SEARCH · PubMed Health

Results for “Regression Analysis”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 181 records · Page 10Linked to original sources

PubMed1996

A survival regression analysis of prognostic factors in colorectal cancer.

BACKGROUND: Many prognostic factors of colorectal cancer are known but their actual clinical validity is still uncertain. The aim of the present study was to verify, on the basis of our experience, the prognostic validy of variables for survival by using survival regression analysis. METHODS: From January 1978 to December 1986 the prognostic factors for 192 patients were analysed. These patients had undergone surgical resection for colorectal cancer. The follow up was completed in every patient by the end of December 1992, with a median follow up of 10 years (range 6-14 years). The prognostic factors considered in the statistical analysis were age, sex, size of tumour, site, grade, direct spread, node involvement and stage (according to Astler-Coller and pTNM). RESULTS: Of the prognostic factors, sex was the only one not to show any prognostic significance. In the survival regression analysis we have used an accelerated failure time model (equivalent to the Cox proportional hazard model); age, grade and stage were significant covariables. CONCLUSIONS: Although clinical pathological staging (pTNM) appears as a pre-eminent prognostic factor, and as our analysis shows, it needs a further variable (grading), which has been shown to affect the prognosis in a significant way.

Adult↗

PubMed2006

Mechanism of action and determination of the best substrate for a thrombin-like enzyme from Lachesis muta muta venom by regression analysis of the kinetic parameters determined with peptidyl p-nitroanilide substrates.

The kinetic behavior of a thrombin-like enzyme from Lachesis muta muta venom has been studied with 13 tripeptidyl p-nitroanilide substrates. Eight substrates were unprotected at the N terminus and were used for the regression analysis of the experimentally determined kinetic parameters 1/Km, kcat and kcat/Km. The individual contribution of each amino acid side chain to the kinetic parameters was calculated. The amino acid sequence of the ideal substrate (D-Pro-Leu-Arg-pNA) was determined from a regression analysis for each kinetic parameter. This result was confirmed experimentally. The structural analysis of the tripeptides showed that the binding to the S3 sub-site had a small effect on Km. The binding of L-Leu to the S2 sub-site increased kcat without changing the value of Km. The analysis of the kinetic parameters revealed that, in the binding of L-Leu to the S2 sub-site, the enzyme bound the transition state configuration of the substrate/product transformation more tightly than that of the substrate.

Anilides↗

PubMed1998

Comparative study of different weighting methods in non-linear regression analysis: implications in the parametrization of carebastine after intravenous administration in healthy volunteers.

The influence of different weighting methods in non-linear regression analysis was evaluated in the pharmacokinetics of carebastine after a single intravenous dose of 10 mg in 8 healthy volunteers. Plasma concentrations were measured by HPLC using an on-line solid-phase extraction method and automated injection. The analytical method was fully validated and the function of the analytical error subsequently determined. The parametric approach was performed using different weighting methods, including the homoscedastic method (W = 1) and heteroscedastic methods using weights of 1/C, 1/C2, and the inverse of the concentration variance calculated through the analytical error function (1/V), and the results were statistically evaluated according to the normal distribution. Statistically significant differences were observed in the representative parameters of the disposition kinetics of carebastine. The use of a multiple comparison test for statistical analysis of all differences among group means indicated that differences were generated between the homoscedastic method (W = 1) and the heteroscedastic methods (1/C, 1/C2, and 1/V). The results obtained in the present study confirmed the utility of the analytical error function as a weighting method in non-linear regression analysis and reinforced the importance of the correct choice of weights to avoid the estimation of imprecise or erroneous pharmacokinetic parameters.

Adult↗

PubMed2004

[Study on the distribution of serum homocysteine and on multi-stepwise regression analysis of the associated factors in the population of community areas in Wuhan].

