What is HIDE, the government’s new AI-assisted Covid-19 early warning system?

A health worker puts a test tube into biohazard plastic after collecting a sample for Covid-19 testing in Jalan Pudu, Kuala Lumpur, January 18, 2021. — Picture by Shafwan Zaidon
A health worker puts a test tube into biohazard plastic after collecting a sample for Covid-19 testing in Jalan Pudu, Kuala Lumpur, January 18, 2021. — Picture by Shafwan Zaidon

Follow us on Instagram and subscribe to our Telegram channel for the latest updates.


PETALING JAYA, May 7 — As the number of Covid-19 cases in Malaysia continues to soar amid a worsening epidemic, the federal government recently launched a big data analytics tool that could predict the emergence of potential clusters through the use of artificial intelligence (AI).

Dubbed the Hotspot Identification by Dynamic Engagement (HIDE), the government said the system allows for risk assessment as well as forwards and backwards contact tracing of a particular hotspot, and ultimately acts as a pre-emptive measure to contain Covid-19.

However, many are still unclear as to the exact nature of this new warning system and its functionality in the current circumstances.

Malay Mail has prepared a brief explainer of the HIDE system through information available publicly.

What is the HIDE system?

Contrary to the connotation associated with its shortened acronym, the HIDE system is also predictive and not reactive in nature.

In a sense, the early warning system ― developed by the Crisis Preparedness and Response Centre (CPRC) under the Health Ministry ― is aimed at pre-emptively “identifying” a potential Covid-19 hotspot before a possible cluster outbreak is confirmed.

This meant that hotspots identified by HIDE have the potential to contribute to a cluster outbreak if proactive intervention is not taken.

According to Minister of Science, Technology and Innovation Khairy Jamaluddin, the HIDE system would allow the government to take a more precise and more transparent approach in containing the pandemic.

How the HIDE system works?

Khairy disclosed during a special media conference on May 4 that the HIDE system uses a few variables in its predictive analysis to identify potential hotspots ahead of time.

The variables include crowd density, space constraints, and air ventilation, which were used to predict whether a particular location could potentially be a Covid-19 hotspot.

However Khairy did not explain how HIDE would integrate with MySejahtera ― the federal government’s foremost contact tracing and vaccination app ― and it is most likely HIDE would cross-reference with the app’s QR code check-in data to simulate a prediction.

The above hypothesis appeared to be supported when Khairy urged business premises to immediately register with MySejahtera to provide QR code scanning for visitors to enable more accurate HIDE analysis.

To simulate an example scenario, HIDE’s analytical approach would perform comparison of places where multiple individuals have tested positive for Covid-19 through backward contact tracing and subsequently predict whether said location would be responsible for future cluster outbreaks.

It would also fully utilise the MySejahtera application where close contacts of a particular hotspots will also be notified following a successful prediction.

“By using this system, we will be able to identify potential hotspots ahead of time. These potential hotspots which are identified from HIDE will be publicised so that broad-based intervention can take place by the premise owners,” Khairy was quoted as saying in the special briefing.

According to Minister of Science, Technology and Innovation Khairy Jamaluddin, the HIDE system would allow the government to take a more precise and more transparent approach in containing the pandemic. — Bernama pic
According to Minister of Science, Technology and Innovation Khairy Jamaluddin, the HIDE system would allow the government to take a more precise and more transparent approach in containing the pandemic. — Bernama pic

So, when do individuals know their premises were potential hotspots?

According to Khairy, the government will publish a list of premises identified as potential Covid-19 hotspots under HIDE from May 7 to enable premise owners and the general public to take action to prevent emerging clusters in the predicted area.

As of May 3, Khairy said HIDE already identified 1,660 potential Covid-19 hotspots nationwide.

They included shopping malls, food and beverage outlets, supermarkets, bazaars, as well as office premises.

However, the government has assured those listed under HIDE that they would not necessarily be shuttered by authorities as this depended on risk evaluations by the Health Ministry or other relevant agencies.

Why the official launch now?

With daily Covid-19 cases hovering above the 2,500 mark, the government is hopeful that potential hotspots identified would spur broad-based proactive intervention to be taken by the premise owners to rectify the situation and ultimately contain any possible Covid-19 outbreak.

As part of his briefing, Khairy listed five preemptive measures that one must undertake once a particular premise is listed by HIDE as a potential Covid-19 hotspot.

The first two measures are screening workers at the affected premises, increasing gatekeeping measures to inspect the MySejahtera risk status at check-in and to refuse high risk individuals as well as HSO (Home Surveillance Order) breakers and perhaps also close contacts individuals.

The remaining three measures were to tighten crowd control, self regulate to lower the number of occupants at all times and categorising those who visited said premises as “casual contacts”.

For those flagged as “casual contacts”, they are advised to get themselves tested if they showed symptoms.

To summarise in a general layman sense, HIDE allows for early identification and intervention measures ― for both authorities and premise owners ― in times when community and workplace infections have contributed significantly to the increase in Covid-19 cases and clusters nationwide.

You May Also Like

Related Articles