The Rapid Testing procedure is one of the first steps undertaken by
the Government of Jakarta in reducing the spread of the COVID-19
pandemic. However, throughout its implementation, several challenges
have arisen, including determining who has the priority to take the
test.
So, to optimize the use of a limited number of test kits,
the Provincial Government is collaborating with Harvard CLM Team. From
this collaboration, the latest innovation called COVID-19 Likelihood
Meter (CLM) - an independent check application that is supported by
machine learning-based technology.
How Harvard CLM Team Contributes to Pandemic Mitigation
As the name implies, Harvard CLM Team consist of Indonesian
students from Harvard University, United States. Now, they return,
voluntarily participate and contribute in the effort to mitigate the
coronavirus in Indonesia, especially Jakarta.
"Each of us feels burdened and wants to help our own
country to deal with this pandemic. We also realize we have more
opportunities - that we can study at Harvard, so we also want to use
what we learn to help our own country, "said Jessica Wijaya, who is
currently pursuing a Master of Science in Data Science program.
"I feel, from what I've learned in Data Science, to make
predictions from machine learning, that it's really important and the
impact is quite big in tackling this pandemic in this country," she
added.
Meanwhile, Dr. Nanda Lucky Prasetya is a Master of Medical
Science (MMSc) from Harvard Medical School in 2019. Late last March, he
led the Indonesians at Harvard team in a meeting with the Governor of
Jakarta, Anies Baswedan, to pitch the concept of Data Science in
determining Rapid Test priority.
Dr. Nanda, who also has a high interest in mathematics,
appreciates the open attitude and commitment of the provincial
government in handling pandemics with a scientific approach.
"Initiatives that are more directed towards epidemiology
and something innovative is really welcomed by the Government of
Jakarta. So I can appreciate their efforts in overcoming COVID-19, and
have a good trust in the Provincial Government," he said about the
meeting.
(Picture: Indonesians at Harvard)
Why does Jakarta Need CLM?
During this coronavirus outbreak, there must be some of you
who have tried telemedicine and their self-assessment services to find
out if your symptoms are similar to COVID-19. But did you know that this
service is not efficient enough in filtering who is eligible to take
the rapid test?
Therefore, through collaboration with Jakarta Smart City,
Health Department of Jakarta, and Klakklik.id, Harvard CLM Team then
developed CLM with several advantages in it.
"First, compared to other telemedicine platforms in
general, the calculation of whether or not people need to take a test
through CLM has a scientific basis. Some telemedicine only does
calculations with a less strong basis. For example, if symptom A is
given a score of 4, symptom B is given a score of 2, then if passing a
score of 10 means it must be tested. CLM is not like that, the machine
learning model is made from previous real data owned by the Health
Office of Jakarta so that the accuracy of calculation will be more
scientific and precise," explained Dr. Nanda.
"So this CLM has a database or like a scientific backup to
calculate whether this patient is positive or negative. Second, we can
predict how many people are positive from the data we send.
"Third, this is a comprehensive self-checker. So this
doesn't just categorize people as OTG, ODP, PDP, or just give people
advice. But it also gives them a score and details about facilities
where they can take rapid tests because we are working with the Health
Department.
"Fourth, CLM is dynamic. The more data, the more precise
and accurate it will be. I believe the people of Jakarta with their good
social values will answer the questions honestly so that CLM can
continue to be developed when new data comes in. And also so that the
CLM model can become a true form of from and for the community."
Honesty is Key in Recovery
In the near future, you can use the CLM-based Kalkulator
COVID-19 application through JAKI. Dr Nanda said that one of the keys to
the success of this program will depend on the honesty of the user when
answering questions. Because in addition to improving the efficiency of
the rapid test implementation, he believes the accuracy of the data
from the CLM will also help the Health Office of Jakarta in conducting
better contact tracing.
This was echoed by Jessica. She believes if every user can
use the application wisely, then it will indirectly help the recovery
process in the capital.
"For those who will use CLM, I hope everyone will give
honest answers. Do not deliberately exacerbate symptoms just to increase
the chances of getting a test. Because by helping other communities
around us, it will actually benefit ourselves," she concluded.