OBJECTIVE: To study the serum homocysteine (Hcy) distribution and characteristics in different sex and age groups in the community residents in Wuhan, and to analyse its associated factors with multi-stepwise regression analysis. METHODS: The population under study was from three community areas in Wuhan. Demographic distribution and the correlation with other risk factors of serum Hcy were analyzed statistically. RESULTS: (1) Geometric mean of serum Hcy was 14.43 micromol/L in males and 10.89 micromol/L in females with P <0.001. (2) Hcy of per age level in males was also higher (P <0.001). (3) The prevalence rate of hyperhomocysteinemia was 23.94% in the general population in Wuhan. The prevalence rate of hyperhomocysteinemia in males was 2.62 times higher than in females. (4) Multi-stepwise regression analysis showed that Hcy had different affecting factors in males and females. The affecting factors of Hcy in males were daily cigarettes smoking, urine micro-albumin (UMALB) and times of exercise per week. The affecting factors of Hcy in females were duration of exercise each time, weight, triglyceride (TG), high-density lipoprotein (HDL), urine micro-albumin (UMALB) and age. CONCLUSIONS: (1) Hcy at the population level was significantly different by sex and age. (2) Population living in the community in Wuhan had a higher serum level and prevalence rate of Hcy comparing to some other cities in China and even in developed countries. (3) The important affecting factors of Hcy in population also showed sex difference, unlike the reports from other countries or other areas in China. Serum Hcy seemed to be affected by environmental and other factors.

Adolescent↗

PubMed1997

Regression analysis of discrete time survival data under heterogeneity.

This paper concerns the regression analysis of discrete time survival data for heterogeneous populations by means of frailty models. We express the survival time for each individual as a sequence of binary variables that indicate if the individual survived at each time point. The main result is that the likelihood for these indicators can be factored into contributions that involve the conditional survival probabilities integrated over the frailty distribution of the risk set (population-averaged). We then model these population-averaged conditional probabilities as a function of covariates. The result justifies the practice of treating the failure indicators as independent Bernoulli trials and fitting binary regression models for the conditional failure probabilities at each time point. However, we must interpret the regression coefficients as population-averaged rather than subject-specific parameters. We apply the method to the Framingham Heart Study on risk factors for cardiovascular disease.

Algorithms↗

PubMed2002

Use of the regression analysis technique to determine the true phosphorus digestibility and the endogenous phosphorus output associated with corn in growing pigs.

The objectives of this study were to determine true phosphorus (P) digestibility and the endogenous P output associated with corn for growing pigs using the regression analysis technique. Four barrows, average initial body weight of 25 kg, were fitted with a T-cannula and fed four diets according to a 4 x 4 Latin square design. Four cornstarch-based diets, containing four levels of P at 0.7, 1.5, 2.2 and 2.8 g/kg dry matter intake (DMI), were formulated from corn. Each experimental period comprised 8 d with 4-d adaptation and 4-d collection of ileal digesta and fecal samples. The apparent ileal and fecal P digestibility values in corn were affected (P < 0.05) by P contents in the assay diets. The apparent ileal and fecal P digestibility values increased from -51.0 to 33.3% and from -41.4 to 39.1%, respectively, as P content increased from 0.7 to 2.8 g/kg DMI. Linear relationships (P < 0.05), expressed as g/kg DMI, between the apparent ileal and fecal digestible P and the total intake of dietary P, suggested that true P digestibility and the endogenous P outputs associated with corn can be determined by the regression analysis technique. There were no differences (P > 0.05) in true P digestibility values (54.0 +/- 6.5 vs. 59.8 +/- 8.5%) and the endogenous P outputs (0.693 +/- 0.128 vs. 0.670 +/- 0.160 g/kg DMI) between the ileal and the fecal determinations. The endogenous fecal P output represented 12.3% and 25.8% of the daily total and available P requirements in growing pigs recommended by the National Research Council in 1998. The present literature data of apparent digestibility and availability underestimate the true digestive utilization of P in corn for growing pigs by approximately 35%. Current diet formulation on the bases of total, apparent P digestibility and availability values in corn inevitably leads to P overfeeding and excessive P excretion in pigs.

Animals↗

PubMed2002

A three-dimensional model for the tooth-loss patterns by multi-plane regression analysis.

The construction of dental condition models is one of the useful methods for analysis of epidemiological surveys. The purpose of this investigation was to make a simple model with clear turning points for the longitudinal tooth-loss patterns of Japanese adults by means of multi-plane regression analysis. Since 1957, the Japanese Ministry of Health and Welfare has carried out national surveys of dental conditions every six years. The data of present tooth numbers by age (24-79 years) and sex from these surveys were used for this study. When there are turning points between two variables, intersecting straight lines regression is a valid means. However, a new method was developed so that the data of this study had three variables. The new three-dimensional model by multi-plane regression analysis seemed to fit tooth-loss patterns of Japanese adults within three phases. Younger subjects are represented in the first phase followed by the third phase of elders, where tooth loss was rather slow. However, in the second phase, middle-aged subjects, people lost their teeth rapidly. Thus, prolongation of the first phase could be an important factor to improve overall dental health.

