Taking a Human-Centered Approach to Medical Informatics - an Interview with Mario A. Cypko
Dr. Mario A. Cypko is a postdoctoral researcher at the HCC Research Group. His work focuses on the future of robotic-assisted surgery. Mario's work is part of the project »Filtering« at the cluster of excellence »Matters of Activity« (MoA).Mario, you have a background in medical informatics, how does this field relate to research in human-computer interaction (HCI)?
In short, medical informatics acts as an interface between medicine and computer science. There are basically two main areas: hospital IT (information technology), which deals with the technical infrastructure and data exchange in hospitals, and computer-assisted medicine, which develops technologies to support medical professionals in the prevention, diagnosis, treatment and rehabilitation of patients. My research focuses on the latter area, computer-assisted medicine, and I am particularly interested in how we can use so called »artificial intelligence« (AI) in medicine in a meaningful way. Research in this area has the goal to develop computer programs that support clinicians in elaborate or complicated activities. Applications that integrate »AI« are, for example, being used to detect anomalies in medical images and other data. Such applications are referred to as »decision support systems« that help experts in their medical decision making. Another area of my research focuses on the development of surgical »robots«. Today, robotic-assisted surgery mostly means that surgeons operate by controlling a robotic arm through a console. Surgical »robots« are thought of as instruments that enable surgeons to perform tasks with a higher degree of accuracy, for example in cases when the manual manipulation would be too imprecise if done by hand. Also, the robotic instruments can reach angles that would not be possible with conventional methods and thus open up new intervention possibilities. One area of research in medical informatics relates to the technical feasibility and touches on questions of implementation and engineering. My current research has a slightly different focus: I focus more on the human-centered aspects of medical informatics that relate to questions of how we can integrate robots in the operating room in such a way that takes into account the requirements of physicians, patients and other stakeholders.What are the challenges of robotic-assisted surgery that you address in your research?
In line with the human-centered approach in our research group, we address the challenge of how we can enable interaction between surgeon and »robot« while also taking into account the effects on the whole operating team. The so called »robot« is actually a telemanipulator. This means that the robotic arm with the camera and instruments is placed directly above the patient and is used to perform the surgery, but it does not move autonomously. The movement of the robotic arm is controlled by a human surgeon through a console, which is placed in a corner of the room away from the patient. When the surgeon sits at the console, they are distanced from the patient and from the other members of the operating team. Integrating a »robot« in the operating room means that certain information is filtered and communication is more difficult. This means, for example, that the surgeon might not be aware of visual or auditory information that is out of their sight behind the console, or the communication between the team and the surgeon might be impaired - this is especially noticeable when problematic and stressful situations occur. Our research takes into account all actors in the operating room, including the surgical team, the patient, the »robot« and other surrounding machines and potential influences. What happens in the operating room between these actors can be thought of as »entangled interaction«. In the current phase of our research, we examine and seek to understand the interdependencies and the interaction in the operating room and communicate it in a visual language, which is the first step. As part of our future research, we will also explore the ways in which AI might be able to support the actors in the operating room in a meaningful and responsible manner.How do you go about in conducting your research in this context?
I work closely with clinicians and designers. I conduct observations in the operating room and also conduct interview studies with surgeons and other clinicians that are part of the operating room team. This helps us to first get a comprehensive understanding of the context and of how things work in the operating room. One challenge that we've identified is that there is no predefined form of notation for observations of interactions between human and non-human actors in the operating room. This makes it difficult to discuss situations or even problems, especially in such an interdisciplinary setting with different expertises and forms of communication. For this reason, I am currently working with designers from the Kunsthochschule Weißensee. Together, we are developing a visual vocabulary of surgical procedures that can be understood across disciplines and helps us to »map« the entangled interactions. This development and evaluation of the visual language is very exciting and happens in several cycles. In each cycle, I sit down with the designers and we go through different visualization options. Our goal is to create a visual language that is rich enough to capture the medical complexity while it should be understandable by people with different expertises and academic backgrounds. We will evaluate the visual vocabulary with clinicians and stakeholders from other disciplines. The visual language then becomes a tool that we can use to discuss the observations with our clinical partners and other disciplines. This helps us to conceptualize what kind of »AI« could be useful and reasonable to integrate in robotic-assisted surgery, in what form, and with what influences - the goal of our research is to ultimately design the interaction between humans and non-human actors.
Additional information and links:
Mario has talked about medical informatics and his research on robotic-assisted surgery with the team from the "BMBF MINT Magie" project, which you can watch on Youtube.
Katrin has conducted a workshop that engaged with the filtering effects of robotic assisted surgery as part of the two-day workshop on »Automation, Control, and Incompetence – Socio-Technical Ecologies of Responsible Filtering in the 20th and 21st Century« at the cluster of excellence »Matters of Activity« (MoA) - you can read the workshop report here.