Project
Customer churn prediction and breast cancer classification projects
Customer churn prediction (C5.0, ANN) and breast cancer classification projects. Solved 300+ problems on Codeforces.
Md Sanaul Haque 201__69Green University of Bangladesh
Md Sanaul Haque 201__69
Executive - Business Analyst
Green University of Bangladesh
Project
Customer churn prediction (C5.0, ANN) and breast cancer classification projects. Solved 300+ problems on Codeforces.
Project
FasNet, a deep learning model for liver tumor segmentation. Accepted at the 6th International Conference on Sustainable Technologies for Industry 5.0 and published in IEEE Xplore.
Liver tumors remain a major challenge in medical imaging due to their high fatality rate and rapid progression. This study introduces FasNet, a hybrid deep learning model developed to improve the accuracy of liver tumor segmentation from CT scans. FasNet combines the strengths of VGG-16 (for fine-grained feature extraction) and ResNet-50 (for residual learning) and integrates 3D convolutional layers to better capture spatial and contextual information from volumetric liver data.
The model was evaluated on a large liver imaging dataset and achieved high performance with a Dice score of 0.854, precision of 85.1%, recall of 81%, and an F1-score of 99.5%. These results highlight FasNet’s strong potential for clinical use in improving diagnostic accuracy and treatment planning for liver disease. Published in IEEE Xplore, this work represents a significant step forward in applying AI to real-world healthcare challenges.