Adult↗

PubMedDate not supplied

A method for cooperative or noncooperative binding studies using nonlinear regression analysis on a microcomputer.

Ligand-macromolecule interactions are studied by nonlinear regression analysis performed using a microcomputer (SYMAG-Micromachine 3000/Z). The basic phenomenon is described by the Clark equation B = (Formula: see text), where B and F represent bound and free ligand, respectively, Ni the total concentration of binding sites, and Ki their corresponding affinity constant. The programs of calculation have been extended also for Hill and for Adair equations using the Gauss algorithm described by E. E. Beaulieu and J. P. Raynaud [Eur. J. Biochem. 13, 293 (1970)]. A statistical test of F type is introduced to test the quality of the fit and compare the representations of the phenomenon using the different equations.

Animals↗

PubMed1986

Evaluation of a Bayesian regression-analysis computer program for predicting phenytoin concentration.

A microcomputer program using Bayesian regression analysis to predict serum phenytoin concentrations was evaluated. Phenytoin concentration-time data from nine healthy male volunteers and one male patient were obtained from published studies. For two different dosage regimens that each subject received, the last available predose concentration on the sixth day of the regimen was predicted using observed predose concentrations on both the morning of the third day and on each of the first three days of phenytoin administration. In nine subjects who received at least 10 days of phenytoin therapy, observed concentrations after more than 10 days of therapy were predicted using both one and three observed serum concentrations. Also, in six subjects, the observed predose concentrations for the first three days of an initial phenytoin regimen were used to predict the last predose concentration observed during each subject's second regimen. Predictive performance of the program was evaluated using mean error (m.e.) as a measure of bias, mean absolute error (m.a.e.) as a measure of precision, and root mean square error (r.m.s.e.) as a composite measure of bias and precision. The majority of the predicted serum concentrations were accurate. Predictions of serum concentrations after six days and after more than 10 days of phenytoin therapy were somewhat more accurate when three serum concentrations were used than when only one concentration was used. In the six subjects for whom concentrations from an initial regimen were used to predict those in a second regimen, the largest prediction error was 5 mg/L (m.e. 0.88, m.a.e. 1.9, and r.m.s.e. 2.4).(ABSTRACT TRUNCATED AT 250 WORDS)

Bayes Theorem↗

PubMed2006

Pointwise linear regression analysis for detection of visual field progression with absolute versus corrected threshold sensitivities.

PURPOSE: To compare the performance of point-wise linear regression analysis (PLR) with total deviation (TD) versus corrected or pattern deviation (PD) threshold sensitivities for detection of visual field progression. METHODS: Four hundred two eyes (402 patients) enrolled in the Advanced Glaucoma Intervention Study (AGIS) were selected. Criteria for progression according to PLR were a slope or=2 worsening points within the same Glaucoma Hemifield Test cluster. PLR was performed on TD and PD threshold sensitivities and compared to clinical evaluation. Eyes were classified into three groups based on mean deviation (MD): mild (MD>or=-6 dB), moderately advanced (-6 dB>MD>or=-12 dB), and advanced (MD<-12 dB) glaucoma. RESULTS: Visual field progression was observed in 154 (38%), 85 (21%), and 175 (44%) eyes, according to PLR(TD) and PLR(PD), and clinical evaluation. The pair-wise agreement between clinicians and PLR(TD) was significantly greater than that of clinicians and PLR(PD) (kappa=0.48, 95% CI: 0.44-0.52 vs. kappa=0.31, 95% CI: 0.27-0.35). Agreement between PLR(TD) and PLR(PD) decreased with increasing glaucoma severity: kappa (95% CI)=0.60 (0.52-0.67), 0.41 (0.35-0.47), and 0.33 (0.27-0.40) for mild, moderately advanced, and advanced glaucoma, respectively. CONCLUSIONS: Point-wise linear regression analysis on TD threshold sensitivities performed better than the same analysis on PD when clinical evaluation was used as a reference. Agreement between the two methods was less in moderately advanced and advanced glaucoma.

