/thesis-Coverage-Analysis

Primary LanguageKotlin

Thesis : Framework for Coverage Analysis in Digital Phenotyping

This project is part of my DTU MSc thesis, and contains the implementation of a framework for coverage analysis in digital phenotyping. The framework is written in Kotlin, and contains a key dependency to CARP Core.

Project Structure

The project is structured as a Gradle project, and contains the following modules:

  • src: Contains the source code of the framework.
  • test: Contains unit tests for the framework.
  • test_data: Contains test data for the evaluation use cases.
  • catch_code.py: Contains the Python code generated by ChatGPT for the CATCH study.
  • diafocus_code.py: Contains the Python code generated by ChatGPT for the DiaFocus study.

Architecture

The framework follows the Onion Architecture, and is thus divided into three layers:

  • Domain Layer: Contains the core business logic of the framework.
  • Application Layer: Contains the services of the framework.
  • Infrastructure Layer: Contains the concrete implementations.

Evaluation

The framework was evaluated using three use cases:

  • CAMS: Coverage Analysis for CARP Mobile Sensing (CAMS).
  • CATCH: Coverage Analysis for the CATCH study.
  • DiaFocus: Coverage Analysis for the DiaFocus study.

These three use cases are implemented in the TestApplication.kt file, and can be run using the main function.

Documentation

While all classes are documented via code comments, a more detailed description of the project can be found in the corresponding thesis.