Colloquium IS (June 1, 2018) – EHR-data driven clinical risk prediction for CVD

Dear all,

We are pleased to invite you to our next Colloquium IS that will take place on Friday,

June 1, 2018, 16:00 – 17:00 (Paviljoen K.16).

Speaker: Zhengxing Huang from Zhejiang University

Title: EHR-data driven clinical risk prediction for CVD

Abstract:

With the rapid development of hospital information systems, a large volume of electronic health records (EHRs) has been produced, which provides a comprehensive source for clinical risk prediction. In this talk, I am concerned with the problem of utilizing the heterogeneous EHR data to assist clinical risk prediction for cardiovascular disease patients.  Specifically, two challenges are addressed in this study: 1) how to annotate unstructured EHR data for generating a benchmark dataset, and 2) how to train a predictive model on the benchmark patient samples which are typically imbalanced and correlated to each other.

Colloquium IS (May 4, 2018) – Managing variability in knowledge-intensive processes

Dear all,

We are pleased to invite you to our next Colloquium IS that will take place on Friday,

May 4, 2018, 16:00 – 17:00 (Paviljoen K.16).

Speaker: Rik Eshuis

Title: Managing variability in knowledge-intensive processes

Abstract:

In modern organizations, the importance of knowledge workers and the knowledge-intensive processes (KiPs) they perform is ever increasing. Examples of KiPs are new product development processes, smart maintenance processes, diagnosing and treating patients etc. In a KiP a case is processed and evaluated based on human judgement of knowledge workers, who often rely on digitized information to apply their knowledge. Flexible BPM technologies like case management and business artifacts support KiPs. A KiP model typically needs to be adapted to the case being processed, leading to different but related variant models of a KiP. In this talk, I outline an approach based on business artifacts to manage variability in KiPs, in order to avoid rework and improve the quality of KiPs.