search

khuhm / deep-kidney-cancer

Deep Learning for End-to-End Kidney Cancer Diagnosis on Multi-Phase Abdominal Computed Tomography
MIT License
15 stars 0 forks source link

Need Your Guidance on my Project #5

Open MahendraPrajapati42 opened 8 months ago

MahendraPrajapati42 commented 8 months ago

Hi Khuhm, First of all , congratulation for getting second spot in KiTS23 and more power to you. I'm doing a research for AIIMS Delhi (Top Medical College and Hospital in India) on Kidney Cancer I've created models to classify Tumor/NonTumor, Melignant/Benign, RCC/Secondary and getting good accuracy However when i'm trying to classify RCC Clear and Non-Clear Cell, getting biased results yes i have imbalanced data but tried to compensate using augmentation.

However i didn't use Segmentation to classify for above models, luckily i came across your article and found interesting and definitely it can help me to increase my accuracy , Only challenge i have right now is, I don't have Segmentation Mask for each images, So i thought of using Trained models or other dataset to train a segmentation models

If you can help me with your process and collaborate together to complete this research would be very helpful.

Please let me know if you have some time to discuss.

Best regards, Mahendra

khuhm commented 8 months ago

Hi, Mahendra. Thank you for your interest in my project. It's nice to meet a researcher studying kidney cancer like me. After hearing about your situation, I wonder if it would be possible to learn the segmentation model using the KiTS23 segmentation masks, which are publicly available. Let me know if you have any other questions. You can also talk via email. khuhm@dali.korea.ac.kr

Best, Kwang-Hyun