Physiological age prediction model based on physical examination indicators and its application

Abstract

This invention provides a physiological age prediction model that predicts the physiological age of a subject by detecting physical examination indicators. These indicators include measurements of blood cells, glucose metabolism, tumor markers, urine components, anthropometric measurements, lipids, blood components, and tissue function. Preferably, these indicators include one or more selected from the following: glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and go, light reaction time, fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid, lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP). More preferably, all of the aforementioned physical examination indicators are included. This invention also provides a physiological age prediction device applying the above-mentioned physiological age prediction model and a computer storage medium.

Description

Technical Field

[0001] This invention belongs to the field of medicine and relates to predictive models and their applications for assessing an individual's physiological age and health status, predicting health risks, and evaluating the effectiveness of anti-aging and revitalizing therapies. Background Technology

[0002] Rising global life expectancy is leading to population growth and an increasing proportion of elderly people in various countries. It is projected that the population aged 60 and over will reach 1.4 billion by 2030 and 2.1 billion by 2050. China, as one of the fastest-aging countries, had 254 million people aged 60 and over in 2019, and this number is projected to increase to 402 million by 2040, accounting for approximately 28% of the total population. The elderly exhibit a wide range of characteristics; some 80-year-olds have physical and cognitive abilities comparable to many 30-year-olds, while others experience significant decline earlier. Therefore, more accurate assessment of aging status and identification of factors influencing healthy aging are crucial for addressing the challenges posed by rapid population aging.

[0003] Aging is a complex process characterized by molecular and cellular changes, tissue alterations, and a decline in organ function. This process leads to the degeneration of physiological functions, a reduction in daily activities, and an increase in age-related diseases and mortality. Accurately measuring aging status is crucial for understanding and potentially intervening in the aging process. However, individuals with the same physiological age (CA) can exhibit significant differences in their rate of aging. In recent years, biological age (BA) has been widely adopted as a means of quantifying the aging process, reflecting the actual degree of aging in different individuals. Measuring BA not only reveals the effects that accelerate aging but also helps identify modifiable factors that influence this process. The “aging clock” provides a proactive and preventative approach to health assessment as a tool for evaluating BA, assessing anti-aging interventions, and predicting disease risk.

[0004] To date, there is no gold standard for measuring biological age (BA). Certain biomarkers, such as genomic instability, telomere length, and DNA methylation (DNAm), have shown significant potential in assessing aging. In 2013, Steve Horvath and colleagues discovered a strong correlation between physiological age and genome-wide DNA methylation levels across various human tissues and cell types. They identified over 350 CpG methylation sites that could predict DNA methylation age in most cells and tissues, thus developing the first aging clock (non-patent literature 1). Since then, various aging clocks have emerged, including blood assays based on proteomics, multi-omics, and immune markers. While these methods are promising, the high cost of detecting DNA methylation and multi-omics markers in blood limits their use in clinical practice and validation in large longitudinal cohorts. Similarly, non-invasive aging clocks based on the microbiome, facial features, and urinary metabolites also face challenges related to cost and susceptibility to external factors such as diet and lifestyle. Therefore, there is a need to develop more stable, convenient, and cost-effective aging clocks for implementation in broader clinical and research settings.

[0005] In recent years, many studies have focused on developing BA models using multidimensional clinical indicators. These models, such as the frailty phenotype and frailty index, have advantages due to their lower cost, greater availability, and direct correlation with health outcomes. For example, one study used principal component analysis (PCA) to develop a clinical aging clock (PCAge) based on 165 clinical parameters, which found that calorie restriction significantly reduced biological age (Non-Patent Literature 2). Another effective BA measurement method is the Klemera and Doubal method (KDM), which outperforms traditional models in predicting disease and mortality risk (Non-Patent Literature 3, Non-Patent Literature 4). Using the KDM method, researchers constructed an aging clock for 12,377 middle-aged and older adults (mean age 57 years) from the Chinese Kadoorie Biobank (CKB), using 16 physical signs and 9 biochemical markers, which effectively captured differences in cardiovascular health and predicted the risk of all-cause mortality over ten years (Non-Patent Literature 5).

[0006] However, these traditional modeling algorithms have certain limitations, including the inability to filter and reduce the raw metrics used in the model. They often fail to identify new potential predictors from a large number of candidate metrics and require the inclusion of physiological age (CA) in the modeling process, which can be controversial when measuring biological age (BA). In contrast, machine learning (ML) algorithms do not require prior mechanistic knowledge or assumptions about the correlation between metrics. They can automatically identify relationships from large datasets, simplifying the inclusion of metrics. Recently, BA models developed using ML algorithms have shown strong effectiveness without including CA and have identified many new predictors. For example, a study involving 59,316 healthy participants (mean age 57 years) used the LightGBM algorithm to identify key predictors of biological age, linking age gap with 70 common health outcomes and mortality rates, and identifying 34 modifiable factors and 9 genomic risk loci associated with age gap (Non-Patent Literature 6).

[0007] Despite some progress in existing research, most basic ML studies remain focused on Western populations with narrow age coverage, and studies covering all age groups in the Chinese adult population are limited. Furthermore, few studies compare differences in BA (biomarkers of age) and modifiable factors influencing aging among different population groups (e.g., sex, urban / rural residency, and regional differences). In addition, most studies focus on identifying associations between accelerated aging and clinical biomarkers, leaving a gap in establishing reference intervals for all age groups in adults.

[0008] Existing technical documents

[0009] Non-patent literature

[0010] Non-patent literature 1: Horvath S. DNA methylation age of human tissues and cell types. Genome Biol 2013; 14(10):R115.

[0011] Non-patent literature 2: Fong S, Pabis K, Latumalea D, et al. Principal component-based clinical aging clocks identify signatures of healthy aging and targets for clinical intervention. Nat Aging 2024; 4(8): 1137-52.

[0012] Non-patent literature 3: Li Z, Zhang W, Duan Y, et al. Biological age models based on a healthy Han Chinese population. Arch Gerontol Geriatr 2023; 107:104905.

