Estimating heterogeneous treatment effect by balancing heterogeneity and fitness
Abstract Background Estimating heterogeneous treatment effect is a fundamental problem in biological and medical applications. Recently, several recursive partitioning methods have been proposed to identify the subgroups that respond differently towards a treatment, and they rely on a fitness criter...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2018-12-01
|
| Series: | BMC Bioinformatics |
| Subjects: | |
| Online Access: | http://link.springer.com/article/10.1186/s12859-018-2521-7 |
| _version_ | 1819289544806105088 |
|---|---|
| author | Weijia Zhang Thuc Duy Le Lin Liu Jiuyong Li |
| author_facet | Weijia Zhang Thuc Duy Le Lin Liu Jiuyong Li |
| author_sort | Weijia Zhang |
| collection | DOAJ |
| description | Abstract Background Estimating heterogeneous treatment effect is a fundamental problem in biological and medical applications. Recently, several recursive partitioning methods have been proposed to identify the subgroups that respond differently towards a treatment, and they rely on a fitness criterion to minimize the error between the estimated treatment effects and the unobservable ground truths. Results In this paper, we propose that a heterogeneity criterion, which maximizes the differences of treatment effects among the subgroups, also needs to be considered. Moreover, we show that better performances can be achieved when the fitness and the heterogeneous criteria are considered simultaneously. Selecting the optimal splitting points then becomes a multi-objective problem; however, a solution that achieves optimal in both aspects are often not available. To solve this problem, we propose a multi-objective splitting procedure to balance both criteria. The proposed procedure is computationally efficient and fits naturally into the existing recursive partitioning framework. Experimental results show that the proposed multi-objective approach performs consistently better than existing ones. Conclusion Heterogeneity should be considered with fitness in heterogeneous treatment effect estimation, and the proposed multi-objective splitting procedure achieves the best performance by balancing both criteria. |
| first_indexed | 2024-12-24T03:08:32Z |
| format | Article |
| id | doaj.art-3ca28185012846c0befdeb7e36f10e88 |
| institution | Directory Open Access Journal |
| issn | 1471-2105 |
| language | English |
| last_indexed | 2024-12-24T03:08:32Z |
| publishDate | 2018-12-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Bioinformatics |
| spelling | doaj.art-3ca28185012846c0befdeb7e36f10e882022-12-21T17:17:55ZengBMCBMC Bioinformatics1471-21052018-12-0119S19617210.1186/s12859-018-2521-7Estimating heterogeneous treatment effect by balancing heterogeneity and fitnessWeijia Zhang0Thuc Duy Le1Lin Liu2Jiuyong Li3School of Information Technology and Mathematical Sciences, University of South AustraliaSchool of Information Technology and Mathematical Sciences, University of South AustraliaSchool of Information Technology and Mathematical Sciences, University of South AustraliaSchool of Information Technology and Mathematical Sciences, University of South AustraliaAbstract Background Estimating heterogeneous treatment effect is a fundamental problem in biological and medical applications. Recently, several recursive partitioning methods have been proposed to identify the subgroups that respond differently towards a treatment, and they rely on a fitness criterion to minimize the error between the estimated treatment effects and the unobservable ground truths. Results In this paper, we propose that a heterogeneity criterion, which maximizes the differences of treatment effects among the subgroups, also needs to be considered. Moreover, we show that better performances can be achieved when the fitness and the heterogeneous criteria are considered simultaneously. Selecting the optimal splitting points then becomes a multi-objective problem; however, a solution that achieves optimal in both aspects are often not available. To solve this problem, we propose a multi-objective splitting procedure to balance both criteria. The proposed procedure is computationally efficient and fits naturally into the existing recursive partitioning framework. Experimental results show that the proposed multi-objective approach performs consistently better than existing ones. Conclusion Heterogeneity should be considered with fitness in heterogeneous treatment effect estimation, and the proposed multi-objective splitting procedure achieves the best performance by balancing both criteria.http://link.springer.com/article/10.1186/s12859-018-2521-7Heterogeneous treatment effectBreast cancerRadiotherapy |
| spellingShingle | Weijia Zhang Thuc Duy Le Lin Liu Jiuyong Li Estimating heterogeneous treatment effect by balancing heterogeneity and fitness BMC Bioinformatics Heterogeneous treatment effect Breast cancer Radiotherapy |
| title | Estimating heterogeneous treatment effect by balancing heterogeneity and fitness |
| title_full | Estimating heterogeneous treatment effect by balancing heterogeneity and fitness |
| title_fullStr | Estimating heterogeneous treatment effect by balancing heterogeneity and fitness |
| title_full_unstemmed | Estimating heterogeneous treatment effect by balancing heterogeneity and fitness |
| title_short | Estimating heterogeneous treatment effect by balancing heterogeneity and fitness |
| title_sort | estimating heterogeneous treatment effect by balancing heterogeneity and fitness |
| topic | Heterogeneous treatment effect Breast cancer Radiotherapy |
| url | http://link.springer.com/article/10.1186/s12859-018-2521-7 |
| work_keys_str_mv | AT weijiazhang estimatingheterogeneoustreatmenteffectbybalancingheterogeneityandfitness AT thucduyle estimatingheterogeneoustreatmenteffectbybalancingheterogeneityandfitness AT linliu estimatingheterogeneoustreatmenteffectbybalancingheterogeneityandfitness AT jiuyongli estimatingheterogeneoustreatmenteffectbybalancingheterogeneityandfitness |