With the rapid advancement of artificial intelligence technology, computational pathology has evolved from task-specific models to foundation models capable of supporting multiple organs, diseases, and downstream tasks. However, existing methods remain limited to single-pass inference on images and are unable to fully emulate the complex diagnostic processes encountered in real-world clinical settings. In recent years, large language models have driven the development of agent-based paradigms. By leveraging capabilities such as navigation, reasoning, and tool use, these frameworks offer a novel approach for modeling complex diagnostic workflows. Framed within the context of clinical pathology diagnostic processes, this article systematically reviews the key challenges and advances in computational pathology research across three stages: preliminary overview, diagnostic analysis, and report writing, and discusses the current functional roles and implementation approaches of agent-based technologies at each stage. Although validation scenarios in existing studies remain limited, the core modules and methods required to construct a pathology diagnostic agent system are largely in place. Agent-related technologies are expected to facilitate the development of a clinically deployable intelligent assistance platform for pathology diagnosis.
- Article type
- Year
- Co-author
Pathology diagnosis is the "gold standard" for clinical disease diagnosis and treatment. Intelligent pathology image analysis has significant clinical value in improving diagnostic efficiency and ensuring consistency. This paper systematically reviews the core multiple instance learning paradigm for whole slide image analysis, and elaborates the pretraining system, core functional branches and latest advances of pathology foundation models. Taking clinical usability as the core evaluation dimension, it establishes a multi-dimensional evaluation framework, analyzes the clinical translation potential of different technical routes, and points out the core contradiction of "excellent laboratory performance but insufficient clinical usability" in current pathology foundation models. It further dissects the key clinical translation bottlenecks of pathology foundation models, outlines future application and development directions, and provides a reference for technological research and development as well as clinical translation in computational pathology.
To propose an intelligent quantitative analysis method of Ki-67 index for breast cancer immunohistochemical whole slide image (WSI).
The pathological sections of patients with breast cancer diagnosed and treated in Peking Union Medical College Hospital from January 2020 to December 2020 were retrospectively collected, and scanned at 40 magnification as WSI images. Manual interpretation of the Ki-67 index was conducted by 2 pathologists according to the guidelines formulated by the International Breast Cancer Ki-67 Working Group in 2019, which is considered the gold standard. According to the ratio of 5:8, WSI was randomly divided into two data sets, A and B (data set A was randomly divided into training set, validation set and test set according to a ratio of 7:1:2). After the hot spot area in WSI of the data set A was manually marked, each WSI randomly cropped 2000 512×512 pixel patches in the 40 field of view, and 50 patches of them were randomly selected to label tumor cells and calculate the Ki-67 index. The conditional random field model was used to fuse the spatial features of the image blocks, the features were extracted by the ResNet34 pre-training model to construct a hot spot recognition model, and its performance (accuracy) was evaluated in the test set. In the hot spot area, 10 fields of view were randomly selected under the high-power field of view (×40), and the model automatically completed the cell classification and calculated the average Ki-67 index. Taking the results of manual interpretation as the gold standard, the accuracy of the Ki-67 index evaluation results of the data set B by the model was calculated, and the Bland-Altman method was used to evaluate the consistency between the results of manual interpretation and model analysis.
A total of 132 pathological sections of patients with breast cancer which met the inclusion and exclusion criteria were selected. There were 50 images in data set A (35, 5, and 10 images in training set, validation set, and test set, including 70 000, 10 000, and 20 000 patches, respectively), and 82 images in data set B. The average accuracy of the model for identifying hot spots in the test set was 81.5%, and the accuracy of the Ki-67 index calculation results for the B data set was 90.2%. Bland-Altman analysis showed that the Ki-67 index calculated by manual interpretation and model was in good agreement.
The intelligent quantitative analysis method of Ki-67 index proposed in this study has high accuracy and can assist pathologists to achieve efficient interpretation of Ki-67 index.
京公网安备11010802044758号