Dr. Fudan wrote 130 lines of code and solved the cumbersome nucleic acid report verification in 2 minutes

Jin Lei is from Oufei Temple

Qubits | Official account QbitAI

2 minutes, "Snap".

More than 800 people completed the screenshot of nucleic acid and reviewed it.

This is the role of a small program developed by a doctoral student majoring in biomedical engineering at Fudan University during the recent joint fight against the epidemic.

And it's just the kind that takes 1 hour, 130 lines of code.

Fudan University's official evaluation of this "anti-epidemic weapon" is:

The efficiency and accuracy of nucleic acid verification have been greatly improved.

The work of this doctoral student has also attracted the "worship" of netizens:

The People's Daily also commented on his work, believing that this wave of "operations is on fire":

2 minutes to get 1 hour of work

The whole thing is that during the recent anti-epidemic period, Fudan University launched a normalized nucleic acid screening work.

And the "pain points" also followed.

That is, to verify that each student's "health cloud" nucleic acid is completed, it takes a lot of time and manpower.

But overall, it's a repetitive, monotonous, and boring job:

A screenshot of a class may take half an hour to verify, and if it is a large number of departments, it may take longer, and it may be misread and missed.

The doctoral student of Fudan, as the counselor of the college's 2019 class of information 1, was responsible for this work during this period.

So he had a plan——

"Make an OCR identification code".

△Source: Fudan University

When I first shared the idea with my fellow students, I was worried that it wouldn't be too difficult.

And he replied:

There is a ready-made library, just import it.

Then, 1 hour passed...

"I wrote it".

Colleagues even sighed and said, "Is this the doctor?"

As a result, his code program began to "take up" in his own class.

After the verification, the accuracy of this procedure is still very high, and even found problems that were not found in the previous manual verification.

Most importantly, the speed of this work has increased dramatically.

For example, it used to take more than 1 hour to check 800 screenshots, and it took more than 1 hour for several staff members, but now, the results can be obtained in 2 minutes!

As for the principle, the PhD student modestly said that it was "not complicated".

The technique he uses is mainly OCR (Optical Character Recognition), and the code language is Python.

According to fudan official introduction, this doctoral student is more specific to use regular expressions in Python.

Regular expressions filter the desired information from the text recognized by the OCR.

Finally, it will be summarized into an Excel file for easy confirmation by the staff.

And in order to facilitate the use of colleagues who do not program, he also encapsulated the program, and only needs to enter a line of commands to use.

PhD student from Fudan University

The doctoral student who developed this "anti-epidemic weapon" is Li Xiaokang from Fudan University.

Interestingly, he is not a computer science student, but a biomedical engineering major.

His research interests are medical imaging and artificial intelligence.

For this work, he argues:

Although the principle is also very simple, as long as the person who can write the code will understand what is going on at the first time, but not doing the relevant work does not feel the time-consuming and laborious of this matter, and naturally will not come up with a way.

I just use what I've learned to solve difficulties in real work.

△ Source: Fudan University, Li Xiaokang himself left a message

According to the official introduction of Fudan University, in the near future, teachers and students can no longer manually collect nucleic acid screenshots, but directly upload pictures through a small program.

……

Finally, to quote Li Xiaokang himself:

We can definitely win this battle "epidemic"!

Reference Links:

[1]https://mp.weixin.qq.com/s/RogQcUAsZszW5HkYwYcV-in

[2]https://mp.weixin.qq.com/s/l8u9JifKDlRDoz32-jZWQg

[3]https://weibo.com/1726918143/Lnn2Ll7KZ?type=comment#_rnd1649380649145