[0013] Non-patent document 4: Levine ME. Modeling the rate of senescence: can estimated biological age predict mortality more accurately than chronological age? JGerontol A Biol Sci Med Sci 2013;68(6):667-74.

[0014] Non-patent literature 5: Chen L, Zhang Y, Yu C, et al. Modeling biological age using blood biomarkers and physical measurements in Chinese adults. EBioMedicine 2023; 89:104458.

[0015] Non-patent document 6: Liu WS, You J, Ge YJ, et al. Association of biological age with health outcomes and its modifiable factors. Aging Cell 2023;22(12):e13995. Summary of the Invention

[0016] The technical problem to be solved by the present invention

[0017] As mentioned above, it is essential to develop an aging clock model that can accurately assess aging status. However, there is currently no physiological age prediction model for all age groups of the Chinese adult population available for clinical use. Therefore, the purpose of this invention is to provide an aging clock model for all age groups of the Chinese adult population that can accurately, conveniently, and effectively predict physiological age based on physical examination indicators.

[0018] Technical solutions for solving technical problems

[0019] To address the aforementioned technical challenges, the inventors conducted repeated and in-depth research. They selected volunteers spanning all age groups within the Chinese adult population for their study, collecting physical examination data for each participant, including anthropometry, blood components, blood cell counts, and urine composition—a total of 52 clinical measurement indicators. Additionally, 10 motor ability tests were conducted, such as grip strength, gait speed, and a timed stand-up-walk test. Furthermore, 92 lifestyle, self-reported health status, psychological, and social support indicators were collected through questionnaires. Based on this, using the ElasticNet model, the inventors constructed a clinical clock in the training set and predicted the biological age of each participant in the validation set. Of the 41 features used in training, 24 key predictive indicators were identified based on their importance in predicting biological age. Subsequently, the difference between predicted age and actual age—the "age difference"—was calculated, and the association between the "age difference" and clinical measurement indicators and motor ability was analyzed. Linear regression analysis identified 14 indicators negatively correlated with the "age difference" and 32 indicators positively correlated. Building upon this, the inventors conducted 1,000 random samplings on the training dataset, recording the frequency with which each feature was selected as a key predictor. The results showed that a simplified model using only the top 16 features could construct an aging clock. These features included renal function indicators such as eGFR, cystatin C, serum creatinine (SCr), and blood urea nitrogen (BUN); physical performance indicators such as timed up and go and light reaction time; and other indicators such as fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid, lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP).

[0020] These results suggest that the above physical examination indicators can accurately, conveniently and effectively predict physiological age. The aging clock model constructed based on these physical examination indicators can be used to accurately assess the aging status of the body, which is of great significance for health assessment.

[0021] This invention was completed based on the above research. Specifically, this invention includes the following:

[0022] [1] A physiological age prediction model, wherein the physiological age prediction model predicts the physiological age of the subject by detecting the physical examination indicators of the subject, the physical examination indicators being indicators that change significantly with age, including blood cells, glucose metabolism, tumor markers, urine components, anthropometric measurements, lipids, blood components and tissue function measurement indicators.

[0023] [2] As described in [1] above, the physiological age prediction model includes physical examination indicators selected from any one or more of the following: glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and go, light reaction time, fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid (Folate), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP).

[0024] [3] As described in [1] above, the physiological age prediction model includes the following physical examination indicators: glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and go, light reaction time, fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid (Folate), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP).

[0025] [4] As described in [1] above, the physiological age prediction model is composed of glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and go, light reaction time, fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid (Folate), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP).

[0026] [5] As described in [1] above, the physiological age prediction model, wherein the physical examination indicators include one or more selected from the following indicators:

[0027] Normal Gait Speed, Fastest Gait Speed, Timed Up and Go, Tandem Stance, Standing on Tiptoes, Sit to Stand Test, Light Reaction Time Time, white blood cell count (WBC), red blood cell count (RBC), platelet count (PLT), hemoglobin (Hb), fasting blood glucose (FBG), triglycerides (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), serum creatinine (SCr), blood urea nitrogen (BUN), uric acid (UA), alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bilirubin (TBIL), albumin (Alb), high-sensitivity C-reactive protein (hs-CRP), gamma-glutamyl transferase (GGT), total bile acids (TBA), lactate dehydrogenase (LDH), cholinesterase (CHE), alkaline phosphatase (ALP), cystatin C (Cystatin C) C) Homocysteine ​​(Hcy), superoxide dismutase (SOD), folic acid, free fatty acids (FFA), insulin (INS), insulin-like growth factor (IGF), β-hydroxybutyrate (D-3-HB), waist-to-hip ratio (WHR), body mass index (BMI), and glomerular filtration rate (eGFR).

[0028] [6] A physiological age prediction device, comprising:

[0029] At least one processor, and

[0030] Storage media that communicates with at least one processor

[0031] The storage medium stores a computer program executable by the at least one processor, which, when executed by the at least one processor, causes the at least one processor to perform the following steps:

[0032] (a) Accepting test values ​​of physical examination indicators from the examinee, wherein the physical examination indicators are indicators that change significantly with age, including blood cells, glucose metabolism, tumor markers, urine components, anthropometric measurements, lipids, blood components and tissue function measurements.

[0033] (b) Predict the physiological age of the subject based on the detection value of (a);

[0034] (c) Prepare a report that includes the predicted physiological age of the subject obtained in (b).

[0035] [7] The physiological age prediction device as described in [6] above, wherein the physical examination indicators include any one or more selected from glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and go, light reaction time, fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid (Folate), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP).

[0036] [8] The physiological age prediction device as described in [6] above, wherein the physical examination indicators include glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and go, light reaction time, fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid (Folate), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP).

[0037] [9] As described in [6] above, the physiological age prediction device, wherein the physical examination indicators consist of glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and go, light reaction time, fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid (Folate), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP).

