tutorial

OARD : An Open Real-world based Annotation for Rare Diseases and its Associated Phenotypes

How to use OARD web application

Input your concepts

• OARD can be used to query single concept frequency or paired concept frequency. The widgets will be automatically adjusted according to your input.
  • For single concept frequency, input the query concept in the first query box
    • You are allowed to pick more than one items;
    • If no input provided in the first box, the concept most frequently occurred will be returned;
  • For paired concept frequency (or association), input the query concept in both boxes;
    • You are allowed to pick more than one items;
    • At least one concept is required in the first box;
    • If no inputs provided in the second box, the concept most frequently co-occurred (or most associated) with the first one will appear;

Select your sevice/method

• There are two service provided by OARD, frequency or association.
  • Frequency works for a single concept or a paired concepts
  • Association only works for paired concepts
    • There are multiple options (statistics) available for association calculation;

Select your dataset

• There are many datasets available to display the frequency or association information;
• They are classified according to clinical site, source, subpopulation; details can be found by clicking how to select dataset;
  • e.g. dataset 2 is compiled by extracting concepts from all clinical notes from CUIMC and generating a population-level concept frequency and co-occurrences (or association) for all cuimc populations
  • e.g. dataset 22 is compiled by extracting concepts from all clinical notes (when patient is in their teenage) from CUIMC and generating a population-level concept frequency and co-occurrences (or association) for all cuimc populations
  • e.g. dataset 199 is compiled by extracting concepts from all clinical notes from CUIMC, aggregating the concept by their hierachical position, and generating a population-level concept frequency and co-occurrences (or association) for all cuimc populations
  • e.g. datatset 190000707 is compiled by converting structured OHDSI database (mainly contains billing and lab codes) to HPO/MONDO concepts, and generating pulation-level concept frequency and co-occurrences (or association) for all cuimc populations having a phenotype related to "abnormality of the nervous system"

Select your domain and return

• When in the mode of returning most frequency or most associated concepts, you can specific the domain of concepts you want to return; you can specify the number of hits you want to return

Examples

Example 1 - find single concept frequency for muscle weakness

Example 2 - find most associated (ranked by obs/exp ratio) diseases with muscle weakness and skeletal muscle atrophy

Example 3 - find co-occurrence frequency of a concept pair of muscle weakness and Duchenne muscular dystrophy

Example 4 - find relative frequency of muscle weakness in duchenne muscular dystrophy patients