During this period, the development of the epidemic across the country has affected people's hearts. In order to better cooperate with epidemic prevention and control, Fudan University has launched regular nucleic acid screening since early March.
This job requires counselors to check the screenshots of students' "Health Cloud" nucleic acid completion one by one to ensure that "no one is missed." It sounds simple but is difficult to do. Faced with dozens or even hundreds of highly repetitive screenshots, manual verification is often time-consuming and labor-intensive, and mistakes may be made accidentally.
In order to solve this problem, Li Xiaokang, a doctoral student in the School of Information Science and Engineering, wrote 130 lines of code and quickly developed a small program.
This small program has greatly improved the speed and accuracy of nucleic acid verification. It took several people to check 800 screenshots for more than an hour, but now it only takes 2 minutes to get the results.
## Source: Weibo screenshot
Subsequently, Li Xiaokang’s deeds became a hot topic, and the People’s Daily also praised him: “Knowledge It’s power!”
Once the idea struck, I just did it
According to the official introduction of Fudan University, Li Xiaokang is a doctoral student in the School of Information Science and Engineering, majoring in biomedical engineering. The research direction is medical imaging and artificial intelligence. In his daily scientific research life, he is often exposed to many image processing methods.
In addition, Li Xiaokang is also the counselor of the 2019 information class 1 of the college. After the school entered quasi-closed management, he also served as a volunteer.
Li Xiaokang is working as a volunteer (Source: Fudan University official account)
Faced with the daily tedious anti-epidemic work, thanks to the long-term Due to his scientific research habits and coding sensitivity, Li Xiaokang came up with the idea of writing a code program to automatically check nucleic acid completion screenshots.
Li Xiaokang’s chat with his academic and engineering colleagues (Source: Fudan University official account)
Just do what you say, on the evening of March 15th, Li Xiaokang It took Xiaokang more than an hour to get the initial code. As soon as the program was written, he verified the nucleic acid screenshot data of his class and found that the accuracy was very high, and even detected problems that had not been discovered during previous manual verification.
At the same time, the running time of the program is also very short. It only takes more than 20 seconds to process more than 80 pictures, which greatly improves the accuracy and speed of verification and reduces the workload of relevant personnel.
The program implementation principle is not complicated
How is such an efficient program implemented? Speaking of the principle, Li Xiaokang thinks it is not complicated.
He first thought of OCR (Optical Character Recognition, optical character recognition) technology, which can identify text in images and convert it into text information.
Since not all the information in the picture is useful, Li Xiaokang thought of regular expressions in Python, which can filter out the desired information from the text recognized by OCR.
The actual running process of the program (Source: Fudan University official account)
Based on the idea of OCR text recognition regular expression screening, Li Xiaokang developed The program implements filtering out the required information from the recognized text. After confirming the relevant information of each screenshot, output everyone's results to an Excel file to facilitate manual confirmation.
##Excel file output by the program (Source: Fudan University official account)
Apply what you learn and actively fight the epidemic
Talk about The original intention of developing the program, Li Xiaokang said that he just wanted to reduce the workload of himself and the teachers around him. Because the program is written in Python and the code comments are complete, those who know how to use Python can get started quickly.
At the same time, in order to facilitate the use of teachers who cannot program, Li Xiaokang also encapsulated the program so that it can be run by simply entering a line of code.
In Li Xiaokang’s original words: “Although the principle is very simple, anyone who can write code will understand what is going on at the first time, but if you don’t do related work, you won’t feel it. Since the matter is time-consuming and labor-intensive, I will naturally not come up with a solution. I just use the knowledge I have learned to solve the difficulties in practical work."
Fudan University officials revealed that the school information office has contacted Li Xiaokang. New mini programs are being developed based on demand. It is expected that in the near future, teachers and students will no longer have to manually collect nucleic acid screenshots. They can simply upload pictures directly through the mini program to view statistical results.
The above is the detailed content of 130 lines of Python code to complete nucleic acid statistics, solving one hour of manual work in two minutes!. For more information, please follow other related articles on the PHP Chinese website!
