IEEE Robotics & Automation Magazine - March 2021 - 31

The robot has collected 80 throat swab samples since it was
put into operation in the Guangzhou Institute of Respiratory Health on 28 February 2020 and showed improved
speed (9 versus 14 s) and comfort for comparable data
quality [25].
In Denmark, researchers from Syddansk Universitet
developed a throat-swabbing robot [Figure 3(b)], which
led them to set up a commercialization company [21],
[24]. In South Korea, researchers at the South Korean
Institute of Machinery and Materials and the Daegu Convergence Technology Research Center created a swabbing
robot [20]-[24], [71]. Similar devices were reported by a
Taiwanese company [27] and researchers in China [26]
without further data.
Automated Drive-In Sampling
As testing volumes have increased, discussion has moved
to establishing drive-in testing facilities for population sampling (although these are, strictly speaking,
sample-collection rather than testing stations). Two further projects, both in Germany, have targeted the automation of drive-in testing: one in the Hamburg area by
Freise Automation [Figure 4(a)] and the other by BoKa
Automatisierung in Bavaria [Figure 4(b)]. These add a
booth, swab stock, and sample management [22], [23].
These act as automated sample handlers for swabbing by
health-care workers or test subjects, with the main goal of
increasing test capacity (12/h) while preventing cross-contamination. Considerable care has been paid to subject
identification and the transfer of results-important matters that have been a weakness in conventional systems as
well as for the economic case. However, the designs are prototypes awaiting clinical validation and certification. In
parallel, there have been efforts to develop alternative sampling approaches that make use of, e.g., saliva. As often
happens in the history of robotics, efforts to automate one
procedure risk being displaced by changes in the process.
Close coordination between users and technologists is
required to address this.
Sample Storage and Biobanking
Once a sample has been analyzed and the diagnostic result
obtained, it is normal to archive the sample in a coldstorage facility. This is demanded in case reanalysis is
required. With the rise of genomic techniques, access to
such samples has become an invaluable resource for gaining new insight into the relationships among the disease,
genetics, and other factors, often within collaborations
between clinical and scientific teams. Being able to track
and recover large numbers of such samples has led to the
burgeoning field of biobanking and the development of a
range of robotic tools to meet this need [7], [8], [28].
Thanks to such biobanks, it has been possible to create
detailed maps of the spread of SARS-CoV-2 that exploit
small changes in the virus genome over time, detect entry
points, and assess mitigation [29].

New Laboratory Concepts-OpenCell
The traditional approach to clinical testing has involved a
central hub laboratory that can afford to install the most
modern high-capacity machines, which are run with a high
load by a critical mass of skilled staff. Although this is efficient, it can lead to some challenges in the sample logistics,
tracking, and reporting of results. During the pandemic,
where geographically dispersed populations are to be
screened, there is some advantage in bringing the laboratory
closer to the sampling site.
One innovative approach proposed has been for the use of
modular labs in containers, an idea pioneered by OpenCell.

(a)

(b)
Figure 4. Robotized drive-in sampling stations developed
by (a) Freise and (b) BoKa. [Sources: (a) W. Kazemi, Freise
Automation and (b) C. Hartung, BoKa Automatisierung; used
with permission.]

RNA
Accessioning Extraction

RT-PCR

Figure 5. The OpenCell lab in a container. (Source: OpenCell
Biomedical; used with permission.)

MARCH 2021

*

IEEE ROBOTICS & AUTOMATION MAGAZINE

*

31
IEEE Robotics & Automation Magazine - March 2021

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - March 2021
