Daan Hekking

Colab - introducing collaborative robotics in the clinical microbiology laboratory

Daan Hekking
Colab - introducing collaborative robotics in the clinical microbiology laboratory
 
 

Why this project?

How would you feel if you obtain a master degree after years of intense studying, but end up spending most of the time during your job doing repetitive tasks that do not require all these years of intense education?

Setup

A 10-week term project in collaboration with ABB Corporate Research and tutoring from Biligi Karan (Former Above employee). Executed at the Umeå Institute of Design in Sweden, from Mid-October 2017 till December 2017. 

In collaboration with classmate Marc Saboya Feliu

Opportunity

In the age of automation and digitalisation, people are scared their jobs are being taken over machines and can’t compete with high output these machines/robots/algorithms have. What if we could transform these worries into something positive. Can these people work together with these ‘competitors’ in order to elevate and achieve a greater performance in the context they are working in?

Result

Colab is a workstation for clinical microbiologists that targets the issue described above in the laboratory context. This workstation allows them to reduce the number of repetitive tasks and improve the overall quality during inoculation. Colab also gives microbiologists a better insight into bacteria growth during the incubation phase with the help of AI. The Time To Result (TTR) will be reduced and the overall quality of a sample will be improved, which not only benefits the patient due to quicker diagnosis but also allows the microbiologists to focus on tasks they love to do within their field.

 

 
 
DaanHekking_Portfolio2018.jpg
 

 

 

what is this laboratory about?

Clinical microbiology in a nutshell

 

Clinical microbiology is a field within the medical science where clinical microbiologists focus on the prevention, diagnosis and treatment of infectious diseases. When a patient is sick, often his/her blood/urine/feces etc. is sent to the clinical microbiology laboratory for analysis. The microbiologists identify the bacteria that possibly makes this patient sick and prescribe a treatment accordingly, which is in many cases antibiotics. Before prescribing the right treatment to a patient, a highly complex procedure goes prior to this.

 
 

 

 

how have we tackled this?

Research strategy

As we believe in including the users of the final solution in the product development process, we have decided to set up a close collaboration with the Umeå Clinical Microbiology laboratory with several checkpoints in this ten-week project. The first visit to the laboratory gave us a better understanding of clinical microbiology in general, what it takes to prescribe treatment, what kind of equipment they are using etc. 

 
DaanHekking_Portfolio2018_Process.jpg
 

 

 

Understanding their daily drivers

Why do they do what they do? And how?

After synthesising, a flowchart of the clinical microbiology laboratory has been created together with a 3D layout of the laboratory with LEGO. Two weeks later we spoke to most of the microbiologists again in the laboratory during our ethnographic research session in order to understand preferences, the reasoning behind becoming a clinical microbiologist, what gives them joy in life besides work etc. As we were aiming to create a solution that solves a problem and aligns with their preferences and integrates within their lifestyle, it is important to deeply understand a clinical microbiologist’s behaviour, thoughts, preferences and general wellbeing. We showed the clinical microbiologists different robots and machines to find out how willing they are in working together with robotics. 

 
 

 

 

working with constraints

Smaller laboratories can't afford full automation

The goal of this project was to identify the most appropriate way of implementing collaborative robotics in the clinical microbiology by prioritising the clinical microbiologist’s preference so he or she goes to work every day with excitement and chances to expand their intellectual capabilities. Implementing a robot or introducing automation is something that is easy to imagine. In many fields, automating tasks is the first step to scale up productivity. Three of the problems we have identified relate to the level of needed productivity, financial status and spatial environment of the laboratory. A small scale laboratory does not require the same sample productivity as a large scale laboratory and would therefore over invest when buying new laboratory equipment. As these small scale laboratories handle fewer samples, their budgets are smaller and have difficulties acquiring the latest technology that improves quality and speed. Also, in person user-research has shown that a small scale laboratory has spatial limitations.

 

 
DaanHekking_Portfolio20181.jpg
 

 

 

creative exploration

Learn by doing

After our desk and field research and synthesising the information, ideas were generated. During this ideation phase, we explored several ways of communicating ideas. As we were focussing on collaborative robotics, the human-machine interaction was of great importance. We held internal ideation workshops at the university and had several role-play exercises to embody the idea, its scale and visualise the human-machine interaction in order to streamline and improve a clinical microbiologist’s workflow. Everything was documented, both on video and on still images for further analysis. After these internal workshops the ideas (sketches, 1:1 models and videos) we visited the clinical microbiology laboratory for the third time and discussed the generated ideas. Within the same week, we also presented our work to ABB, who is a leading player in the field of robotics. Feedback from both parties was taken into account and a realistic, innovative and slightly provocative, but still accepted by its users, concept was created.

 

 
 

 

 

introducing

colab - making repetition less repetitive

Colab is a workstation for clinical microbiologists that targets the issue described above in the laboratory context. Our desk and field research showed us that up to 49% of a clinical microbiologist’s time in the laboratory is spent on repetitive tasks that do not require the number of intellectual capabilities clinical microbiologists have. This workstation allows them to reduce the number of repetitive tasks and improve the overall quality during inoculation. Colab also gives microbiologists a better insight into bacteria growth during the incubation phase with the help of Artificial Intelligence (AI). The Time To Result (TTR) will be reduced and the overall quality of a sample will be improved, which not only benefits the patient due to quicker diagnosis but also allows the microbiologists to focus on tasks they love to do within their field.