[0038]

[10] The physiological age prediction device as described in [6] above, wherein the physical examination indicators include one or more selected from the following indicators:

[0039] Normal Gait Speed, Fastest Gait Speed, Timed Up and Go, Tandem Stance, Standing on Tiptoes, Sit to Stand Test, Light Reaction Time Time, white blood cell count (WBC), red blood cell count (RBC), platelet count (PLT), hemoglobin (Hb), fasting blood glucose (FBG), triglycerides (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), serum creatinine (SCr), blood urea nitrogen (BUN), uric acid (UA), alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bilirubin (TBIL), albumin (Alb), high-sensitivity C-reactive protein (hs-CRP), gamma-glutamyl transferase (GGT), total bile acids (TBA), lactate dehydrogenase (LDH), cholinesterase (CHE), alkaline phosphatase (ALP), cystatin C (Cystatin C) C) Homocysteine ​​(Hcy), superoxide dismutase (SOD), folic acid, free fatty acids (FFA), insulin (INS), insulin-like growth factor (IGF), β-hydroxybutyrate (D-3-HB), waist-to-hip ratio (WHR), body mass index (BMI), and glomerular filtration rate (eGFR).

[0040]

[11] A computer storage medium storing a computer program, characterized in that the computer program causes a computer connected to an external computer issuing a data input / output request and a storage device storing data to perform the following steps:

[0041] The physiological age of the examinee is predicted based on the examinee's physical examination indicators.

[0042] The physical examination indicators are those that change significantly with age, including blood cell count, glucose metabolism, tumor markers, urine composition, anthropometric measurements, lipids, blood components, and tissue function measurements.

[0043]

[12] The computer storage medium as described in

[11] above, wherein the physical examination indicators include any one or more selected from glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and go, light reaction time, fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid (Folate), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP).

[0044]

[13] The computer storage medium as described in

[11] above, wherein the physical examination indicators include glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and go, light reaction time, fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid (Folate), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP).

[0045]

[14] The computer storage medium as described in

[11] above, wherein the physical examination indicators consist of glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and go, light reaction time, fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid (Folate), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP).

[0046]

[15] The computer storage medium as described in

[11] above, wherein the physical examination indicators include one or more selected from the following indicators:

[0047] Normal Gait Speed, Fastest Gait Speed, Timed Up and Go, Tandem Stance, Standing on Tiptoes, Sit to Stand Test, Light Reaction Time Time, white blood cell count (WBC), red blood cell count (RBC), platelet count (PLT), hemoglobin (Hb), fasting blood glucose (FBG), triglycerides (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), serum creatinine (SCr), blood urea nitrogen (BUN), uric acid (UA), alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bilirubin (TBIL), albumin (Alb), high-sensitivity C-reactive protein (hs-CRP), gamma-glutamyl transferase (GGT), total bile acids (TBA), lactate dehydrogenase (LDH), cholinesterase (CHE), alkaline phosphatase (ALP), cystatin C (Cystatin C) c) Homocysteine ​​(Hcy), superoxide dismutase (SOD), folic acid, free fatty acids (FFA), insulin (INS), insulin-like growth factor (IGF), β-hydroxybutyrate (D-3-HB), waist-to-hip ratio (WHR), body mass index (BMI), and glomerular filtration rate (eGFR).

[0048] The effects of the invention

[0049] According to the present invention, an aging clock model based on physical examination indicators can be provided to accurately, conveniently and effectively predict physiological age. It has important application value for assessing individual aging status, predicting health risks, realizing self-health monitoring, early identification of the occurrence of age-related diseases, and exploring aging and anti-aging targets. Attached Figure Description

[0050] Figure 1 This refers to the project design of the present invention.

[0051] Figure 2 This section represents the analysis of age-related clinical characteristics and the construction of the clinical clock. (A) represents the classification of included clinical indicators; (B) represents the correlation between each category of indicators and age; (C) represents the intersection of age-related clinical indicators in the BJ and CS cohorts; (D) represents the changes of each category of age-related indicators with age; (E) represents the indicators with higher contribution from the clinical clock; (F) represents the correlation between the physiological age predicted by the clinical clock and the actual age in each dataset; (G) represents the correlation between each clinical indicator and the "age difference".

[0052] Figure 3 The following data represent the correlation between increased "age difference" and increased disease prevalence: (A) represents the division of the population into different aging groups based on "age difference," including accelerated aging, normal aging, and decelerated aging; (B) represents the difference in "age difference" between populations with different numbers of diseases; (C) represents the difference in prevalence among different aging groups across all participating populations; (D) represents the change in "age difference" caused by disease; (E) represents the age difference in the same "intrinsic abilities" between men and women in each group, based on the decelerated aging group; (F) represents the difference in prevalence among different aging groups in the population under 60 years of age; (G) represents the difference in prevalence among different aging groups in the population aged 60 years and older; (H) represents the change in "intrinsic abilities" among different aging groups in the populations under 60 years of age and aged 60 years and older; (IK) represents the differences in all-cause mortality, ADL, and IADL among different aging groups.

[0053] Figure 4 The following represent modifiable factors influencing aging. (A) represents modifiable factors that are significantly associated with the "age difference" among all participants; (B) represents modifiable factors that are significantly associated with the "age difference" among participants under 60 years of age and (C) participants 60 years of age or older.

[0054] Figure 5 The percentile curve represents age-related indicators, with the shaded area indicating the current reference range.

[0055] Figure 6 The simplified physiological age prediction model is represented as follows: (A) represents the frequency of each indicator identified as a high-contribution feature of the aging clock after 1000 iterations; (B) represents the correlation between the simplified clock predicted age and the actual age; (C) represents the different aging groups based on "age difference"; (D) represents the changes in "intrinsic abilities" in different aging groups for people under 60 years old and over 60 years old; (EG) represents the differences in ADL, IADL and all-cause mortality in different aging groups. Detailed Implementation

[0056] The present invention will now be described in detail.

