ICDeep 10

Application for Clinical Trial Eligibility of Potential Participants

Get Started

What is ICDeep 10?

ICDeep 10 is a system for clinical trials management system created for the Department of Oncology at the Karolinska University Hospital. It gathers all ongoing trials in a singular location in order to make managing and performing trials less tedious and timeconsuming. It also assists in matching potential patients with trials using information obtained from the TakeCare system.

ICDeep 10 was designed with many of the useful features found in other similar products, such as IBM Watson. It is intended to work with any Electronic Health Record (EHR) system at any hospital, however since it was developed in cooperation with Karolinska University Hospital the TakeCare system and the Department of Oncology was used for the development of the prototype. Further information about this and everything else can be found in the report covering the entire project.


This project has been a principal investigation to determine how the process for matching patients with clinical trials could be improved. The presented results are a prototype showcasing a possible way to design this system as well as a report explaining the design process.

Features

CMS for Clinical Trials

ICDeep 10 provides a Content Management System for creating, editing and deleting clinical trials within the system. It also provides an overview of all the currently ongoing trials. The system aim to simplify the current process of managing trials by digitalizing a process that still in large part is done with physical papers. It is also in our interest to make the trials more easily accessible for those working that the hospital, since the process of finding patients for trials is dependant on doctors and physicians knowing that they exist.

Semi-automatic patient matching

ICDeep 10 uses the API available for the TakeCare EHR to obtain basic information about patients in order to automatically filter away some of the exsisting clinical trials. ICDeep 10 uses three different parameters for automatic filtering, Age, Sex and Diagnosis (ICD-10 format). In addition to the automatic filtering ICDeep 10 provides an interface for further filtering patients. Displaying inclusion/exclusion criteria together with the patients journal in one view to make the filtering process as smooth as possible.

Shortlist functionality

To simplify the transition from the current routine to our system it implements a shortlist functionality. Patients are assigned to clinical trials (or treatment) at a weekly multi-disciplinary conference (MDC). For this conference there is always a list of patients to be discussed. The shortlist functionality of ICDeep 10 allows for easy management of said lists and, most importantly, ICDeep 10 notifies which patients are very likely to be eligable for a clinical trials on the list.

Prototype

To demonstrate our proposed design and our solutions we created a prototype of our system. The interactive prototype has been used as the common ground for discussion with client and stakeholders. During the course of the project the system has been presented, evaluated and iterate with help from the prototype.

The prototype was built with Axure RP Pro V7.

Prototype


Supported browsers:
  

Paper

Abstract

The purpose of this paper is to investigate the possibilities improving the system for matching clinical trials with patients at the Department of Oncology at the Karolinska University Hospital. From investigating the current process we learned that the system used i mainly based on manual labor and paper based clinical trials. Therefore different computer based solutions such as: IBM Watson, Epic Beaker and SAP Hana are research and a State of the Art analysis of IBM Watson is performed. Based on the State of the Art we decided that it would be most fitting to develop a new system for use at the Department of Oncology. With the use of parallel prototyping we created a prototype for managing clinical trials. In addition to the prototype, further contextual observations and interviews were performed in order to increase our understanding of the target environment. Furthermore, the prototype was iterated over based on feedback gathered during workshop sessions with the clients as well as an interview with a research nurse. The prototype created generated positive feedback overall during both the workshops and the interview. Therefore, in the last sections of the paper, we discuss what needs to be research further before this system could be implemented as well as how to test it's effectiveness if it were to be implemented. We also propose different areas to further investigate in order to expand the scope and possibilities of our proposed system.

Full paper

Contact us



Contact Us!

About us

Created by Niklas Blomqvist, Gustav Hammarlund, Aydin Heydari, Marcus Rönnmark, Elizaveta Sigova, Huiting Wang & Charles Windlin

Created for the courses DH2655 Cooperative IT-Design & DH2460 Software Design - Business - Leadership at KTH in cooperation with Rolf Lewensohn at the Department of Oncology at the Karolinska University Hospital in Solna.