no code implementations • 21 Nov 2024 • Honglin Li, Yuting Gao, Chenglu Zhu, Jingdong Chen, Ming Yang, Lin Yang
Multimodal large language models (MLLMs) are closing the gap to human visual perception capability rapidly, while, still lag behind on attending to subtle images details or locating small objects precisely, etc.
1 code implementation • 18 Oct 2024 • Honglin Li, Yunlong Zhang, Pingyi Chen, Zhongyi Shui, Chenglu Zhu, Lin Yang
Our analysis shows that the local mask aligns with the attention patterns in the lower layers of the Transformer.
no code implementations • 28 Jul 2024 • Honglin Li, Yusuan Sun, Chenglu Zhu, Yunlong Zhang, Shichuan Zhang, Zhongyi Shui, Pingyi Chen, Jingxiong Li, Sunyi Zheng, Can Cui, Lin Yang
Though computer-aided automated diagnostic models can serve as strong complement for pathologists, their effectiveness is hampered by the paucity of extensive and detailed annotations, coupled with the limited interpretability and robustness.
1 code implementation • 8 Jul 2024 • Pingyi Chen, Chenglu Zhu, Sunyi Zheng, Honglin Li, Lin Yang
Abundant experience is required for pathologists to achieve accurate and reliable diagnostic results of whole slide images (WSI).
3 code implementations • 18 Jun 2024 • Yunlong Zhang, Zhongyi Shui, Yunxuan Sun, Honglin Li, Jingxiong Li, Chenglu Zhu, Lin Yang
While existing methods to alleviate this issue introduce complex modules or processing steps, such as multiple-stage training and teacher-student distillation, this paper proposes a simple yet effective regularization: Attention Entropy Maximization (AEM).
no code implementations • 5 Jan 2024 • Sunyi Zheng, Xiaonan Cui, Yuxuan Sun, Jingxiong Li, Honglin Li, Yunlong Zhang, Pingyi Chen, Xueping Jing, Zhaoxiang Ye, Lin Yang
Additionally, we assess the robustness of PathCLIP in the task of image-image retrieval, revealing that PathCLIP performs less effectively than PLIP on Osteosarcoma but performs better on WSSS4LUAD under diverse corruptions.
no code implementations • 3 Jan 2024 • Shichuan Zhang, Sunyi Zheng, Zhongyi Shui, Honglin Li, Lin Yang
Using multi-modal data, whole slide images (WSIs) and clinical information, can improve the performance of deep learning models in the diagnosis of axillary lymph node metastasis.
1 code implementation • 27 Nov 2023 • Zhongyi Shui, Yunlong Zhang, Kai Yao, Chenglu Zhu, Sunyi Zheng, Jingxiong Li, Honglin Li, Yuxuan Sun, Ruizhe Guo, Lin Yang
In this paper, we present a novel prompt-driven framework that consists of a nucleus prompter and SAM for automatic nucleus instance segmentation.
1 code implementation • 27 Nov 2023 • Pingyi Chen, Honglin Li, Chenglu Zhu, Sunyi Zheng, Zhongyi Shui, Lin Yang
We benchmark our model on the largest subset of TCGA-PathoText.
no code implementations • 21 Nov 2023 • Honglin Li, Yunlong Zhang, Chenglu Zhu, Jiatong Cai, Sunyi Zheng, Lin Yang
Histopathology image analysis is the golden standard of clinical diagnosis for Cancers.
1 code implementation • 13 Nov 2023 • Yunlong Zhang, Honglin Li, Yuxuan Sun, Sunyi Zheng, Chenglu Zhu, Lin Yang
In the application of Multiple Instance Learning (MIL) methods for Whole Slide Image (WSI) classification, attention mechanisms often focus on a subset of discriminative instances, which are closely linked to overfitting.
no code implementations • 25 Jun 2023 • Jingxiong Li, Sunyi Zheng, Zhongyi Shui, Shichuan Zhang, Linyi Yang, Yuxuan Sun, Yunlong Zhang, Honglin Li, Yuanxin Ye, Peter M. A. van Ooijen, Kang Li, Lin Yang
This yields a non-trivial reconstruction task, allowing the model to effectively preserve chromosome banding patterns and structure details in the reconstructed results.
no code implementations • 14 Jun 2023 • Zhongyi Shui, Yizhi Zhao, Sunyi Zheng, Yunlong Zhang, Honglin Li, Shichuan Zhang, Xiaoxuan Yu, Chenglu Zhu, Lin Yang
Overall, we use the current models to generate pseudo labels for unlabeled images, which are in turn utilized to supervise the models training.
1 code implementation • 24 May 2023 • Yuxuan Sun, Chenglu Zhu, Sunyi Zheng, Kai Zhang, Lin Sun, Zhongyi Shui, Yunlong Zhang, Honglin Li, Lin Yang
Secondly, by leveraging the collected data, we construct PathCLIP, a pathology-dedicated CLIP, to enhance PathAsst's capabilities in interpreting pathology images.
