Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2510.16421 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912657595957248 |
|---|---|
| author | Yu, Baichen Liu, Jin Wang, Hansheng |
| author_facet | Yu, Baichen Liu, Jin Wang, Hansheng |
| contents | We develop here a semiparametric Gaussian mixture model (SGMM) for unsupervised learning with valuable spatial information taken into consideration. Specifically, we assume for each instance a random location. Then, conditional on this random location, we assume for the feature vector a standard Gaussian mixture model (GMM). The proposed SGMM allows the mixing probability to be nonparametrically related to the spatial location. Compared with a classical GMM, SGMM is considerably more flexible and allows the instances from the same class to be spatially clustered. To estimate the SGMM, novel EM algorithms are developed and rigorous asymptotic theories are established. Extensive numerical simulations are conducted to demonstrate our finite sample performance. For a real application, we apply our SGMM method to the CAMELYON16 dataset of whole-slide images (WSIs) for breast cancer detection. The SGMM method demonstrates outstanding clustering performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_16421 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | A Semiparametric Gaussian Mixture Model with Spatial Dependence and Its Application to Whole-Slide Image Clustering Analysis Yu, Baichen Liu, Jin Wang, Hansheng Methodology We develop here a semiparametric Gaussian mixture model (SGMM) for unsupervised learning with valuable spatial information taken into consideration. Specifically, we assume for each instance a random location. Then, conditional on this random location, we assume for the feature vector a standard Gaussian mixture model (GMM). The proposed SGMM allows the mixing probability to be nonparametrically related to the spatial location. Compared with a classical GMM, SGMM is considerably more flexible and allows the instances from the same class to be spatially clustered. To estimate the SGMM, novel EM algorithms are developed and rigorous asymptotic theories are established. Extensive numerical simulations are conducted to demonstrate our finite sample performance. For a real application, we apply our SGMM method to the CAMELYON16 dataset of whole-slide images (WSIs) for breast cancer detection. The SGMM method demonstrates outstanding clustering performance. |
| title | A Semiparametric Gaussian Mixture Model with Spatial Dependence and Its Application to Whole-Slide Image Clustering Analysis |
| topic | Methodology |
| url | https://arxiv.org/abs/2510.16421 |