Health Data Analytics

Background

The analysis of health/medical data is relevant for many use cases, including clinical research and patient treatment. While these cases typically represent a doctors'/physicians' perspective, there is yet another user group with an immense information need: patients. In addition, other types of stakeholders involved in the medical domain are, e.g., clinics, quality management environments, operational stuff, and insurances.

The analysis of health data has many facets just like data in the medical field have. Individual sources of data include treatments, drugs, histological, clinical, follow-up, or quality-of-life data. These types of data, often referred to as ingredients of electronic health records, are, e.g., often complemented with volume data as a result of medical imaging. Different types of techniques borrowed from Machine Learning, Data Mining, and Information Retrieval may be applied and combined to provide effective and efficient solutions for the analysis of medical and patient-related data.

Despite its benefit the analysis of health data comes with a series of challenges. First, data collections such as electronic health records often have quality problems and require manual (pre-) processing. Second, algorithmic models applied on medical data need to be selected, parameterized, trained, and validated. In most cases this process is iterative and urgently requires humans in the loop. Third, the analysis of model outputs is often not trivial and decisions of models are not always explainable. Finally, involved user groups such as physicians, patients, but also operational stuff have very different levels of expertise and interest, which requires a careful characterization of the domain problem and analysis tasks in order to ensure that the tool to be built is both usable and useful.

Approach

Interactive Visual Data Analysis (IVDA) has proven to be a promising approach to medical data analysis. The particular benefit of IVDA tools is the ability to combine the strengths of both humans and machines in a unified data analysis workflow.

Thesis Goal

The goal of this thesis is to design, implement, and validate a IVDA tool to support a concrete health data analysis use case. From start, a dataset will be available, as a basis for data processing and analysis. The student will have the chance to get in touch with an expert in the medical domain, to learn about the domain problem and desirable analysis tasks. In an iterative design phase, a data science workflow will be designed as well as an interactive visual interface for the visual analysis of the medical data. A validation process needs to be conducted to ensure that the IVDA tool solves the identified medical data analysis problem.

Requirements

• Programming experience in Python or willingness to learn
• Knowledge about interactive data visualization and data processing
• Basic knowledge about machine learning.
• Decent understanding of written English.