View clinical trials related to Predictive Model.
Filter by:The purpose of this 32 week study is to use an innovative experimental design known as SMART (Sequential Multiple Assignment Randomized Trial), which will allow us to determine the best way to sequence the delivery of teleexercise (referred to as an adaptive intervention), combined with predictive analytics on participant adherence in a stepped program of physical activity interventions. All 257 participants will have access to a library of recorded video exercise content, and a weekly wellness article. Some participants will receive health coaching calls (1st randomization). Analytic data will be used to determine which participants are responding or not responding to the intervention. Participants not responding after 4 weeks will receive either live one on one or group exercise training (2nd randomization). After 8 weeks, the participant will receive only pre recorded exercise content and articles for another 8 weeks. After final surveys, participants will have open access to the website for another 16 weeks where we will passively observe their fitbit and website data. The study outcomes are: The effectiveness of the adaptive interventions Exploring mediating and moderating variables Sensitivity analysis of the predictive analytics
Anxiety and depression in later life are highly prevalent, often appear as comorbid disorders, and have many adverse consequences for both the individual and society. Given the disease burden, the large influx of new cases, and the economic costs, efforts should be made to prevent the onset of anxiety and depression in later life. Preventive interventions are likely to become more cost effective when they are targeted at elderly who have been exposed to risk factors known to be predictive of the onset of anxiety and/or depression. As the population aging is speeding up, senile diseases have become a significant and severe public health problem, influencing national health. More than 20 million elderly patients undergo surgery each year in China, accounting for a quarter of the population who undergo surgery. Therefore, it is necessary to construct a predictive model of anxiety and depression in perioperative elderly hospitalized patients
Cognitive impairment is one of the core early signs of dementia, and it is also a key stage for community-based dementia prevention. Accurate and convenient prediction of cognitive impairment can help the community to identify and manage the high-risk population of dementia. Previous studies had developed several dementia predicting models, but such models may be not suitable for cognitive impairment prediction. Based on the national representative follow-up data of Chinese Longitudinal Healthy Longevity Survey (CLHLS), this project aims to develop and validate a brief cognitive impairment prediction algorithm among the community-dwelling elderly, using machine learning methods (such as Logistic regression, Naïve Bayes model, Extreme Gradient Boosting Tree and so on). Finally, based on the constructed model, an easy-to-use online intelligent assessment tool for predicting cognitive impairment risk will be developed. The general practitioners, social workers and the elderly would be invited to use the tool and we will revise the tool according to their suggestions and comments. This project is expected to provide scientific basis and technical support for community-based dementia prevention, and will also be useful for the elderly to easily understand their cognitive health.
Purpose: The purpose of this study is to create prediction models for when major complications occur after elective colectomy surgery. Justification: After surgery, patients can have multiple complications. Accurate risk prediction after surgery is important for determining an appropriate level of monitoring and facilitating patient recovery at home. Objectives: Investigators aim to develop and internally validate prediction models to predict time-to-complication for each individual major medical complications (pneumonia, myocardial infarction (MI) (i.e. heart attacks), cerebral vascular event (CVA) (i.e. stroke), venous thromboembolism (VTE) (i.e. clots), acute renal failure (ARF) (i.e. kidney failure), and sepsis (i.e. severe infections)) or adverse outcomes (mortality, readmission) within 30-days after elective colectomy. Data analysis: Investigators will be analyzing a data set provided by the National Surgical Quality Improvement Program (NSQIP). Descriptive statistics will be performed. Cox proportional hazard and machine learning models will be created for each complication and outcome outlined in "Objectives". The performances of the models will be assessed and compared to each other.
The study intends to develop a predictive model of spontaneous onset of labor between 39 and 41 weeks of pregnancy in women carrying singletons and without indication of delivery before this date. The main hypothesis is that a combination of clinical, ultrasonographic, biochemical and/or biophysical variables will allow to differentiate women who will spontaneously trigger their labors from those who will require an induction by the term of their pregnancies. A tool of this kind should aid in the individualization of the management of the final weeks of pregnancy and, at the light of recent evidence, provide support to the decision-making processes.
In order to obtain the risk level of patients with intestinal insufficiency through simple indicators, many foreign scholars have studied and developed their own prediction models. However, the current guideline indicates that there is insufficient evidence to recommend the use of a specialized predictive model for clinical practice.There are few studies on external validation of existing prediction models.
The rate of adequate bowel preparation is one of important quality indicators of colonoscopy. Inadequate bowel preparation negatively affects the outcomes of colonoscopy. If patients with inadequate bowel preparation were identified before the procedure, enhanced strategy could be offered to achieve better bowel cleaning. Currently, there were three predicting models of inadequate bowel preparation established based on patient-related factors. So far, none of predictive models have been tested in other than their validation cohort populations, and no study has attempted to apply a different regimen to patients presenting with risk factors for inadequate colon cleanliness. In previous studies, we established a prediction model based on procedure-related factors, which has better accuracy and can better predict the quality of bowel preparation. The aim of this study is to compare the quality of bowel preparation by using a predictive model based on procedure-related factors versus the criterion group in unsedation patients
The rate of adequate bowel preparation is one of important quality indicators of colonoscopy. Inadeqaute bowel preparation negatively affects the outcomes of colonoscopy. If patients with inadequate bowel preparation were identified before the procedure, enhanced strategy could be offerred to achieve better bowel cleasing. Currently, there were three predicting models of inadequate bowel preparation eatablished based on patient-related factors. It remains unclear which model perfroms better in predicting bowel preparation quality. Futhermore, althought those predicting models only composing of patients-related factors are useful for identifing high-risk patients, the preparation-related factors may also be valuable for prediciting inadeqaute bowel preparation before the procedure of colonoscopy. This study aimed: 1) to compare the values of three availlable models (based on patient-related factors) in predicting inadeqaute bowel preparation in a prospective, multicentered cohort of patients undergoing colonoscopy; 2) to investigate whether a new model based on preparation-related or a combined model based on patient-related and preparation-related factors is comparable to previous models based on patient-related factors.
The purpose of this study was to investigate the risk factors for mortality of sepsis and to create mathematical models to predict the survival rate based on electronic health records that extracted from hospital information system. More than 1000 records should be collected and used to data analysis. Univariate and multivariable logistic regression model were applied to risk factors analysis for the outcome, and machine learn algorithms were employed to generate predictive models for the outcome.