Could robots overtake humans in healthcare testing, reducing time and boosting capacity? Promising new research reveals a 65% increase in capacity

Professor Andrea Zanchettin, Politecnico di Milano tried a new approach to monitoring Covid-19 – by developing the YuMi robot to save time and boost capacity in COVID-19 testing.

Due to the high demand for coronavirus testing and a backlog of laboratory work, expert in the field of collaborative robotics, Dr Andrea Zanchettin developed an application based on “YuMi”, a two-armed, seven-axis “cobot” first introduced by ABB in 2015, which automates serological testing.

Dr Zanchettin is an Associate Professor at the Department of Electronics, Information and Bioengineering of the Politecnico di Milano, Italy, who – together with scientists of the European Institute of Oncology (IEO), adapted the serological testing protocol for COVID-19 first created at Mount Sinai Hospital in New York.

Back at the start of the pandemic in March, he thought back at ways in which his team could provide support, as they did for instance in 2012, with the use of a collaborative robot to manipulate pharmaceutical instruments such as syringes and pipettes.

By analysing the time and the amount of pipettes required to improve the serological analysis of Covid-19, Dr Zanchettin realised that the washing phase employed 2/3 of the total task. Despite the lack of public funding, he went on with his plan and started looking for a way to save time and improve ergonomics by programming a robot to automate the well place pipetting.

This was his best option, as the alternative would have required a laboratory technician to operate the piston of the micropipette eight times for each serological test performed on a single patient. This would have implied a considerate amount of stress for the operators as well as for their thumbs, that were at risk of an inflammation of the tendon.

Dr Zanchettin was very pleased with the results of the new procedure using YuMi: in nominal conditions, the robot can automate up to 77% of the testing actions to ensure the analysis of up to 450 samples an hour. Despite being a robot, it is no faster than a human.

Whilst the serological trial was not as balanced as the Professor expected, as it sped up the process by 65%, it still represents a great increase in test numbers.


YuMi: the design

YuMi was supplied to Dr Zanchettin by ABB as part of a long-term strategic partnership with Politecnico di Milano. He already had experience with the robot which was regulated to ensure safety and prevent contamination. Dr Zanchettin and his team could also rely on RobotStudio which is ABB’s offline programming tool that allowed them to work with a virtual robot in a simulation environment, saving a lot of time in this way by working from home.

YuMi is activated when the excess of the patient’s serum must be cleaned out of the plate: the robot collects the plate and moves in into position, then draws the cleaning solution from a reservoir and fills the plate. It then draws the solution back out of each well and discards it, which takes three minutes to complete.

Among the challenges Dr Zanchettin and his team had to face, they had difficulty in replicating a physical task such as pipetting, which has a return spring which is hard to simulate. Once he was back in the lab, he came to the realization that he needed to use a pneumatic piston mounted on the pipette to get one of the robotic arms to hold the test tube and the other pressing the pipette. Moreover, due to the circumstances of the past months, Dr Zanchettin struggled to access the necessary material for the test, and had to source some from a 3D printing specialist in Belgium, whilst others even from Japan.

Dr Zanchettin suggested that in order to make an impact on Covid-19 testing, efforts should be directed to smaller regional hospitals, which may lack lab technicians to take tests. This is where collaborative robotics and the real value of YuMi can be extremely helpful in upgrading testing. Today, the biggest obstacle to progress is represented by the people themselves, who don’t see the benefit in automation yet.