older adults results from an increased rate among those who accrue the fewest annual miles. This study evaluated whether low physical capacity among older drivers, measured by the National Health and Aging Trends Study (NHATS) Expanded Short Physical Performance Battery (SPPB) and Fried’s frailty phenotype, increases the risk of being low-mileage drivers. Data were collected for 2,990 older drivers via questionnaires and assessments in addition to 61,528 person-months of driving data. Multivariable log-binomial regression was used to estimate risk ratios. Those with fair and good function had 0.53 (95% CI: 0.40-0.69) and 0.60 (0.47-0.78) times the risk of driving fewer than 3,000 miles/year and 0.45 (0.26-0.77) and 0.48 (0.32-0.72) times the risk of driving fewer than 1,865 miles/year, respectively, compared to those with poor function. For an increase from not frail to pre-frail and from pre-frail to frail, the risk of driving fewer than 3,000 or 1,865 miles/year increased 1.36 (1.11-1.65) or 2.38 (1.63-3.46) times, respectively. Having low physical capacity is associated with an increased risk of low annual mileage. Given the known association between low-mileage driver status and increased crash rates and the modifiable nature of the risk factors examined in this study, interventions aimed at improving physical capabilities may lead to an improvement in safety among older drivers.

Download Full-text

Reliability and validity of the self-reported frailty screening questionnaire in older adults

Therapeutic Advances in Chronic Disease ◽

10.1177/2040622320904278 ◽

2020 ◽

Vol 11 ◽

pp. 204062232090427 ◽

Cited By ~ 1

Author(s):

Yanhong Zhang ◽

Yaxin Zhang ◽

Yun Li ◽

Piu Chan ◽

Lina Ma

Keyword(s):

Older Adults ◽

Weight Loss ◽

Intraclass Correlation ◽

Area Under The Curve ◽

Reliability And Validity ◽

Adverse Outcomes ◽

Measurement Tool ◽

Frailty Phenotype ◽

Screening Questionnaire ◽

Sensitivity Specificity

Background: Frailty is one of the most important risk factors for adverse outcomes in older adults. Despite a high prevalence, there is still a lack of frailty screening instruments specific to Chinese older adults. We developed a simple frailty screening questionnaire (FSQ) that could predict long-term mortality. We aimed to explore the reliability and construct validity of this new measurement tool. Methods: A total of 205 individuals aged 65 years or older were recruited in this study. The FSQ and frailty phenotype were assessed. The FSQ included self-reported slowness, weakness, weight loss, inactivity, and exhaustion. A subgroup of 109 participants completed the FSQ a second time 2 weeks later for test–retest reliability. Frailty phenotype included slowness, exhaustion, weight loss, weakness, and inactivity. Results: The intraclass correlation coefficient for the FSQ, slowness, weakness, weight loss, inactivity and exhaustion were 0.937, 0.938, 0.934, 0.921, 0.826, and 0.832, respectively. Using a cut-off of 3, the sensitivity, specificity, and area under the curve of the receiver operating characteristic were 52.6%, 93.5%, and 0.883 ( p < 0.001), respectively. The kappa coefficient between the FSQ and frailty phenotype was 0.431 ( p < 0.001). FSQ score was negatively correlated with walking speed and grip strength, and positively correlated with age. Frailty defined by the FSQ was associated with older age, chronic diseases, and worse physical function. Conclusions: The FSQ is a potentially useful, reliable, and valid instrument in screening frailty in older adults, and can be recommended to identify frailty in clinical settings.

Download Full-text