Adult↗

PubMed2002

Predictors of long-term success in catheter ablation of atrioventricular nodal reentrant tachycardia: a multivariate regression analysis.

BACKGROUND: To investigate the predictors of long-term success after catheter ablation of atrioventricular nodal reentrant tachycardia (AVNRT). METHODS: One-hundred and fourteen consecutive patients underwent slow pathway ablation using anteroseptal (n=24), midseptal (n=65) and posteroseptal approach (n=25). The correlation between ablation approaches, electrophysiological characteristics during and after ablation and the recurrence rate of AVNRT was analyzed by a multivariate regression analysis. RESULTS: During ablation, transient AV block in the anteroseptal, midseptal and posteroseptal approach occurred in 8.3, 4.6 and 0%, respectively (P<0.01). AVNRT recurred in seven patients after 5 years follow-up. Five recurrences (20.8%) were from anteroseptal approach group and two (3.1%) were from midseptal approach group. Multivariate regression analysis revealed that anteroseptal ablation approach and residual dual atrioventricular nodal pathway following apparently successful ablation were the predictors for recurrence of AVNRT (R=0.645, P<0.001). CONCLUSION: Anteroseptal approach of slow pathway ablation is associated with a higher incidence of transient AV block and AVNRT recurrence than other approaches. Residual dual atrioventricular nodal pathway after apparently successful ablation also carries a high risk of recurrence.

Adolescent↗

PubMed2000

Evaluating the frequency rate of hypomagnesemia in critically ill pediatric patients by using multiple regression analysis and a computer-based neural network.

OBJECTIVES: To determine the frequency rate of hypomagnesemia in patients admitted to the pediatric intensive care unit (ICU), and to identify subsets of patients (grouped by disease) who are at greatest risk of hypomagnesemia. We also compared a neural network model with multiple regression analysis to identify independent variables that would correlate with hypomagnesemia and to predict serum magnesium values in critically ill pediatric patients overall. DESIGN: Prospective, multicenter study. SETTING: Tertiary level medical/surgical pediatric ICUs. PATIENTS: Data were obtained at admission to the pediatric ICU for 463 patients from newborn to 18 yrs old who were admitted with a variety of surgical and nonsurgical conditions. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Total serum magnesium values were obtained within the first 24 hrs after admission in 463 pediatric patients admitted to four pediatric ICUs. Hypomagnesemia (defined as total serum magnesium <0.75 mmol/L) was found in 51 (11%) of the 463 patients, with the highest frequency rate (72%) and lowest mean serum magnesium level (0.66 +/- 0.17 mmol/L) in patients admitted after surgery with extensive osseous involvement (spinal fusion and craniofacial reconstruction). To determine whether hypomagnesemia could be predicted on the basis of other laboratory and clinical criteria, multiple regression analysis was performed and showed age, weight, and albumin levels weakly associated (r2 = .14, p < .001) with magnesium levels within the different diagnostic groups. These data were used to produce a mathematical model able to predict magnesium levels within 5% of the actual values in 23% of patients. A neural network was also created to compare its predictive capabilities to those of the multiple regression model. Once trained on a random subset (85%) of the patient population, the neural network was able to predict magnesium levels to within 5% of actual values for 88% of the remaining 15% of patients, comparing favorably with the predictions derived from the multiple regression model. CONCLUSIONS: Hypomagnesemia is not uncommon (11%) in critically ill pediatric patients, but is very common (72%) in patients admitted after surgery for spinal fusion or craniofacial reconstruction. Patients who undergo surgery for correction of scoliosis and craniofacial anomalies should have serum magnesium levels monitored closely after surgery. In other patients, a neural network or multiple regression model could help predict which patients would be at risk of developing hypomagnesemia, thereby focusing testing on patients likely to benefit from such testing.

Adolescent↗

PubMed2001

[Ultrasound fetal biometry and birth weight prognosis with the use of Aoki method and classical regression analysis].

