Please enable JavaScript.

Coggle requires JavaScript to display documents.

Data Structure & Statistical Procedures (Statistical Procedures …

- - - - Open Humans OpenAPS Data Commons
        
        User-generated Data' Properties
        
        ID
        
        Metadata
        
        Source
        
        Device
        
        Objects
        
        Types
        
        Structured Data
        
        Unstructured Data
        
        Subclasses
        
        Profil (IC/ISF/BR/ Targets)
        
        Category (Insulin profile/TT: category)
        
        Quantitative (Bolus info: insulin amount, carbs amount // TBR: absolute, relative, amount // Insulin Profile: Duration, Peak // Extended Bolus: amount // Temporary Target: Target)
        
        External Physical Activity (Extra HR values)
        
        Level of correlation
        
        Classes
        
        Types
        
        Samples (Endpoint=Startpoint)
        
        Intervals
        
        Subclasses
        
        Category (Bolus info/ TBR/ Insulin Profile/ Extended Bolus/ TT/ Profile switch)
        
        Occurrences (time-series trends)
        
        Correlations (info ID connected to other entries)
        
        Other info (External activity from another sensor)
        
        Pre-loop data?
    - - WP2
        
        QoL
        
        Diabetes-related complications
        
        Comorbidities
        
        Weight, height? (e.g. to monitor child growth)
      - WP4
        
        Socioeconomic status
    - - Data Entries
        
        Technological data
        
        Demographic data
        
        Medical data
        
        Diabetes Management data
    - - Patient stratification
        
        Age-ranged:
        
        Children (8-12)
        
        Teenagers (13-18)
        
        Parents
- - - - Diabetes-related dimensions
        
        1.- Environment
        
        2.- Economics
        
        3.- Safety
        
        4.- Nature
        
        5.- Infracstructure
        
        6.- Endocrine disruptions
        
        7.- Education
        
        8.- Policy
        
        9.- Diet
        
        10.- Activity
    - - Machine Learning models
        
        Usage
        
        Variable importances (variables for predicting better app decisions and diabetes coaching)
        
        Pattern classification (target data, patient stratification)
        
        Trends in time-series (individual help for diabetes coaching and education)
        
        Which ones do what on Diabetes Research?
        Paper: Artificial Intelligence Methodologies
        and Their Application to Diabetes
        More info on that: https://www.jmir.org/2018/5/e10775/pdf
        others... Decision Trees, Naïve Bayes or K-Nearest Neighbor
        
        ANNs
        
        Hypoglycaemia detection
        
        Glucose Prediction
        
        SVMs
        
        Diagnosis of Diabetes
        
        GA
        
        Estimation of Model Parameters
        
        Foot Ulcers Risk
        
        Glucose Prediction
        
        Bone Mineral Density Prediction
        
        GA Diabetes Retinopathy Detection
        
        Recommender Systems (Unsupervised Learning)
        
        Daily-life support system
        
        EA: evolutionary algorithm.
        
        RA: regression algorithm
        
        DT, ANN
        
        Pattern recognition
        
        RF
        
        NLP
        
        DT
        
        Adverse glycemic events