Hi Kagglers,
\nWe’re happy to announce the conclusion of the Mayo Clinic - STRIP AI competition. Thank you everyone for your participation which helps a good cause of this!
\nThis competition ended with 10,367 registrations and 1,025 participants on 888 teams. We had 6,980 submissions from 81 countries. For 168 users (including 14 in the top 100!), this was their first competition. We also had 3 new Masters at the conclusion of the competition! Thank you all for your hard work in this competition and congratulations to our winners and to those who gained a new ranking!
\nThis competition is challenging with its comparatively smaller public test set and limited signals. However, we are pleased to see the winning solutions and substantial great notebooks produced for working with these Whole Slide images. Most importantly, many Kagglers mentioned they had learned a lot from this competition.
\nThe top potential winning teams have been contacted via email for the next steps. We look forward to learning more about their winning solutions.
\nYou should have already seen your points and medals awarded, however it may take a few hours to propagate through if yours are still missing.
\nHere is a curated list of your work. Should you be looking for takeaways in the future, you have a place to see what surfaced from the work performed.
\nHappy Modeling!
\nKaggle Team
1st Place
\n2nd Place
\n3rd Place
\n5th Place
\n6th Place
\n7th Place
\n13th Place
\n25th Place
EDA, images processing and exploration
\nSTRIP AI - Exploratory Data Analysis
\n[Train + Infer] CoAtNet + EfficientNet
\nCNN STRIP AI [inference]
\nBloodClots🔬: EDA🔎 & load WSI + prune background
\nA fast tile generation
\nMayo Clinic - Removing Background via Seam Carving
\n[MAYO] Simple CNN Starter (Eng/日本語）
\nMayo Clinic STRIP AI - Can I trust my model?
Image Classification Checklist
\nQuestion to organizers - are we sure there is signal in the data?
\nHow I Deal With WSI (Whole Slide Images aka. RAM Crashingly Large Images)
\nMayo Clinic Sliced 1024x1024 JPG Datasets
\nI became the first person to cross the sample submission.csv
\n[placeholder] transformer MIL for WSI
\nMayo Tile Images Dataset [size=1024, N=16]
\nPapers on Image Classification of Stroke Blood 🩸🧐
Hi Kagglers,
\nWe’re happy to announce the conclusion of the Mayo Clinic - STRIP AI competition. Thank you everyone for your participation which helps a good cause of this!
\nThis competition ended with 10,367 registrations and 1,025 participants on 888 teams. We had 6,980 submissions from 81 countries. For 168 users (including 14 in the top 100!), this was their first competition. We also had 3 new Masters at the conclusion of the competition! Thank you all for your hard work in this competition and congratulations to our winners and to those who gained a new ranking!
\nThis competition is challenging with its comparatively smaller public test set and limited signals. However, we are pleased to see the winning solutions and substantial great notebooks produced for working with these Whole Slide images. Most importantly, many Kagglers mentioned they had learned a lot from this competition.
\nThe top potential winning teams have been contacted via email for the next steps. We look forward to learning more about their winning solutions.
\nYou should have already seen your points and medals awarded, however it may take a few hours to propagate through if yours are still missing.
\nHere is a curated list of your work. Should you be looking for takeaways in the future, you have a place to see what surfaced from the work performed.
\nHappy Modeling!
\nKaggle Team
1st Place
\n2nd Place
\n3rd Place
\n5th Place
\n6th Place
\n7th Place
\n13th Place
\n25th Place
EDA, images processing and exploration
\nSTRIP AI - Exploratory Data Analysis
\n[Train + Infer] CoAtNet + EfficientNet
\nCNN STRIP AI [inference]
\nBloodClots🔬: EDA🔎 & load WSI + prune background
\nA fast tile generation
\nMayo Clinic - Removing Background via Seam Carving
\n[MAYO] Simple CNN Starter (Eng/日本語）
\nMayo Clinic STRIP AI - Can I trust my model?
Image Classification Checklist
\nQuestion to organizers - are we sure there is signal in the data?
\nHow I Deal With WSI (Whole Slide Images aka. RAM Crashingly Large Images)
\nMayo Clinic Sliced 1024x1024 JPG Datasets
\nI became the first person to cross the sample submission.csv
\n[placeholder] transformer MIL for WSI
\nMayo Tile Images Dataset [size=1024, N=16]
\nPapers on Image Classification of Stroke Blood 🩸🧐