We have attempted to estimate prognostic value of neonatal birth weight prediction with the use of fetal sonographic data: biparietal diameter, femur length, transverse and longitudinal abdominal diameter. Clinical and ultrasound data from 265 pregnant women with singleton gestation who delivered within 7 days from the last ultrasound scan were analyzed. For the fetal weight estimation we have calculated fetal abdominal area in the equation proposed by Aoki in 1990. Predicted and observed neonatal birth weight centiles and coefficients of determination were calculated with the use of regression analysis. Prediction method of neonatal weight with Aoki method explained 70% of variability in our population. In contrast, regression analysis on ultrasound data explained 99% of observer variability in neonatal birth weight. We conclude that own method of birth weight prediction could be of value, however it should be tested in a new prospectively examined population.

Adult↗

PubMed1994

Use of linear multiple regression analysis on dental survey data.

The use of multiple regression analysis in dental research has increased greatly in recent years. However, in most cases insufficient or no details are given to assess the validity of the results obtained through this statistical approach. Since multiple regression is the method of choice when dealing with three or more variables, provided the assumptions are fulfilled, this study reports the use of this technique on caries data from a random sample of schoolchildren from Andalucia, Spain. Linear multiple regression was used to analyse caries experience, expressed as dmft or DMFT, a quantitative dependent variable. The explanatory or independent variables considered for each age group were gender, type of area, social class and province, following the basic survey method recommended by the World Health Organization. Laborious data transformation procedures had to be employed to comply with necessary assumptions for this statistical technique. The difficulties encountered may be partly due to the fact that all the independent variables considered in the study were categorical or qualitative variables. In the process of ensuring valid results, several helpful statistics were identified. In addition, statistically significant influences on the dependent variable studied were easily detected, as well as the order of importance of the independent variables included in the model. However, the complex data transformation procedures required to validate assumptions in this study suggested that in some instances it could prove extremely difficult to use linear multiple regression on survey data.

Adolescent↗

PubMed2004

Diagnostic accuracy of atypical p-ANCA in autoimmune hepatitis using ROC- and multivariate regression analysis.

INTRODUCTION: Antineutrophil cytoplasmic antibodies (atypical p-ANCA) are detected at high prevalence in sera from patients with autoimmune hepatitis (AIH), but their diagnostic relevance for AIH has not been systematically evaluated so far. METHODS: Here, we studied sera from 357 patients with autoimmune (autoimmune hepatitis n=175, primary sclerosing cholangitis (PSC) n=35, primary biliary cirrhosis n=45), non-autoimmune chronic liver disease (alcoholic liver cirrhosis n=62; chronic hepatitis C virus infection (HCV) n=21) or healthy controls (n=19) for the presence of various non-organ specific autoantibodies. Atypical p-ANCA, antinuclear antibodies (ANA), antibodies against smooth muscles (SMA), antibodies against liver/kidney microsomes (anti-Lkm1) and antimitochondrial antibodies (AMA) were detected by indirect immunofluorescence microscopy, antibodies against the M2 antigen (anti-M2), antibodies against soluble liver antigen (anti-SLA/LP) and anti-Lkm1 by using enzyme linked immunosorbent assays. To define the diagnostic precision of the autoantibodies, results of autoantibody testing were analyzed by receiver operating characteristics (ROC) and forward conditional logistic regression analysis. RESULTS: Atypical p-ANCA were detected at high prevalence in sera from patients with AIH (81%) and PSC (94%). ROC- and logistic regression analysis revealed atypical p-ANCA and SMA, but not ANA as significant diagnostic seromarkers for AIH (atypical p-ANCA: AUC 0.754+/-0.026, odds ratio [OR] 3.4; SMA: 0.652+/-0.028, OR 4.1). Atypical p-ANCA also emerged as the only diagnostically relevant seromarker for PSC (AUC 0.690+/-0.04, OR 3.4). None of the tested antibodies yielded a significant diagnostic accuracy for patients with alcoholic liver cirrhosis, HCV or healthy controls. CONCLUSIONS: Atypical p-ANCA along with SMA represent a seromarker with high diagnostic accuracy for AIH and should be explicitly considered in a revised version of the diagnostic score for AIH.

Adolescent↗

PubMed2005

Rates and risk factors for subsequent tonsillectomy after prior adenoidectomy: a regression analysis.

