SIIM-FISABIO-RSNA COVID-19 Detection - Identify and localize COVID-19 abnormalities on chest radiographs
Finally, after 3 months of brainstorming and innovating, we've come to the end of the Kaggle competition "SIIM-FISABIO-RSNA COVID-19 Detection" which was based on Identifying and localizing COVID-19 abnormalities on chest radiographs. The Kaggle competition "SIIM-FISABIO-RSNA COVID-19 Detection" was completed with 3 months of knowledge and engineering effort. The COVID-19 problem is a very important problem nowadays. Currently, COVID-19 can be diagnosed via polymerase chain reaction to detect genetic material from the virus or chest radiograph. However, it can take a few hours and sometimes days before the molecular test results are back. By contrast, chest radiographs can be obtained in minutes. While guidelines exist to help radiologists differentiate COVID-19 from other types of infection, their assessments vary. COVID-19 looks very similar to other viral and bacterial pneumonias on chest radiographs, which makes it difficult to diagnose. We constructed the model using efficientnet, efficientnet-v2, YOLOv5, cascade r-cnn and centernet-v2. ๐ I am happy to be able to win the silver medal in this important and difficult competition. Thank you to hwigeon, seongchae, WoongBae Yoon and Heonjin Ha for the results we got from our team "[Kor]COVID-solver" members so well done. ๐๐๐ It wouldn't have been possible without my Z by HP Z4 workstation. Looking forward to competing in more such competitions! ๐ฐ - Chanran Kim