1
|
Luna M, Chikontwe P, Park SH. Enhanced Nuclei Segmentation and Classification via Category Descriptors in the SAM Model. Bioengineering (Basel) 2024; 11:294. [PMID: 38534568 DOI: 10.3390/bioengineering11030294] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/22/2024] [Revised: 03/13/2024] [Accepted: 03/19/2024] [Indexed: 03/28/2024] Open
Abstract
Segmenting and classifying nuclei in H&E histopathology images is often limited by the long-tailed distribution of nuclei types. However, the strong generalization ability of image segmentation foundation models like the Segment Anything Model (SAM) can help improve the detection quality of rare types of nuclei. In this work, we introduce category descriptors to perform nuclei segmentation and classification by prompting the SAM model. We close the domain gap between histopathology and natural scene images by aligning features in low-level space while preserving the high-level representations of SAM. We performed extensive experiments on the Lizard dataset, validating the ability of our model to perform automatic nuclei segmentation and classification, especially for rare nuclei types, where achieved a significant detection improvement in the F1 score of up to 12%. Our model also maintains compatibility with manual point prompts for interactive refinement during inference without requiring any additional training.
Collapse
Affiliation(s)
- Miguel Luna
- Department of Robotics and Mechatronics Engineering, Daegu Gyeongbuk Institute of Science and Technology (DGIST), Daegu 42988, Republic of Korea
| | - Philip Chikontwe
- Department of Robotics and Mechatronics Engineering, Daegu Gyeongbuk Institute of Science and Technology (DGIST), Daegu 42988, Republic of Korea
| | - Sang Hyun Park
- Department of Robotics and Mechatronics Engineering, Daegu Gyeongbuk Institute of Science and Technology (DGIST), Daegu 42988, Republic of Korea
- AI Graduate School, Daegu Gyeongbuk Institute of Science and Technology (DGIST), Daegu 42988, Republic of Korea
| |
Collapse
|
2
|
Pan H, Gao B, Bai W, Li B, Li Y, Zhang M, Wang H, Zhao X, Chen M, Yin C, Kong W. WA-ResUNet: A Focused Tail Class MRI Medical Image Segmentation Algorithm. Bioengineering (Basel) 2023; 10:945. [PMID: 37627829 PMCID: PMC10451191 DOI: 10.3390/bioengineering10080945] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/16/2023] [Revised: 07/28/2023] [Accepted: 08/04/2023] [Indexed: 08/27/2023] Open
Abstract
Medical image segmentation can effectively identify lesions in medicine, but some small and rare lesions cannot be well identified. Existing studies do not take into account the uncertainty of the occurrence of diseased tissue, and the problem of long-tailed distribution of medical data. Meanwhile, the grayscale image obtained from Magnetic Resonance Imaging (MRI) detection has problems, such as the features being difficult to extract and invalid features being difficult to distinguish. In order to solve these problems, we propose a new weighted attention ResUNet (WA-ResUNet) and a class weight formula based on the number of images contained in the class, which improves the performance of the model in the low-frequency class and the overall effect of the model by improving the degree of attention paid to the valid features and invalid ones and rebalancing the learning efficiency among the classes. We evaluated our method on an uterine MRI dataset and compared it with the ResUNet. WA-ResUNet increased Intersection over Union (IoU) in the low-frequency class (Nabothian cysts) by 21.87%, and the overall mIoU increased by more than 6.5%.
Collapse
Affiliation(s)
- Haixia Pan
- College of Software, Beihang University, Beijing 100191, China
| | - Bo Gao
- College of Software, Beihang University, Beijing 100191, China
| | - Wenpei Bai
- Department of Obstetrics and Gynecology, Beijing Shijitan Hospital, Capital Medical University, Beijing 100038, China
| | - Bin Li
- Department of MRI, Beijing Shijitan Hospital, Capital Medical University/Ninth Clinical Medical College, Peking University, Beijing 100038, China
| | - Yanan Li
- College of Software, Beihang University, Beijing 100191, China
| | - Meng Zhang
- College of Software, Beihang University, Beijing 100191, China
| | - Hongqiang Wang
- College of Software, Beihang University, Beijing 100191, China
| | - Xiaoran Zhao
- College of Software, Beihang University, Beijing 100191, China
| | - Minghuang Chen
- Department of Obstetrics and Gynecology, Beijing Shijitan Hospital, Capital Medical University, Beijing 100038, China
| | - Cong Yin
- Department of Obstetrics and Gynecology, Beijing Shijitan Hospital, Capital Medical University, Beijing 100038, China
| | - Weiya Kong
- Department of Obstetrics and Gynecology, Beijing Shijitan Hospital, Capital Medical University, Beijing 100038, China
| |
Collapse
|
3
|
Li Y, Zuo H, Han P. A Universal Decoupled Training Framework for Human Parsing. Sensors (Basel) 2022; 22:5964. [PMID: 36015724 PMCID: PMC9413982 DOI: 10.3390/s22165964] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [What about the content of this article? (0)] [Affiliation(s)] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 07/14/2022] [Revised: 08/03/2022] [Accepted: 08/05/2022] [Indexed: 06/15/2023]
Abstract
Human parsing is an important technology in human-robot interaction systems. At present, the distribution of multi-category human parsing datasets is unbalanced, and the samples present a long-tailed distribution, which directly affects the performance of human parsing. Meanwhile, the similarity between different categories leads the model to predict false parsing results. To solve the above problems, a general decoupled training framework called Decoupled Training framework based on Pixel Resampling (DTPR) was proposed to solve the long-tailed distribution, and a new sampling method named Pixel Resampling based on Accuracy distribution (PRA) for semantic segmentation was also proposed and applied to this decoupled training framework. The framework divides the training process into two phases, the first phase is to improve the model feature extraction ability, and the second phase is to improve the performance of the model on tail categories. The training framework was evaluated in MHPv2.0 and LIP datasets, and tested in both high-precision and real-time SOTA models. The MPA metric of model trained by DTPR in above two datasets increased by more than 6%, and the mIoU metric increased by more than 1% without changing the model structure.
Collapse
Affiliation(s)
- Yang Li
- School of Information Engineering, Wuhan University of Technology, Wuhan 430070, China
| | - Huahong Zuo
- Wuhan Chuyan Information Technology Co., Ltd., Wuhan 430030, China
| | - Ping Han
- School of Information Engineering, Wuhan University of Technology, Wuhan 430070, China
| |
Collapse
|