Hasso-Plattner-Institut25 Jahre HPI
Hasso-Plattner-Institut25 Jahre HPI
Login
 

Process Analytics in Healthcare (Wintersemester 2021/2022)

Lecturer: Prof. Dr. Matthias Weidlich (Business Process Technology) , Dorina Bano (Business Process Technology)

General Information

  • Weekly Hours: 4
  • Credits: 6
  • Graded: yes
  • Enrolment Deadline: 01.10.2021 - 22.10.2021
  • Teaching Form: Lecture / Exercise
  • Enrolment Type: Compulsory Elective Module
  • Course Language: English

Programs, Module Groups & Modules

IT-Systems Engineering MA
Data Engineering MA
  • CODS: Complex Data Systems
    • HPI-CODS-K Konzepte und Methoden
  • CODS: Complex Data Systems
    • HPI-CODS-T Techniken und Werkzeuge
  • CODS: Complex Data Systems
    • HPI-CODS-S Spezialisierung
Cybersecurity MA
Digital Health MA

Description

This seminar explores state-of-the-art research topics in process analytics, which is the data-driven assessment of how activities in an organization are coordinated. The seminar focuses particularly on analytics in healthcare, where processes denote, for instance, clinical pathways and treatments of patients. Specific areas to be covered in this seminar are:

 - Data extraction and abstraction. Data to be used for process analytics may first need to be extracted from heterogeneous sources (from health information systems to real-time locating systems in hospitals) and abstracted to an event log that indicates the progress of a process. Here, for instance, techniques for data profiling (to identify correlations) and data mining (e.g., to recognize activities from sensor data) may be employed.

 - Sample-based analytics. Facing large volumes of data, valuable insights may sometimes be derived already from a sample of the data. However, the extraction of a sample that is most useful for a certain analysis question is still rather unexplored.

 - Resource Allocation. Process analytics may target the assignment of the right resource (e.g., medical staff, machines, etc.) to the right task of a process in order to improve the service quality. Various perspectives may serve as an optimization goal, though.

 - Data-enhanced processes. In medical contexts, a process may be driven by changes in certain data, such as patient information. Therefore, models that combine traditional control-flow-oriented views with data-centric views are needed.

During this master seminar, after being assigned a research topic, each student (or a group of two) will perform the following tasks:

 - Literature review based on the references provided by the supervisor

 - Give two presentations in front of BPT group members and other participants (one intermediate and one final presentation)

 - Write a research paper based on the assigned topic

 - Build a proof of concept prototype (if applicable)

 - Participate in all presentation meetings and discussions

 

What you will learn:

 - How to generate research ideas

 - How to work in a team

 - How to give a scientific presentation

 - How to write a research paper

 - Overview of some of the advanced topics in the area of process analytics

Requirements

The seminar is intended for master students, who are familiar with basic notions of process modeling and execution, and who would like to address research challenges in the field of process analytics.

Literature

The literature will be provided in the kick-off meeting for each topic.

Learning

Further details regarding this seminar will be announced in the kick-off meeting, including the concrete topics, registration deadline and other important dates.

Examination

The seminar will be graded based on the following structure

 - Intermediate presentation

 - Final presentation

 - Research paper + prototype

Dates

- Kick-off meeting: 19.10.2021, 15:15h - done

- How to Give a Scientific Presentation: 26.10.2021, 15:15h - done

- How to Write a Scientific Paper: 02.11.2021, 15:15h - done

- Intermediate presentation: 04.01.2022, 15:15h - done

- Final presentation: 15.02.2022, 15:15h

- Research paper + prototype: 18.03.2022, 23:59h

Zurück