TY - GEN
T1 - DF-SegDiff
T2 - 2024 IEEE International Conference on Cognitive Computing and Complex Data, ICCD 2024
AU - Mi, Hancang
AU - Gan, Hong Seng
AU - Wang, Xiaoyi
AU - Shimizu, Akinobu
AU - Ramlee, Muhammad Hanif
AU - Zubeyir Unlu, Mehmet
N1 - Publisher Copyright:
© 2024 IEEE.
PY - 2024
Y1 - 2024
N2 - Brain tumours are among the most life-threatening diseases, and automatic segmentation of brain tumours from medical images is crucial for clinicians to identify and quantify tumour regions with high precision. While traditional segmentation models have laid the groundwork, diffusion models have since been developed to better manage complex medical data. However, diffusion models often face challenges related to insufficient parallel computing power and inefficient GPU utilization. To address these issues, we propose the DF-SegDiff model, which includes diffusion segmentation, parallel data processing, a distributed training model, a dynamic balancing parameter and model fusion. This approach significantly reduces training time while achieving an average Dice score of 0.87, with several samples reaching Dice values close to 0.94. By combining BRATS2020 with the Medical Segmentation Decathlon dataset, we also integrated a comprehensive dataset containing 800 training samples and 53 test samples. Evaluation of the model using Dice, IoU, and other relevant metrics demonstrates that our method outperforms current state-of-the-art techniques.
AB - Brain tumours are among the most life-threatening diseases, and automatic segmentation of brain tumours from medical images is crucial for clinicians to identify and quantify tumour regions with high precision. While traditional segmentation models have laid the groundwork, diffusion models have since been developed to better manage complex medical data. However, diffusion models often face challenges related to insufficient parallel computing power and inefficient GPU utilization. To address these issues, we propose the DF-SegDiff model, which includes diffusion segmentation, parallel data processing, a distributed training model, a dynamic balancing parameter and model fusion. This approach significantly reduces training time while achieving an average Dice score of 0.87, with several samples reaching Dice values close to 0.94. By combining BRATS2020 with the Medical Segmentation Decathlon dataset, we also integrated a comprehensive dataset containing 800 training samples and 53 test samples. Evaluation of the model using Dice, IoU, and other relevant metrics demonstrates that our method outperforms current state-of-the-art techniques.
KW - Brain Tumour
KW - Diffusion
KW - Distributed Algorithms
KW - Segmentation
UR - http://www.scopus.com/inward/record.url?scp=85218089593&partnerID=8YFLogxK
U2 - 10.1109/ICCD62811.2024.10843429
DO - 10.1109/ICCD62811.2024.10843429
M3 - Conference Proceeding
AN - SCOPUS:85218089593
T3 - 2024 IEEE International Conference on Cognitive Computing and Complex Data, ICCD 2024
SP - 13
EP - 17
BT - 2024 IEEE International Conference on Cognitive Computing and Complex Data, ICCD 2024
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 28 September 2024 through 30 September 2024
ER -