An Improved Multi-Task Learning Approach with Applications in Medical Diagnosis
Abstract
We propose a family of multi-task learning algorithms for collaborative computer aided diagnosis which aims to diagnose multiple clinically-related abnormal structures from medical images. Our formulations eliminate features irrelevant to all tasks, and identify discriminative features for each of the tasks. A probabilistic model is derived to justify the proposed learning formulations. By equivalence proof, some existing regularization-based methods can also be interpreted by our probabilistic model as imposing a Wishart hyperprior. Convergence analysis highlights the conditions under which the formulations achieve convexity and global convergence. Two real-world medical problems: lung cancer prognosis and heart wall motion analysis, are used to validate the proposed algorithms.
Cite
Text
Bi et al. "An Improved Multi-Task Learning Approach with Applications in Medical Diagnosis." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2008. doi:10.1007/978-3-540-87479-9_26Markdown
[Bi et al. "An Improved Multi-Task Learning Approach with Applications in Medical Diagnosis." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2008.](https://mlanthology.org/ecmlpkdd/2008/bi2008ecmlpkdd-improved/) doi:10.1007/978-3-540-87479-9_26BibTeX
@inproceedings{bi2008ecmlpkdd-improved,
title = {{An Improved Multi-Task Learning Approach with Applications in Medical Diagnosis}},
author = {Bi, Jinbo and Xiong, Tao and Yu, Shipeng and Dundar, Murat and Rao, R. Bharat},
booktitle = {European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases},
year = {2008},
pages = {117-132},
doi = {10.1007/978-3-540-87479-9_26},
url = {https://mlanthology.org/ecmlpkdd/2008/bi2008ecmlpkdd-improved/}
}