EVOLVE Job Shadowing Series: Mahlet Birhanu on her visit to the Population Imaging Node Valencia – job roles, health data management tools, and seeing your blind spots.

The EVOLVE Job Shadowing program is a fantastic opportunity for Node staff to immerse in the daily life and operations of other Nodes of the Euro-BioImaging community. Between September 2024 and March 2025, during the programme's first phase, twelve participants will share, learn, explore and discuss topics ranging from innovative imaging processes to facility management best practices. As participants returned to their home institutions, we asked them to reflect on their experiences and share their insights. Discover how this unique program ignited inspiration, sparked new ideas and brought colleagues closer together.

In today’s edition of our EVOLVE Job Shadowing Stories Series, we have the pleasure of talking with Mahlet Birhanu, from the Population Imaging Node Rotterdam, about her week with Pedro Mallol Roselló.

Mahlet’s electrical and computer engineering background gives her a unique perspective on medical imaging. After a Master’s program hosted in three different European countries, she began an internship in the Netherlands and now works as a research software engineer at the Erasmus Medical Center in Rotterdam. 
Euro-BioImaging: Nice to talk with you, Mahlet. What motivated you to join the EVOLVE Job Shadowing Program?  

Mahlet Birhanu: I'm glad to be here! I work as a research software engineer, and this role fits between researchers and clinicians. Clinicians directly take care of patients and usually collect data for diagnoses and health-related issues. Researchers need this data to advance their work. We, research software engineers, are exactly in between these teams, playing a somewhat less defined but extremely important role!

I was eager to explore how other facilities operate and learn their approach to our issues. Gaining these insights and returning them to my home institution was super helpful. 

Euro-BioImaging: The importance of these roles will undoubtedly grow with the increasing importance of data management and analysis. When you made the application, you already knew where you wanted to go. You specifically wanted to go to Pedro Mallol Roselló in Valencia. What drove you to choose that host?

Mahlet Birhanu: We both participate in many of the same European projects but have never met in person. When this Job Shadowing program came up, it was very easy for me to choose because we work in similar areas. 

Euro-BioImaging: Had anyone from your two Nodes met in person previously?

Mahlet Birhanu: Not really, perhaps an introduction in a hallway or coffee. There had been some interactions but not really working together, like during my visit, but they have done so since my application.

Euro-BioImaging: It's great that the Job Shadowing Programme was a nice opportunity to strengthen ties between your Nodes! Moving on, what were your expectations when you applied for the program?

Mahlet Birhanu: I did not have particular expectations! I decided to go to Valencia, meet with Pedro and his team, and spend the whole day at their office browsing their systems, with someone guiding me through their software and processes. Something relatively low-key! 

Euro-BioImaging: And were your expectations met?  

Mahlet Birhanu: Absolutely! The whole experience exceeded my expectations! The whole visit was amazing. Everything was well organised, and the Population Imaging Node Valencia team took the program seriously. The five days of my visit were well-structured and perfectly scheduled, with no downtime. I met with the three most relevant teams for my activities. We had time slots scheduled for the team members to spend time with me, explaining their approaches in detail and discussing our common issues. 

Euro-BioImaging: It’s fantastic! You would give top marks to the visit! What struck you most about the Node?

Mahlet Birhanu: A lot of things stood out! I was amazed to see the size of the Data Scientist team there. They are organized into different teams with specific assignments. There were a lot of young people there, and the teams showed a good gender balance, which was also really nice. 

Euro-BioImaging: What new insights did you bring back to your Node from your visit?

Mahlet Birhanu: Regarding the technical implementation, it was easy for us to understand the hosting team because we face similar challenges in our work. We spoke about how to face these common issues in general, but one specific example stood out: an open-source viewer that we, in Rotterdam, want to integrate with other tools, enabling clinicians to review whether segmentation has been performed correctly. The team in Valencia are also working on a very similar project that allows clinicians to directly interact with results and includes AI algorithms that you can interface with. Their implementation was very detailed, and it proved helpful for me to learn directly from them. I immediately shared this with my home team when I went back! 

Euro-BioImaging: Did you come across any challenges during your visit?

Mahlet Birhanu: I noticed that many of the tools created at the Valencia Node were developed in-house rather than as open-source software. This makes it difficult to delve into the code. However, seeing that similar solutions have already been successfully implemented elsewhere was reassuring. This boosted our confidence that the tools we’re developing will work well for us!

Euro-BioImaging: Would you recommend the job shadowing program to other node staff members, and if so, why?

Mahlet Birhanu: For sure, I would recommend it! It's enriching to collaborate with another team on the same challenges. Also, you notice many things when you get away from your group. Our team also works on specific ontologies for radiology images, such as MRI, CT, and those kinds of images. We try to define the data cataloguing standards and ontologies for specific imaging data. Now, when you're meeting over Zoom, obviously we're from different countries, we have different standards to comply with, and the way we think is a bit different sometimes, and everyone's not always on the same page. But then, when you actually go to the centre and see why others think in a certain way, how they are organized, you understand a lot more clearly why they use one ontology rather than another.

I have identified many biases and some challenges I may have overlooked, such as how data is organized. Moreover, other people have a different mindset when organizing their teams. You suddenly see a lot of your blind spots when you go to a different Node and see how it works.