[0057] To provide an aging clock model that accurately, conveniently, and effectively predicts biological age based on physical examination indicators across all age groups in the Chinese adult population, the inventors selected volunteers covering all age groups in the Chinese adult population for the study. Physical examination data was collected for each participant, including anthropometry, blood components, blood cell counts, and urine components, totaling 52 clinical measurement indicators. In addition, 10 motor ability tests were conducted, such as grip strength, gait speed, and timed stand-up walking tests. Furthermore, 92 lifestyle, self-reported health status, psychological, and social support indicators were collected through questionnaires. Based on this, using the ElasticNet model, the inventors constructed a clinical clock in the training set and predicted the biological age of each participant in the validation set. The 41 features used in the training are shown in Table 1 below, among which 24 key predictive indicators were identified based on their importance in predicting biological age. Subsequently, the difference between the predicted age and the actual age—the "age difference"—was calculated, and the correlation between the "age difference" and clinical measurement indicators and motor ability was analyzed. Linear regression analysis identified 14 indicators that were negatively correlated with "age difference" and 32 indicators that were positively correlated.

[0058] Considering the limited applicability of using 41 features to assess physiological age, the inventors performed 1,000 random samplings on the training dataset to construct a simplified aging clock and recorded the frequency with which each feature was selected as a key predictor. The results showed that a simplified model using only the top 16 features could construct the aging clock. These features included renal function indicators such as eGFR, cystatin C, serum creatinine (SCr), and blood urea nitrogen (BUN); physical performance indicators such as timed up and go and light reaction time; and other indicators such as fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid, lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP).

[0059] These results suggest that the above physical examination indicators can accurately, conveniently and effectively predict physiological age. The aging clock model that includes these physical examination indicators can be used to accurately assess the aging status of the body, which is of great significance for health assessment.

[0060] Building upon the aforementioned research, the applicant has successfully constructed an aging clock model based on physical examination indicators that can accurately, conveniently, and effectively predict physiological age. This model can be used for non-invasive assessment of individual aging status, prediction of health risks, self-health monitoring, early identification of age-related diseases, and discovery of aging and anti-aging targets, demonstrating significant application value.

[0061] [Table 1]

[0062] Normal Gait Speed normal walking speed Fastest Gait Speed Fastest walking speed Timed Up and Go Standing and walking time Tandem Stance Standing test Standing on Tiptoe Standing on tiptoe Sit to Stand Test Sitting test Light Reaction Time Light reaction time WBC White blood cell count RBC Red blood cell count PLT Platelet count Hb hemoglobin FBG fasting blood glucose TG Triglycerides TC Total cholesterol HDL-C High-density lipoprotein cholesterol LDL-C Low-density lipoprotein cholesterol SCr serum creatinine BUN urea nitrogen UA uric acid ALT alanine aminotransferase AST Aspartate aminotransferase TBIL Total bilirubin Alb albumin hs-CRP High-sensitivity C-reactive protein GGT Gamma-glutamyl transferase TBA Total bile acids LDH lactate dehydrogenase CHE Cholinesterase ALP alkaline phosphatase Cystatin C Cystatin C Hcy homocysteine SOD Superoxide dismutase Folate folic acid FFA Free fatty acids INS insulin IGF Insulin-like growth factor D-3-HB β-hydroxybutyric acid WHR Waist-to-hip ratio BMI Body Mass Index eGFR Glomerular filtration rate

[0063] That is, the present invention relates to a physiological age prediction model, which predicts the physiological age of the examinee by detecting the examinee's physical examination indicators, including blood cells, glucose metabolism, tumor markers, urine components, anthropometric measurements, lipids, blood components and tissue function measurement indicators. The aforementioned physical examination indicators are those that change significantly with age, preferably including one or more selected from the following: glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and go, light reaction time, fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid, lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP); and even more preferably including glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and go, and light reaction time. The following are the parameters: fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid (Folate), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP); more preferably, they consist only of glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and go, light reaction time, fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid (Folate), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP).

[0064] Furthermore, in the physiological age prediction model of the present invention, the aforementioned physical examination indicators may also include one or more selected from the following indicators: Normal Gait Speed, Fastest Gait Speed, Timed Up and Go, Tandem Stance, Standing on Tiptoe, Sit to Stand Test, and Light Reaction Time. Reaction Time, White Blood Cell Count (WBC), Red Blood Cell Count (RBC), Platelet Count (PLT), Hemoglobin (Hb), Fasting Blood Glucose (FBG), Triglycerides (TG), Total Cholesterol (TC), High-Density Lipoprotein Cholesterol (HDL-C), Low-Density Lipoprotein Cholesterol (LDL-C), Serum Creatinine (SCr), Blood Urea Nitrogen (BUN), Uric Acid (UA), Alanine Aminotransferase (ALT), Aspartate Aminotransferase (AST), Total Bilirubin (TBIL), Albumin (Alb), High-Sensitivity C-Reactive Protein (hs-CRP), Gamma-Glutamate Transferase (GGT), Total Bile Acids (TBA), Lactate Dehydrogenase (LDH), Cholinesterase (CHE), Alkaline Phosphatase (ALP), Cystatin C C) Homocysteine ​​(Hcy), superoxide dismutase (SOD), folic acid, free fatty acids (FFA), insulin (INS), insulin-like growth factor (IGF), β-hydroxybutyrate (D-3-HB), waist-to-hip ratio (WHR), body mass index (BMI), and glomerular filtration rate (eGFR).

[0065] The present invention also relates to a physiological age prediction device, comprising at least one processor and a storage medium communicatively connected to the at least one processor, wherein the storage medium stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to cause the at least one processor to perform the following steps:

[0066] (a) Accept the test values ​​of physical examination indicators from the examinee, wherein the above physical examination indicators are indicators that change significantly with age, including blood cells, glucose metabolism, tumor markers, urine components, anthropometric measurements, lipids, blood components and tissue function measurements.

[0067] (b) Predict the physiological age of the subjects based on the test values ​​in (a) above;

[0068] (c) Prepare a report that includes the predicted physiological age of the subject obtained in (b) above.

[0069] In the above (a), the physical examination indicators preferably include one or more selected from the following: glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and go, light reaction time, fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid, lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP); and more preferably include glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and go, and light reaction time. The study found that the fasting blood glucose (FBG), insulin-like growth factor (IGF), and superoxide dismutase (SOD) are all of these; more preferably, the study found that the fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid (Folate), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP) are all of these.