1 code implementation • CVPR 2023 • Honglin Li, Chenglu Zhu, Yunlong Zhang, Yuxuan Sun, Zhongyi Shui, Wenwei Kuang, Sunyi Zheng, Lin Yang
Our framework is evaluated on five pathology WSI datasets on various WSI heads.
no code implementations • 5 Mar 2023 • Zhongyi Shui, Sunyi Zheng, Chenglu Zhu, Shichuan Zhang, Xiaoxuan Yu, Honglin Li, Jingxiong Li, Pingyi Chen, Lin Yang
Unlike mainstream PCD methods that rely on intermediate density map representations, the Point-to-Point network (P2PNet) has recently emerged as an end-to-end solution for PCD, demonstrating impressive cell detection accuracy and efficiency.
1 code implementation • 30 Jun 2022 • Yunlong Zhang, Yuxuan Sun, Honglin Li, Sunyi Zheng, Chenglu Zhu, Lin Yang
Evaluated on two resulting benchmark datasets, we find that (1) a variety of deep neural network models suffer from a significant accuracy decrease (double the error on clean images) and the unreliable confidence estimation on corrupted images; (2) A low correlation between the validation and test errors while replacing the validation set with our benchmark can increase the correlation.
no code implementations • 27 Feb 2022 • Shichuan Zhang, Chenglu Zhu, Honglin Li, Jiatong Cai, Lin Yang
We have evaluated our framework on immunohistochemical cytoplasm staining images, and the results demonstrate that our method outperforms recent cell recognition approaches.
no code implementations • 19 Oct 2021 • Francesca Palermo, Honglin Li, Alexander Capstick, Nan Fletcher-Lloyd, Yuchen Zhao, Samaneh Kouchaki, Ramin Nilforooshan, David Sharp, Payam Barnaghi
Agitation is one of the neuropsychiatric symptoms with high prevalence in dementia which can negatively impact the Activities of Daily Living (ADL) and the independence of individuals.
no code implementations • 29 Sep 2021 • Honglin Li, Frieder Ganz, David J. Sharp, Payam M. Barnaghi
The proposed model can continually learn and embed new tasks into the model without losing the information about previously learned tasks.
no code implementations • 6 Jul 2021 • Jiatong Cai, Chenglu Zhu, Can Cui, Honglin Li, Tong Wu, Shichuan Zhang, Lin Yang
In addition, the model is optimized by fine-tuning on merged domains to eliminate the interference of class mismatching among various domains.
1 code implementation • 18 Jan 2021 • Honglin Li, Roonak Rezvani, Magdalena Anita Kolanko, David J. Sharp, Maitreyee Wairagkar, Ravi Vaidyanathan, Ramin Nilforooshan, Payam Barnaghi
We have developed an integrated platform to collect in-home sensor data and performed an observational study to apply machine learning models for agitation and UTI risk analysis.
no code implementations • 27 Nov 2020 • Honglin Li, Magdalena Anita Kolanko, Shirin Enshaeifar, Severin Skillman, Andreas Markides, Mark Kenny, Eyal Soreq, Samaneh Kouchaki, Kirsten Jensen, Loren Cameron, Michael Crone, Paul Freemont, Helen Rostill, David J. Sharp, Ramin Nilforooshan, Payam Barnaghi
Machine learning techniques combined with in-home monitoring technologies provide a unique opportunity to automate diagnosis and early detection of adverse health conditions in long-term conditions such as dementia.
no code implementations • 2 Nov 2020 • Yuchen Zhao, Hanyang Liu, Honglin Li, Payam Barnaghi, Hamed Haddadi
In this paper, we propose an activity recognition system that uses semi-supervised federated learning, wherein clients conduct unsupervised learning on autoencoders with unlabelled local data to learn general representations, and a cloud server conducts supervised learning on an activity classifier with labelled data.
no code implementations • 19 Oct 2020 • Honglin Li, Yifei Fan, Frieder Ganz, Anthony Yezzi, Payam Barnaghi
The robustness of neural networks is challenged by adversarial examples that contain almost imperceptible perturbations to inputs, which mislead a classifier to incorrect outputs in high confidence.
no code implementations • 8 May 2020 • Honglin Li, Payam Barnaghi, Shirin Enshaeifar, Frieder Ganz
The changes in goals or data are referred to as new tasks in a continual learning model.
no code implementations • 9 Oct 2019 • Honglin Li, Payam Barnaghi, Shirin Enshaeifar, Frieder Ganz
The catastrophic forgetting is an inevitable problem in continual learning models for dynamic environments.
no code implementations • 20 May 2019 • Honglin Li, Shirin Enshaeifar, Frieder Ganz, Payam Barnaghi
The results show that our approach enables the model to continually learn and adapt to the new changes without forgetting the previously learned tasks.
no code implementations • 6 Nov 2018 • Honglin Li, Frieder Ganz, Shirin Enshaeifar, Payam Barnaghi
Learning in a non-stationary environment is an inevitable problem when applying machine learning algorithm to real world environment.