The medical imaging technologist was working with a UI that had been designed and added on, over the better part of 40 years. The result was a fragmented, Frankenstein-esque experience.

CT scanner for GE Healthcare

UI for CT scanner focused on improving the tech workflow and reducing risk for over-radiation

UI design focused on improving radiation dose management and techs' workflow. The tech's interface span over two monitors, had once been a touch interface and various windows from different eras would even make the cursor take shapes I had never seen from long forgotten times. 

The industry had been plagued with terrible, accidental incidents of over-dosing radiation to patients. Over a 5-month project we worked closely with the client team to research, prototype, design and document.

We designed improved information and safeguards against overdosing are now designed into the system through various ways.

Head techs' and radiologist can lock certain parameters for protocols that should not be altered on the fly. This was expected to be hugely controversial by the client team and the idea was thrown in as more or less of a sacrificial concept but techs agreed that some parameters could be locked to avoid mistakes.

Beyond clarifying the total dose to be administered, we designed in various moments and places in the interface to alert and educate techs' about what changes and actions altered radiation dose.

One example of a design feature we introduced was the Dose Rage indicator showing the tech what the current settings would suggest the size of the patient to be. If the settings indicate that the patient should be above average size and the actual person on the table does not match this physical description, this should be cause of concern.

The new design also improves the techs' workflow by decoupling the scanning phase when the patient is on the table and post processing tasks. This allows techs' to better manage peaks and troughs throughout the day by leaving non-time critical post processing work to be done during slow periods.

Above you can see the first level of resolution on our first round of research where we tested flow and various mental models.

The second round focused on a few key modules in the new design with the chosen hierarchy. The resolution varied depending on what was necessary for each module.

The last and final major round of research had a high resolution visually. We used two monitors in an extensive click-trough we had memorised the path for. The experience for the techs were quite realistic and convincing.


• Develop the future vision of the premium CT scanner UI 
• Optimise the tech’s workflow and the patient experience
• Reduce the patient’s overall radiation exposure

Time and team

• Another interaction designer and I, a software designer (also project lead), a design researcher as well as an external visual designer contractor
• Close collaboration with the client team
• The initial engagement lasted 5 months
• We were re-engaged through various sprints to dive deep into specialised protocols


This project was very complex. In order to intelligently design a new approach we had to understand the underlying scanning technology, anatomy and related terms as well as the clinical procedures involved in the scanning a patient and "image" needs of specialists like radiologist who would ultimately use the images to diagnose. In addition, the outputs i.e. the images for the radiologists has very specific demands. 

I can honestly say I loved this project. Diving into a field I knew nothing about with a steep learning curve, taking to experts and people at the frontline and having a chance to design something that truly matters and maybe even saves lives — I couldn't have asked for more.