Salvato in:
Dettagli Bibliografici
Autori principali: Li, Jiawen, Guan, Tian, Xia, Qingxin, Wang, Yizhi, Ling, Xitong, Li, Jing, Huang, Qiang, Wang, Zihan, Shen, Zhiyuan, Ma, Yifei, Zhao, Zimo, Lei, Zhe, Chen, Tiandong, Tan, Junbo, Wang, Xueqian, Bian, Xiu-Wu, Wang, Zhe, Guo, Lingchuan, He, Chao, He, Yonghong
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2412.20430
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913628921266176
author Li, Jiawen
Guan, Tian
Xia, Qingxin
Wang, Yizhi
Ling, Xitong
Li, Jing
Huang, Qiang
Wang, Zihan
Shen, Zhiyuan
Ma, Yifei
Zhao, Zimo
Lei, Zhe
Chen, Tiandong
Tan, Junbo
Wang, Xueqian
Bian, Xiu-Wu
Wang, Zhe
Guo, Lingchuan
He, Chao
He, Yonghong
author_facet Li, Jiawen
Guan, Tian
Xia, Qingxin
Wang, Yizhi
Ling, Xitong
Li, Jing
Huang, Qiang
Wang, Zihan
Shen, Zhiyuan
Ma, Yifei
Zhao, Zimo
Lei, Zhe
Chen, Tiandong
Tan, Junbo
Wang, Xueqian
Bian, Xiu-Wu
Wang, Zhe
Guo, Lingchuan
He, Chao
He, Yonghong
contents Foundation models have revolutionized the paradigm of digital pathology, as they leverage general-purpose features to emulate real-world pathological practices, enabling the quantitative analysis of critical histological patterns and the dissection of cancer-specific signals. However, these static general features constrain the flexibility and pathological relevance in the ever-evolving needs of clinical applications, hindering the broad use of the current models. Here we introduce PathFiT, a dynamic feature learning method that can be effortlessly plugged into various pathology foundation models to unlock their adaptability. Meanwhile, PathFiT performs seamless implementation across diverse pathology applications regardless of downstream specificity. To validate PathFiT, we construct a digital pathology benchmark with over 20 terabytes of Internet and real-world data comprising 28 H\&E-stained tasks and 7 specialized imaging tasks including Masson's Trichrome staining and immunofluorescence images. By applying PathFiT to the representative pathology foundation models, we demonstrate state-of-the-art performance on 34 out of 35 tasks, with significant improvements on 23 tasks and outperforming by 10.20% on specialized imaging tasks. The superior performance and versatility of PathFiT open up new avenues in computational pathology.
format Preprint
id arxiv_https___arxiv_org_abs_2412_20430
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Unlocking adaptive digital pathology through dynamic feature learning
Li, Jiawen
Guan, Tian
Xia, Qingxin
Wang, Yizhi
Ling, Xitong
Li, Jing
Huang, Qiang
Wang, Zihan
Shen, Zhiyuan
Ma, Yifei
Zhao, Zimo
Lei, Zhe
Chen, Tiandong
Tan, Junbo
Wang, Xueqian
Bian, Xiu-Wu
Wang, Zhe
Guo, Lingchuan
He, Chao
He, Yonghong
Image and Video Processing
Computer Vision and Pattern Recognition
Foundation models have revolutionized the paradigm of digital pathology, as they leverage general-purpose features to emulate real-world pathological practices, enabling the quantitative analysis of critical histological patterns and the dissection of cancer-specific signals. However, these static general features constrain the flexibility and pathological relevance in the ever-evolving needs of clinical applications, hindering the broad use of the current models. Here we introduce PathFiT, a dynamic feature learning method that can be effortlessly plugged into various pathology foundation models to unlock their adaptability. Meanwhile, PathFiT performs seamless implementation across diverse pathology applications regardless of downstream specificity. To validate PathFiT, we construct a digital pathology benchmark with over 20 terabytes of Internet and real-world data comprising 28 H\&E-stained tasks and 7 specialized imaging tasks including Masson's Trichrome staining and immunofluorescence images. By applying PathFiT to the representative pathology foundation models, we demonstrate state-of-the-art performance on 34 out of 35 tasks, with significant improvements on 23 tasks and outperforming by 10.20% on specialized imaging tasks. The superior performance and versatility of PathFiT open up new avenues in computational pathology.
title Unlocking adaptive digital pathology through dynamic feature learning
topic Image and Video Processing
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.20430