[0070] In (a) above, the physical examination indicators may also include one or more selected from the following indicators: Normal Gait Speed, Fastest Gait Speed, TimedUp and Go, Tandem Stance, Standing on Tiptoe, Sit to Stand Test, and Light Reaction Time. Time, white blood cell count (WBC), red blood cell count (RBC), platelet count (PLT), hemoglobin (Hb), fasting blood glucose (FBG), triglycerides (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), serum creatinine (SCr), blood urea nitrogen (BUN), uric acid (UA), alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bilirubin (TBIL), albumin (Alb), high-sensitivity C-reactive protein (hs-CRP), gamma-glutamyl transferase (GGT), total bile acids (TBA), lactate dehydrogenase (LDH), cholinesterase (CHE), alkaline phosphatase (ALP), cystatin C (Cystatin C) C) Homocysteine ​​(Hcy), superoxide dismutase (SOD), folic acid, free fatty acids (FFA), insulin (INS), insulin-like growth factor (IGF), β-hydroxybutyrate (D-3-HB), waist-to-hip ratio (WHR), body mass index (BMI), and glomerular filtration rate (eGFR).

[0071] The present invention also relates to a computer storage medium storing a computer program that causes a computer connected to an external computer issuing data input / output requests and a storage device storing the data to perform the following steps to predict the physiological age of a subject based on the aging clock model of the present invention. The steps performed by the computer include predicting the physiological age of the subject based on the subject's physical examination indicators. These indicators preferably include one or more selected from the following: glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and walk time (Timed Up and Go), light reaction time, fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid, lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP); and more preferably include glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and walk time (Timed Up and Go), and light reaction time (Timed Up and Go). The following are the parameters: fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid (Folate), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP); more preferably, they consist only of glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and go, light reaction time, fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid (Folate), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP).

[0072] Furthermore, in the computer storage medium of the present invention, the aforementioned physical examination indicators may also include one or more selected from the following indicators: Normal Gait Speed, Fastest Gait Speed, Timed Up and Go, Tandem Stance, Standing on Tiptoe, Sit to Stand Test, and Light Reaction Time. Reaction Time, White Blood Cell Count (WBC), Red Blood Cell Count (RBC), Platelet Count (PLT), Hemoglobin (Hb), Fasting Blood Glucose (FBG), Triglycerides (TG), Total Cholesterol (TC), High-Density Lipoprotein Cholesterol (HDL-C), Low-Density Lipoprotein Cholesterol (LDL-C), Serum Creatinine (SCr), Blood Urea Nitrogen (BUN), Uric Acid (UA), Alanine Aminotransferase (ALT), Aspartate Aminotransferase (AST), Total Bilirubin (TBIL), Albumin (Alb), High-Sensitivity C-Reactive Protein (hs-CRP), Gamma-Glutamate Transferase (GGT), Total Bile Acids (TBA), Lactate Dehydrogenase (LDH), Cholinesterase (CHE), Alkaline Phosphatase (ALP), Cystatin C C) Homocysteine ​​(Hcy), superoxide dismutase (SOD), folic acid, free fatty acids (FFA), insulin (INS), insulin-like growth factor (IGF), β-hydroxybutyrate (D-3-HB), waist-to-hip ratio (WHR), body mass index (BMI), and glomerular filtration rate (eGFR).

[0073] Example

[0074] The present invention will now be described in further detail with reference to the embodiments. It should be understood that the specific embodiments described below are only for explaining the present invention and are not intended to limit the present invention. Furthermore, the experimental methods in the following embodiments, unless otherwise specified, are all performed according to conventional methods known in the art.

[0075] The project design and cohort characteristics of this study are as follows.

[0076] For each participant, the inventors collected physical examination data, including anthropometry, blood components, blood cell counts, and urine components, totaling 52 clinical measurement indicators. Figure 1 In addition, 10 motor ability tests were conducted, such as grip strength, walking speed, and timed stand-up walking tests. Information on 92 lifestyle, self-reported health status, psychological and social support indicators was also collected through questionnaires.

[0077] Based on this, the inventors conducted the following research.

[0078] Example 1: Analysis of age-related clinical characteristics and construction of a clinical clock

[0079] The inventors divided the clinical measurement indicators into eight groups: blood cells, glucose metabolism, tumor markers, urine components, anthropometric measurements, lipids, blood components, and tissue function. Figure 2 A). The correlation between age and these measurements and 10 motor ability tests was calculated in both cohorts. Further analysis of age-related clinical characteristics revealed that motor ability indicators showed the strongest correlation with age, ranking highest overall. Figure 2 B). For example, the light reaction time test (hand), the timed stand-up walking test, and the five-repetition sit-up test increase with age ( Figure 2 C). Conversely, grip strength, walking speed, tiptoeing time, and eye-open, foot-to-toe standing time decrease with age. Figure 2 C). Furthermore, blood cell markers such as red blood cells, white blood cells, and platelets, as well as routine blood indices such as IGF, SOD, and hemoglobin, decrease with age. Figure 2 C). Anthropometric parameters (systolic blood pressure and waist-to-hip ratio) and glucose metabolism-related parameters (fasting blood glucose, glycated hemoglobin, and hydroxybutyrate) increase with age. Figure 2 D). Some lipid metabolism indicators, such as LDL-C and TC, show non-monotonic changes, decreasing with age and after age 60. Figure 2 D). Tissue function indicators related to liver, kidney, and heart function also showed significant age-related changes. The important renal function marker eGFR begins to decline around age 40. Figure 2 D). Albumin (ALB), a plasma protein mainly synthesized by the liver, decreases with age, indicating a decline in liver function. Figure 2 D). Lactate dehydrogenase (LDH) and high-sensitivity C-reactive protein (hs-CRP) are common biomarkers of tissue damage and inflammation, respectively. With increasing age, they indicate a decline in various tissue functions and the presence of chronic inflammation. Figure 2 D).

