We believe the next series of breakthroughs in healthcare are locked deep within clinical data.

We believe the next series of breakthroughs in healthcare are locked deep within clinical data.

We believe the next series of breakthroughs in healthcare are locked deep within clinical data.

We believe the next series of breakthroughs in healthcare are locked deep within clinical data. 

KAIKO_HEADER_2

Overview

At Kaiko.ai, I worked with world-leading Machine Learning Engineers and did research at the National Cancer Institute in the Netherlands (NKI) with oncologists, surgeons, radiologists, pathologists, nurse specialists, and other medical specialists. 
When I joined Kaiko.ai in the summer of 2023, the company grew fast, and so were the different ongoing projects. My job as a Senior Product Designer was to understand how those abstract and technical heavy efforts could be applied at hospitals in the short-, mid-, and long-term future.

RESPONSIBILITIES

User Research – Remote & On-Site
Developing & visualizing the product vision
Goal setting for the Design Team
Creating & deciding on UX Roadmap
Supporting engineers in planning & priorization
User journeys, UX audits, information architecture
Team development
Building Design System
Hiring engineers
Communicating with stakeholders in hospitals

HIGHLIGHTED PROJECTS

User Research for the Tumor Board
MVP of the Dicom Viewer for Radiology
Visionary Kaiko Insight App based on our FM
Company-wide Design Principles

HIGHLIGHTED PROJECTS

User Research for the Tumor Board
MVP of the Dicom Viewer for Radiology
Creating a visionary workflow for radiologists
First Consultation product based on our FM 

Hospital Processes

At Kaiko.ai, teams focus on areas of greatest impact. While our ML Engineers build a foundation model, the design team began exploring hospital processes by interviewing surgeons, oncologists, and other medical specialists. We developed personas and patient journeys to understand how specialists collaborate, make decisions, and communicate with patients. After many interviews and observations, we identified a key bottleneck in the patient journey: the tumor board.

PATHOLOGY_PROCESS

Together with my colleague, Alex, I shadowed and interviewed a pathologists for an afternoon.

WHAT IS A TUMOR BOAR

The tumor board is a weekly hospital meeting. In Europe, it's a medical standard, often required by law. The goal is to reach the best decision regarding the diagnosis and treatment plan of a patient by leveraging the extensive expertise of diverse medical specialists. With a former Head & Neck surgeon, we observed these boards and interviewed doctors to understand how they prepare patient documents, hypothesizing that this is when critical data is evaluated and key decisions are made.

MDO ROOM SKETCH

Sketch of the room layout for the tumor board.

We want to get an overview of the tumor board process and spot emerging opportunities where data can make impactful improvements. 

Get an overview of the tumor board process and spot emerging opportunities where data can make impactful improvements. 

34+
Interviews

34+
Interviews

8
Roles

8
Roles

659
Sticky notes

659
Sticky notes

FINDING OPPORTUNITIES

We interviewed head and neck surgeons, nurse specialists, oncologists, radiologists, pathologists, and many more; we shadowed different departments. Together with the team, I created a huge affinity map, grouped sticky notes and started ideating with the team – always in-sync with our engineers efforts.

KAIKO_PROCESS_CRAZY_8_5

Examples of a Crazy 8 session with my cross-functional team.

CONCEPTS & WIREFRAMES

Over the next months we started building conceptional prototypes and wireframes to understand what content can technically be displayed in what form and how this will be most valuable to medical specialists. We iterated fast, by interviewing dozens of doctors across different departments.

WIREFRAME_1

A version of a wireframe for a head and neck surgeon. We tested the content, and hypothesises – no interactions. 

04 HI-FI PROTOTYPE

In the Summer of 2024, the design team took the conceptional prototype a step further, added more detailed interactions, and applied our design principles and design system. These are ongoing efforts until today and will be again tested with medical specialists. I’m happy to share the current password-protected link to the latest prototype.

KEY LEARNINGS

Working with machine intelligence is the weirdest profession ever. Crafting minds. Diving deep ino the unknown, not only hypothesing, but observing the fundamental phonemena of information and data. As a designer I am trained to see processes and patterns, that don't exist yet. The potential of things which could be. Never in my career I have aimed to thing this far out there while going as deep as possible to understand my users.

MOOD_PROCESS_2
MOOD_PROCESS_1
MOOD_PROCESS_3

MY PERSONAL VISIOIN

I believe in the full potential of multimodal foundation models. We will be able to treat cancer and predict and prevent this highly complex disease. I’m thinking about superhuman technology that will eventually make healthcare available to everyone, and let doctors fully focus on the wellbeing and quality of life of a patient – always.