OBJECTIVE: To determine the role of adenoidectomy without concurrent tonsillectomy in the treatment of upper airway obstruction, by determining rates and risk factors for subsequent tonsillectomy. DESIGN: Retrospective cohort study with nested case-control study. Data were evaluated using Kaplan-Meyer curves with Cox proportional hazards regression analysis, as well as contingency table and logistic regression analysis. SETTING: Tertiary care pediatric hospital with satellite clinics and surgical centers. PATIENTS: A total of 2462 patients aged 5 months to 18 years undergoing adenoidectomy without concurrent tonsillectomy. MAIN OUTCOME MEASURES: A 5-year database was searched for birth dates, dates of initial surgery, and dates of subsequent tonsillectomy (if performed) or latest follow-up. Cases (tonsillectomies) were then matched 1:1 by age with controls (no subsequent tonsillectomy). Medical charts were reviewed to identify potential predisposing factors, including sex, tonsil size, and adenoidectomy or tonsillectomy indication. RESULTS: Within 5.4 years, 108 patients underwent subsequent tonsillectomy. The relative risk of subsequent tonsillectomy decreases by 0.83 (95% confidence interval, 0.78-0.88) for each increasing year of age at adenoidectomy. The odds of undergoing a future tonsillectomy significantly increase with increasing tonsil size at the time of adenoidectomy. There was a trend toward doubling the risk of subsequent tonsillectomy when the adenoids were removed for upper airway obstruction (including obstructive sleep apnea) compared with other indications (P = .06). CONCLUSION: Knowledge of the rates and risk factors for subsequent tonsillectomy will allow more informed counseling of parents regarding whether tonsillectomy should be performed or deferred at the time of an indicated adenoidectomy.

Adenoidectomy↗

PubMed1992

Time to loco-regional recurrence after resection of Dukes' B and C colorectal cancer with or without adjuvant postoperative radiotherapy. A multivariate regression analysis.

Factors influencing time to loco-regional recurrence were identified in a multivariate regression analysis of data from a series of 468 radically operated patients (260 Dukes' B and 208 Dukes' C) with carcinoma of the rectum and the rectosigmoid. A number of clinical and pathological characteristics were prospectively collected and recorded. In addition, carcinoembryonic antigen (CEA) was measured within 1 week before surgery. The endpoint used was recurrence below the level of the umbilicus. All patients were followed for at least 5 years or until time of death. The two Dukes' stages B and C were analysed in two separate analyses using the Cox proportional hazards model. In patients with Dukes' B tumours, an increased risk of loco-regional recurrence was associated with perineural invasion, tumour located less than 10 cm from the anal verge, patient aged above 70 years, and small tumour size. In patients with Dukes' C tumours, the necessity to resect neighbour organs, perineural and venous invasion, tumour located less than 10 cm from the anal verge, and large tumour size were all associated with a poor loco-regional outcome. Postoperative radiotherapy was not a significant prognosticator for loco-regional control. An update of the 5-year results of the randomised study of post-operative radiotherapy (50 Gy with 2 Gy per fraction in an overall treatment time of 7 weeks) showed no survival benefit from adjuvant radiotherapy in either Dukes' category and no statistically significant improvement in the 5-year loco-regional control rate. However, when the comparison was restricted to a group of high-risk patients there was a statistically significant benefit from radiotherapy with respect to loco-regional control (P = 0.03) but not with respect to survival (P = 0.23). The potential advantage, in terms of the required number of patients, of restricting clinical trials of intensified loco-regional therapies to the high-risk patients, is illustrated.

Adult↗

PubMed1975

[Use of regression analysis in the quantitative evaluation of the activity of biological preparations].

Use of regression analysis in the assessment of the activity of biological preparations under experimental conditions permitted not only to assess the quantitative effect (ED50) more strictly, but also to find other parameters of importance for the results of comparison, for example with the standard, i.e. in standardization. To these belong regression coefficient, parallelism of regressions, and the relative potency. By the presence of a parallelism one can judge the similarity between the activity mechanism of the active principle of the preparations being compared. Relative potency characterizes the activity of the preparation in the relative values in comparison with the standard, with a statistical evaluation of this value with the aid of the confidence interval. The authors suggest a program for Mir-2 computer facilitating the calculations in using the analystical method which is more objective than the graphic method of assessment of the linear dosage-response curve.

Animals↗