[0080] Although most age-related clinical indicators showed similar trends in both cohorts, some differences remained. For example, participants in the CS cohort, especially older adults, had lower physical fitness compared to participants in the BJ cohort. Figure 2 D). Furthermore, in the CS cohort, older volunteers had lower platelet and SOD levels, while having higher LDH and hs-CRP levels. Figure 2 D). Regarding lipid metabolism, the levels in the CS cohort were higher than those in the BJ cohort, while glucose metabolism was higher in the BJ cohort than in the CS cohort. Figure 2 D).

[0081] To determine whether clinical measurements could predict biological age, the inventors selected 41 indicators with the fewest missing values ​​from two cohorts to construct a clinical clock. All healthy individuals (n=645) from the CS cohort were divided into a training set (50%) and a validation set (50%). The remaining volunteers from the CS cohort (n=1,816) and the BJ cohort (n=1,884) served as the internal and external validation sets, respectively. Using the ElasticNet model, the inventors constructed a clinical clock in the training set and predicted the biological age of each participant in the validation set. Figure 2 F). Of the 41 features used in training, 24 key predictive metrics were identified based on their importance in predicting biological age. Figure 2 E). These key predictive indicators, such as cystatin C, fasting blood glucose (FBG), eGFR, and IGF, are also highly correlated with age. Figure 2 CD). In the validation set of healthy participants in the CS cohort, the mean absolute error (MAE) of the model was 5.45 ( Figure 2 F). Compared to the BJ cohort's validation set (MAE = 6.99), this model performs better on the CS cohort's validation set of unhealthy participants (MAE = 6.41). Figure 2 F).

[0082] Subsequently, the inventors calculated the difference between predicted age and actual age—the "age difference"—and analyzed its correlation with clinical measurement indicators and motor ability. Through linear regression analysis, they identified 14 indicators negatively correlated with the "age difference" and 32 indicators positively correlated. Figure 2 G). Fasting blood glucose (FBG) is positively correlated with the "age difference," with each 1 mmol / L increase in FBG increasing the "age difference" by 3.6 years. Each 0.1 mg / L increase in cystatin C increases the "age difference" by 1.5 years. Conversely, each 10-unit decrease in eGFR increases the "age difference" by 1.5 years. Each 100 ng / ml decrease in IGF increases the "age difference" by 4 years. HDL-C is negatively correlated with the "age difference," with each 1 mmol / L increase in HDL-C decreasing the age difference by 2.8 years.

[0083] Example 2: Correlation between increased "age difference" and increased disease prevalence

[0084] To explore the relationship between age difference and disease risk in two cohorts, the inventors divided the validation cohorts into three groups: an accelerated group (age difference > 7.6), a normal group (-7.6 >= age difference <= 7.6), and a slowed group (age difference < -7.6). Figure 3 A).

[0085] The study found that volunteers who reported a higher number of diseases had a greater "age difference" (p<0.01). Figure 3B). Among all individuals, compared with the slowing group, the accelerating group had a higher prevalence of hypertension (p<0.001), diabetes (p<0.001), arthritis (p<0.05), metabolic disorders (p<0.01), heart disease (p<0.001), and cataracts (p<0.01). Figure 3 C). Compared to disease-free individuals, the "age difference" of diseased individuals increased by an average of 1.3 years. Figure 3 D). Among the affected population, the "age difference" increased the most, averaging 6 years. Figure 3 D). Secondly, chronic nephritis increases the age difference by about 2 years. Figure 3 D). Heart disease and hypertension each increase the "age difference" by approximately 1.2 years, while metabolic disorders increase it by 0.9 years. Figure 3 D). In individuals under 60 years of age, the disease prevalence in the accelerated group was similar to the findings observed in the entire cohort. Figure 3 F). However, among individuals over 60 years of age, except for diabetes (p<0.001) and heart disease (p<0.05), a larger “age difference” did not lead to a higher prevalence of disease. Figure 3 G).

[0086] For older adults, intrinsic capabilities are more significantly affected by the age gap. Studies have found that intrinsic capabilities gradually decline as the age gap increases. Figure 3 H). Compared to individuals under 60, individuals over 60 experienced a more significant decline in intrinsic capabilities. Figure 3 H).

[0087] The inventors further analyzed the association between age difference and all-cause mortality, activities of daily living (ADLs), and instrumental activities of daily living (IADLs). Compared with the slowing group, the accelerating group showed a significant increase in all-cause mortality, with an increase of 3.7%. Figure 3 K). Furthermore, a positive correlation was found between "age difference" and ADL scores, with the accelerated group having significantly higher ADL scores than the slowed group (K). Figure 3 I).

[0088] Example 3: Adjustable factors affecting aging

[0089] The inventors collected 91 modifiable factors through a questionnaire survey, covering nine categories, including basic information, lifestyle, physical condition, medical and disease status, psychological state, economic status, social network and support, living environment, and cognitive appraisal. After adjusting for age and gender, modifiable factors related to the age gap were identified. Figure 4A). Social networks are the most significant factor influencing the age gap. For example, those who cannot obtain help from close relatives when facing major problems experience an average increase in the age gap of 3.8 years. Poor marital relationships also lead to an increase in the age gap, averaging 2.5 years. Lifestyle factors also significantly affect the age gap. Poor working conditions (2 years), smoking (2 years), primarily consuming sugary drinks (1.5 years), lack of participation in social organizations and activities (1.3 years), and engaging in heavy physical labor (0.8 years) all contribute to an increase in the age gap. Conversely, consuming protein (soybeans [-3.3 years], milk [-1.7 years]), fruit (-1.9 years), and engaging in physical exercise (-0.5 years) can reduce the age gap. Interestingly, the use of sleeping pills is associated with a significantly reduced age gap. Employment status also affects the age gap; both unemployment (1.6 years) and part-time work (2.2 years) lead to an increase in the age gap. Higher levels of education (-1.3 years) are associated with a smaller age gap. Furthermore, regardless of current marital status, individuals who had been married had a smaller "age gap" than those who had never been married. We found that individuals taking medication and who had been hospitalized in the past two years had a larger "age gap" than those taking less medication and who had not been hospitalized. Those with a positive health awareness had a smaller "age gap." Physical condition, mental state, and economic status also significantly affected the "age gap." For example, those with urinary incontinence (4 years), missing teeth (0.7 years), and hearing loss (0.65 years) had a larger "age gap." In addition, those without social insurance (0.9 years) and housing (1 year) also had a larger "age gap."

