Saved in:
Bibliographic Details
Main Authors: Li, Yongcheng, Cai, Lingcong, Lu, Ying, Lin, Cheng, Zhang, Yupeng, Jiang, Jingyan, Dai, Genan, Zhang, Bowen, Cao, Jingzhou, Zhang, Xiangzhong, Fan, Xiaomao
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2408.07467
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910565420498944
author Li, Yongcheng
Cai, Lingcong
Lu, Ying
Lin, Cheng
Zhang, Yupeng
Jiang, Jingyan
Dai, Genan
Zhang, Bowen
Cao, Jingzhou
Zhang, Xiangzhong
Fan, Xiaomao
author_facet Li, Yongcheng
Cai, Lingcong
Lu, Ying
Lin, Cheng
Zhang, Yupeng
Jiang, Jingyan
Dai, Genan
Zhang, Bowen
Cao, Jingzhou
Zhang, Xiangzhong
Fan, Xiaomao
contents Accurate classification of blood cells is of vital significance in the diagnosis of hematological disorders. However, in real-world scenarios, domain shifts caused by the variability in laboratory procedures and settings, result in a rapid deterioration of the model's generalization performance. To address this issue, we propose a novel framework of domain-invariant representation learning (DoRL) via segment anything model (SAM) for blood cell classification. The DoRL comprises two main components: a LoRA-based SAM (LoRA-SAM) and a cross-domain autoencoder (CAE). The advantage of DoRL is that it can extract domain-invariant representations from various blood cell datasets in an unsupervised manner. Specifically, we first leverage the large-scale foundation model of SAM, fine-tuned with LoRA, to learn general image embeddings and segment blood cells. Additionally, we introduce CAE to learn domain-invariant representations across different-domain datasets while mitigating images' artifacts. To validate the effectiveness of domain-invariant representations, we employ five widely used machine learning classifiers to construct blood cell classification models. Experimental results on two public blood cell datasets and a private real dataset demonstrate that our proposed DoRL achieves a new state-of-the-art cross-domain performance, surpassing existing methods by a significant margin. The source code can be available at the URL (https://github.com/AnoK3111/DoRL).
format Preprint
id arxiv_https___arxiv_org_abs_2408_07467
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Domain-invariant Representation Learning via Segment Anything Model for Blood Cell Classification
Li, Yongcheng
Cai, Lingcong
Lu, Ying
Lin, Cheng
Zhang, Yupeng
Jiang, Jingyan
Dai, Genan
Zhang, Bowen
Cao, Jingzhou
Zhang, Xiangzhong
Fan, Xiaomao
Computer Vision and Pattern Recognition
Accurate classification of blood cells is of vital significance in the diagnosis of hematological disorders. However, in real-world scenarios, domain shifts caused by the variability in laboratory procedures and settings, result in a rapid deterioration of the model's generalization performance. To address this issue, we propose a novel framework of domain-invariant representation learning (DoRL) via segment anything model (SAM) for blood cell classification. The DoRL comprises two main components: a LoRA-based SAM (LoRA-SAM) and a cross-domain autoencoder (CAE). The advantage of DoRL is that it can extract domain-invariant representations from various blood cell datasets in an unsupervised manner. Specifically, we first leverage the large-scale foundation model of SAM, fine-tuned with LoRA, to learn general image embeddings and segment blood cells. Additionally, we introduce CAE to learn domain-invariant representations across different-domain datasets while mitigating images' artifacts. To validate the effectiveness of domain-invariant representations, we employ five widely used machine learning classifiers to construct blood cell classification models. Experimental results on two public blood cell datasets and a private real dataset demonstrate that our proposed DoRL achieves a new state-of-the-art cross-domain performance, surpassing existing methods by a significant margin. The source code can be available at the URL (https://github.com/AnoK3111/DoRL).
title Domain-invariant Representation Learning via Segment Anything Model for Blood Cell Classification
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2408.07467