Is it difficult to count nucleic acid results？ The operation of Fudan doctoral students is on fire.
Manual verification of nucleic acid reports is time-consuming, laborious and error-prone. What should I do?
On April 7th, Guan Wei of Fudan University issued a document. Since the beginning of March, the school has started the normalized nucleic acid screening work, requiring class counselors to check the students’ "healthy cloud" nucleic acid to complete screenshots to ensure that "no one is missed".
Li Xiaokang, a doctoral student in the School of Information Science and Engineering, is on fire. In a few minutes, he can quickly check the nucleic acids of hundreds of people to complete screenshots, greatly improving the efficiency and accuracy of verification!
From 1 hour to 2 minutes, a string of codes is done.
It is understood that Li Xiaokang is currently a counselor of Class 1 of Information in 2019.
Regarding the work of checking the nucleic acid report, he said: "It sounds simple, but when it is actually done, it may take half an hour to check the screenshot of a class. If it is a department with a large number of people, it may take longer, and it may be misread."
Li Xiaokang is volunteering.
As a doctoral student majoring in biomedical engineering, his research direction is medical imaging and artificial intelligence, and he usually comes into contact with many image processing methods. Li Xiaokang immediately thought of the OCR (Optical Character Recognition) technology he had learned before.
"OCR can recognize the text in the image and convert it into text information, which is convenient for verification. Moreover, because the nucleic acid screenshots are printed fonts, the recognition rate is very high, and it can be almost 100% accurate. "
He also thought of the regular expression in Python language — — You can search for specific pattern content in the string.
"Using regular expressions, you can filter out the desired information from the text recognized by OCR. Finally, after confirming the information such as the name, detection time and whether it has been sampled in each screenshot, output the results of everyone to an Excel file for manual confirmation. "
Finally, Li Xiaokang determined the program idea of "OCR character recognition+regular expression screening". On the evening of March 15th, it took him more than an hour to write the initial code, with a total of 130 lines, and it was found that it really worked, and it was very efficient.
Li Xiaokang’s chat with his colleagues when he first wrote the program.
As soon as the program was written, Li Xiaokang verified it on the nucleic acid screenshot data of his class, and the accuracy rate was really high, even detecting the problems that were not found by manual verification before. Moreover, the running time of the program is very short, and it only takes more than 20 seconds for more than 80 pictures.
Fudan University: It is expected to be covered by the whole school.
At present, the program has served in the college for 2 weeks. Li Xiaokang once asked other teachers to try their own programs for verification. 800 screenshots, it used to take more than an hour for several people to check, but now it only takes 2 minutes to get the results.
After he told this story to a circle of friends, many colleagues expressed great interest, and he also shared the code. In order to facilitate the use of teachers who can’t program, Li Xiaokang finally packaged the program — — When you use it, you just need to enter a line of code on the command line to run it.
Excel file output by the program
It is understood that the school information office also docked with Li Xiaokang. It is expected that in the near future, teachers and students will not need to manually collect screenshots of nucleic acids, but upload pictures directly through small programs, and they can check the statistical results at any time in the background.
This afternoon, when China Youth Daily reporter tried to get in touch with Li Xiaokang, he was serving as a volunteer in the school building to deliver meals.
He said: "The principle is also very simple, as long as people who can write code will understand what is going on at the first time, but those who don’t do related work will not feel the time-consuming and laborious work of this matter, and naturally they will not come up with a solution. I just use what I have learned to solve the difficulties in practical work. "