[0090] Age 60 is a critical juncture in aging, at which point molecular markers of aging become significantly imbalanced. The inventors analyzed age-related characteristics in individuals under 60 and those 60 and over. Figure 4 B represents modifiable factors that are significantly associated with the "age difference" in people under 60 years of age; Figure 4 C represents modifiable factors significantly associated with the "age gap" in individuals aged 60 and older. Among those under 60, those consuming more vegetables showed the most significant reduction in the "age gap." Higher education levels and marriage significantly reduced the age gap. Conversely, poor marital relationships, lack of family support and emotional assistance, unemployment or part-time work, and smoking significantly increased the "age gap." Fewer factors had a significant impact on individuals aged 60 and older compared to those under 60. For example, soy and fruit intake were the most important factors significantly reducing the "age gap" in this age group. Older adults living in buildings with elevators have a younger physiological age. Conversely, limited participation in social activities significantly increased the "age gap."

[0091] Example 4: The effects of age and sex distribution on the values ​​of age-related biomarkers

[0092] To determine reference ranges for age-related clinical indicators and to compare the "health curves" (P25-P75) of non-accelerated aging individuals with existing reference intervals, the inventors excluded accelerated aging samples and established reference ranges for the remaining indicators. Figure 5 ).

[0093] Indicators related to mobility fluctuate most significantly with age, showing a clear inflection point around age 60, after which their reference ranges change significantly. Insulin-like growth factor (IGF), superoxide dismutase (SOD), and estimated glomerular filtration rate (eGFR) lack clearly defined reference ranges and show a trend of continuously decreasing reference ranges with age. For individuals under 60 years of age, the hemoglobin (Hb) reference range is higher than existing standards, while in individuals over 80 years of age, the lower limit of red blood cell (RBC) count is lower than the existing reference range. Among indicators related to tissue function, with significant age-related changes, the upper limits of β2-microglobulin (β2-MG), alkaline phosphatase (ALP), high-sensitivity C-reactive protein (hs-CRP), and lactate dehydrogenase (LDH) in individuals over 80 years of age exceed existing reference ranges, while the upper limit of cystatin C in individuals over 60 years of age also exceeds existing standards. The ideal upper limit for lipid levels is slightly higher than existing standards. For individuals over 40 years of age, the upper limits for total cholesterol (TC) and low-density lipoprotein cholesterol (LDL-C), and for individuals over 70 years of age, the upper limits for free fatty acids (FFA) are higher than existing standards. Our results show that in 99% of individuals, TC is above 2.5 mmol / L and LDL-C is above 1 mmol / L. The ideal lower limits indicate that TC should be at least above 3.9 mmol / L and LDL-C should be above 2.1 mmol / L. For blood glucose indicators (fasting blood glucose [FBG], glycated hemoglobin [HbA1c]) in non-accelerated aging individuals, the ideal ranges for different ages are all within the existing reference ranges, with the upper limit for FBG remaining below 5.5 mmol / L, and even lower in individuals under 60 years of age (approximately 5.1-5.5 mmol / L). However, the study found a significant difference between the ideal range for blood pressure and existing standards, with the upper limit for individuals over 40 years of age beginning to exceed existing standards.

[0094] Given the gender differences in these indicators, the inventors further analyzed the ideal ranges for male and female participants of different ages. Grip strength showed significant gender differences, with men exhibiting higher average grip strength than women across all age groups. Younger men had higher levels of red blood cells (RBC) and platelets (PLT) than women, but these indicators declined more rapidly in men with age. Conversely, lipid-related indicators (LDL-C, HDL-C, TC) were higher in women. Men had higher levels of renal function indicators (β2-microglobulin [β2-MG], blood urea nitrogen [BUN], serum creatinine [SCr], uric acid [UA]), while liver function indicators (albumin [ALB], total bilirubin [TBIL], alanine aminotransferase [ALT]) fluctuated more with age in men than in women.

[0095] Example 5: Simplified physiological age prediction model

[0096] Considering that using 41 features to assess physiological age is not very applicable, the inventors performed 1,000 random samplings on the training dataset to construct a simplified aging clock and recorded the frequency with which each feature was selected as a key predictor.

[0097] The inventors sorted the features according to frequency and, using different numbers of features, progressively calculated the similarity between the predicted physiological age and the actual age. Figure 6 A). The results show that a simplified model using only the top 16 features is able to achieve a similarity of over 0.9 between physiological age and chronological age in both the CS and BJ cohorts. Figure 6 B). These characteristics include renal function indicators such as glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), and blood urea nitrogen (BUN); physical performance indicators such as timed up and go and light reaction time; and other indicators such as fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid, lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP). Figure 6 A).

[0098] Correlation analysis of the physiological age predicted by the simplified clock with that of the comprehensive clock showed that the simplified clock could effectively reproduce the prediction results of the comprehensive clock. Figure 6 B). Consistent with the results of the comprehensive model, the accelerated group over 60 years of age experienced a faster decline in intrinsic capacity, leading to a higher mortality rate. Similarly, the accelerated group had a higher rate of disability and higher ADL scores. Figure 6 CG).

[0099] The above research clearly demonstrates that the simplified clock constructed based on these 16 physical examination indicators can accurately predict physiological age, proving that these 16 physical examination indicators play an important role in assessing aging.

[0100] After reading the above statement about the present invention, those skilled in the art can make various modifications or changes to the present invention, and these equivalent forms also fall within the scope defined in the appended claims.

[0101] Industrial availability

[0102] The non-invasive aging clock model of this invention can accurately, conveniently and effectively predict physiological age based on physical examination indicators. It has important value for assessing individual aging status, predicting health risks, realizing self-health monitoring, early identification of the occurrence of age-related diseases, and exploring aging and anti-aging targets. It has high practicality in industry.