 
DaanHekking_Portfolio20184.jpg
 

 

 

what is inside?

Workstation setup

The workspace of Colab is the area where the two collaborative robots, Zack & Sarah, interact with the clinical microbiologists. We were inspired when playing a game of ping pong and used this ‘back and forth’ as a leading principle for the interaction between the microbiologist and Zack & Sarah. The area on the left side is the workstation where the clinical microbiologists interact with the collaborative robotics. The three different incubation areas offer space for storage of agar dishes.

 
Workstation.jpg
 

 

 

how do you work together?

Defining the human-robot interaction

As mentioned before, we envisioned the interact to be like a ping-pong principle. The clinical microbiologist prepares a tray, fills it with a sample, a number of agar dishes and a microscopic slide. Once this is done, the microbiologist presses a capacitive sensor, which tells Zack and Sarah to pick up the prepared task and start the inoculation process. The microbiologist has the possibility to decide which and how many agar plates he or she wants to use during this process and has the ability to personalise rather than standardise their process. While Zack and Sarah are inoculating the agar plates, the microbiologist can prepare the next tray, fill it up with agar plates, blood/urine/feces and a microscopic plate and press on the capacitive sensor. This tray will be placed in the queue and completed by Zack and Sarah when they are done with the first one. The process of inoculation will speed up tremendously, which allows a laboratory to increase their productivity and capabilities adequately.

 
 

 

 

your helping hands!

Collaborative robotics - Zack & Sarah

Zack & Sarah are the two collaborative robots that reduce the amount of repetition on a daily base for microbiologists in their laboratory. This duo has a friendly character, moves on a rail and has six degrees of axial movement. This creates a flawless, smooth and elegant movement of the robots. Zack and Sarah can pick up agar dishes, drip blood on the agar dishes and microscopic plates, inoculate (the process of spreading and streaking a liquid on the agar dish) and transport them to the incubation area. 

 
DaanHekking_Portfolio20186.jpg
 

 

 

aesthetics

Character design

When doing ethnographic research studies, we realised the interviewed microbiologists are sceptical towards implementing robots. Though, they understood robots could improve their daily life and make their job more enjoyable, there seemed to be a reluctancy in working with them. After synthesising the research results we focussed on creating a sense of trust between the microbiologist and the collaborative robot.

 
171217_Portfolio25.jpg
 

 

 

benefits of good design

Building trust through design

Based on our research, we found out that robots are often working in closed environments (industrial welding or assembly robots or at-home 3D-printers) due to its safety issues and predictable but incommunicable upcoming movements. We tried to build a sense of trust by communicating the upcoming movements of the robots and with flawless movements thanks to its six degrees of axial movement. The bubbles in the collaborative robot’s head, can show the status of the task and communicate where it will be going. It’s friendly and soft character, expresses a sense of calmness and its arms are welcoming you when Colab is activated. Though Zack and Sarah work in their designated space, the microbiologist can reach that space as well. Once he or she enters this space, Zack and Sarah will slowly move away from your hand and stay always 15 centimetres away.

 
Arm rotation.jpg
 

 

 

the 'storage area'

Incubation that improves time efficiency

As mentioned before, once an agar dish is inoculated, it’s transported to the incubation area. The function of an incubator is to offer an environment in which a bacteria or fungi can grow as quickly as possible. Colab has three incubators with different conditions/environments. Every agar plate that has been inoculated will be placed in one of the incubators. During our desk and field research, we found out that plate reading takes up to 25% of a microbiologists time. By implementing a system that simply takes photos of these agar plates on set intervals and compares the photos, growth can be identified. A microbiologist will be notified once there is enough bacterial growth. If there is not enough growth, the agar plate will be discarded and removed from the incubator. This way the microbiologist only looks at samples that have sufficient growth, rather than the ones without enough growth, and are ready for the next phase: susceptibility testing.

 
core77_colab5.jpg
DaanHekking_Portfolio20188.jpg
 

 

 

made for small, benefits for all

deployable and scalable rail system

Colab currently focusses on the inoculation and incubation process within clinical microbiology. We have designed this system in such a way it could be implemented in the other process within the laboratory. We envision the railway system and two collaborative robots to be implemented in the registration, bacteria identification, susceptibility testing and anaerobic workspaces as well. Rather than creating an all-in-one solution that requires a large investment (both financially and spatially), we created a solution which can be adopted by not only large-scale laboratories but also smaller ones.

 
171104_Behance1.jpg
DaanHekking_Portfolio20189.jpg
 

 

 

Don't you want to do what you have been educated for?

No one likes to do tasks they haven’t been educated for, especially when it takes years to become a professional microbiologist. The implementation of two collaborative robots are taking over tasks that do not require years of intense education, let alone a masters or PhD degree. As our research has shown, the profession of a microbiologist looks often different than what they see in their books. Zack and Sarah allows them to use the in-depth knowledge they have acquired more frequently and can therefore change the structure of their working day.