Human Centered Design in AI-Based Healthcare
In conversations with medical professionals today, one persistent challenge is the timely and accurate diagnosis of patients. With the growing need for care today, the availability of specialized physicians, their time with the patient, and access to good medical care equipment have become a significant challenge. The healthcare industry often lacks a fully personalized treatment plan for the patients. With the hectic schedules of medical professionals, a patient’s medical history is easy to miss, making the treatment in a ‘parts-and-pieces’ approach with doctors bucketing the patient according to similar cases they have seen earlier. Healthcare innovation needs to be designed around the day-to-day reality of the doctors, seamlessly fitting into their workflow. In an environment that needs split-second decision-making with high stakes, we need to build tech rooted in a deep understanding of the clinical context and caters to the doctors’ emotional and cognitive needs.
HCD necessitates observing individuals as humans. The design strategy focuses on humans and their needs, behavior, experiences, emotions, perspectives, irrational decision-making, ambitions, motivations and lifestyles rooted within distinct cultural settings in producing a design. HCD is a shift of viewing humans not as a part of the system but central in every aspect of the design.
Empowering Human Decisions
Doctors, Physicians, and Technicians are at the core of using AI tech in their workflows. It is crucial to design a system that centers on understandability and straightforward interpretations. We need to look at algorithms that work on human terms and have clarity on what data it gives doctors to make precise decisions. Designers need to design AI that studies the problem from a system perspective and then translates it as a helping hand into actual clinical practice. Designing for cross-department consultation, cross doctor references, and looking at the patient as a whole with their past medical history will enable doctors to give patients more personalized recommendations and better quality healthcare that is standardized and scalable. Humans fear what they don’t understand and readily embrace what they can control. Designers need to explain the benefit of the technology and show the users ‘how’ the systems works and how they can control the process rather than explaining the technology.
Role Of Empathy
Being human-centered allows designers to look at the mess of the natural world and deal with real ground complexities. In day-to-day practice, doctors lose sight of individual people and individual bodies. Empathizing with the user is a never-ending pursuit and cannot be limited to one design phase. Empathy is less about the giver and more about the receiver. We can broadly look at three types of empathy — Emotional Empathy (What is Felt), Cognitive Empathy (What is Thought), and Behavioural Empathy (What is Expressed). We need to learn about the shifting needs and goals of doctors and what their lifestyles demand from them. It is essential to know how they reason, communicate, and collaborate.
Designing For Error
Making rules do not always embrace real-world complexity. Designers need to evaluate the potential unintended consequences of AI and its limitations in a clinical setting. In a high-risk and unpredictable environment, doctors need to be given an informed point of view and oversight about the possible errors, stakes and impact. In such situations, designers need to design manual controls and a system that recommends a way forward to avoid an analysis paralysis state of the user.
Designing For Context, Not In Isolation
For doctors and patients, it is essential to examine human needs, behavior, interactions, experience, and cognition in-situ. The approach allows designers to examine the societal influence of AI in healthcare. When designers look at the larger context of the collective medical minds, there will always be more decision-making. Technology does not need to be looked an add-on, but a complement to human skill and efficiency such that fewer people can do more in a shorter time. We need to build machines that serve a human group’s specific behavior, letting AI do the work people cannot.
Designing For Shaping a Practice and Lifestyle
Integrating AI-based tools in healthcare can primarily change the way the medical community organizes and practices medicine. By designing AI, we design a doctors decision-making and how technology mediates their knowledge and understanding. The impact it will have on doctors is the way they perceive, experience, and behave while making decisions, enabling them to know when and why a patient needs a particular diagnosis.
Readings and References
[1] Tricog, A revolution in cardiac care, https://www.tricog.com
[2] Niramai, Breast cancer screening, https://www.niramai.com/technology/
[3] Qure.ai, AI for medical Imaging, https://qure.ai
[4] Design Research Society, Human-Centered AI: The role of Human-Centered Design Research in the development of AI, https://dl.designresearchsociety.org/cgi/viewcontent.cgi?article=1178&context=drs-conference-papers
[5] Philips, Why AI in healthcare needs human-centered design, https://www.philips.com/a-w/about/news/archive/blogs/innovation-matters/2021/20210419-why-ai-in-healthcare-needs-human-centered-design.html
[6] Google, People+AI Guidebook, https://pair.withgoogle.com/guidebook
[7] Elon Musk, Neuralink, https://neuralink.com/approach/
[8] Shivani Gupta, Consumer Mindset’s in a Post Covid World, https://www.linkedin.com/posts/shivanigupta25_5-industries-in-focus-consumer-mindsets-activity-6689159537536098304-qJXj