Claims

1. A physiological age prediction model, characterized in that, The physiological age prediction model predicts the physiological age of the examinee by detecting the examinee's physical examination indicators. The physical examination indicators are those that change significantly with age, including blood cell count, glucose metabolism, tumor markers, urine composition, anthropometric measurements, lipids, blood components, and tissue function measurements.

2. The physiological age prediction model as described in claim 1, characterized in that, The physical examination indicators include any one or more selected from the following: glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and go, light reaction time, fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid (Folate), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP).

3. The physiological age prediction model as described in claim 1, characterized in that, The physical examination indicators include glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and go, light reaction time, fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid (Folate), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP).

4. The physiological age prediction model as described in claim 1, characterized in that, The physical examination indicators include one or more selected from the following indicators: Normal Gait Speed, Fastest Gait Speed, Timed Up and Go, Tandem Stance, Standing on Tiptoes, Sit to Stand Test, Light Reaction Time Time, white blood cell count (WBC), red blood cell count (RBC), platelet count (PLT), hemoglobin (Hb), fasting blood glucose (FBG), triglycerides (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), serum creatinine (SCr), blood urea nitrogen (BUN), uric acid (UA), alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bilirubin (TBIL), albumin (Alb), high-sensitivity C-reactive protein (hs-CRP), gamma-glutamyl transferase (GGT), total bile acids (TBA), lactate dehydrogenase (LDH), cholinesterase (CHE), alkaline phosphatase (ALP), cystatin C (Cystatin C) C) Homocysteine ​​(Hcy), superoxide dismutase (SOD), folic acid, free fatty acids (FFA), insulin (INS), insulin-like growth factor (IGF), β-hydroxybutyrate (D-3-HB), waist-to-hip ratio (WHR), body mass index (BMI), and glomerular filtration rate (eGFR).

5. A physiological age prediction device, characterized in that, include: At least one processor, and Storage media that communicates with at least one processor The storage medium stores a computer program executable by the at least one processor, which, when executed by the at least one processor, causes the at least one processor to perform the following steps: (a) Accepting test values ​​of physical examination indicators from the examinee, wherein the physical examination indicators are indicators that change significantly with age, including blood cells, glucose metabolism, tumor markers, urine components, anthropometric measurements, lipids, blood components and tissue function measurements. (b) Predict the physiological age of the subject based on the detection value of (a); (c) Prepare a report that includes the predicted physiological age of the subject obtained in (b).

6. The physiological age prediction device as described in claim 5, characterized in that, The physical examination indicators include any one or more selected from the following: glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and go, light reaction time, fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid (Folate), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP).

7. The physiological age prediction device as described in claim 5, characterized in that, The physical examination indicators include glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and go, light reaction time, fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid (Folate), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP).

8. The physiological age prediction device as described in claim 5, characterized in that, The physical examination indicators include one or more selected from the following indicators: Normal Gait Speed, Fastest Gait Speed, Timed Up and Go, Tandem Stance, Standing on Tiptoes, Sit to Stand Test, Light Reaction Time Time, white blood cell count (WBC), red blood cell count (RBC), platelet count (PLT), hemoglobin (Hb), fasting blood glucose (FBG), triglycerides (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), serum creatinine (SCr), blood urea nitrogen (BUN), uric acid (UA), alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bilirubin (TBIL), albumin (Alb), high-sensitivity C-reactive protein (hs-CRP), gamma-glutamyl transferase (GGT), total bile acids (TBA), lactate dehydrogenase (LDH), cholinesterase (CHE), alkaline phosphatase (ALP), cystatin C (Cystatin C) C) Homocysteine ​​(Hcy), superoxide dismutase (SOD), folic acid, free fatty acids (FFA), insulin (INS), insulin-like growth factor (IGF), β-hydroxybutyrate (D-3-HB), waist-to-hip ratio (WHR), body mass index (BMI), and glomerular filtration rate (eGFR).

9. A computer storage medium storing a computer program, characterized in that the computer program causes a computer connected to an external computer issuing a data input / output request and a storage device storing the data to perform the following steps: The physiological age of the examinee is predicted based on the examinee's physical examination indicators. The physical examination indicators are those that change significantly with age, including blood cell count, glucose metabolism, tumor markers, urine composition, anthropometric measurements, lipids, blood components, and tissue function measurements.

10. The computer storage medium as claimed in claim 9, characterized in that, The physical examination indicators include any one or more selected from the following: glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and go, light reaction time, fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid (Folate), lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP).

11. The computer storage medium as claimed in claim 9, characterized in that, The physical examination indicators include glomerular filtration rate (eGFR), cystatin C, serum creatinine (SCr), blood urea nitrogen (BUN), timed up and go, light reaction time, fasting blood glucose (FBG), insulin-like growth factor (IGF), superoxide dismutase (SOD), insulin (INS), albumin (Alb), alkaline phosphatase (ALP), folic acid, lactate dehydrogenase (LDH), aspartate aminotransferase (AST), and high-sensitivity C-reactive protein (hs-CRP).

12. The computer storage medium as claimed in claim 9, characterized in that, The physical examination indicators include one or more selected from the following indicators: Normal Gait Speed, Fastest Gait Speed, Timed Up and Go, Tandem Stance, Standing on Tiptoes, Sit to Stand Test, Light Reaction Time Time, white blood cell count (WBC), red blood cell count (RBC), platelet count (PLT), hemoglobin (Hb), fasting blood glucose (FBG), triglycerides (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), serum creatinine (SCr), blood urea nitrogen (BUN), uric acid (UA), alanine aminotransferase (ALT), aspartate aminotransferase (AST), total bilirubin (TBIL), albumin (Alb), high-sensitivity C-reactive protein (hs-CRP), gamma-glutamyl transferase (GGT), total bile acids (TBA), lactate dehydrogenase (LDH), cholinesterase (CHE), alkaline phosphatase (ALP), cystatin C (Cystatin C) C) Homocysteine ​​(Hcy), superoxide dismutase (SOD), folic acid, free fatty acids (FFA), insulin (INS), insulin-like growth factor (IGF), β-hydroxybutyrate (D-3-HB), waist-to-hip ratio (WHR), body mass index (BMI), and glomerular filtration rate (eGFR).

Images