Development and Validation of an Artificial Intelligence Foundation Model for Hierarchical Classification of Central Nervous System Tumors Using Hematoxylin and Eosin Whole-Slide Images
试验速览
- 阶段
- 不适用
- 状态
- 尚未招募
- 入组人数
- 20,000
- 主要终点
- Hierarchical CNS tumor classification performance on the internal hold-out test set
研究概览
简要总结
This is a multi-center, retrospective, observational study to develop and internally validate an artificial intelligence (AI) foundation model for hierarchical classification of central nervous system (CNS) tumors using approximately 20,000 hematoxylin and eosin (H&E) whole-slide images (WSIs) collected at Huashan Hospital Fudan University and Shandong Provincial Hospital. Archived pathology slides and linked de-identified clinical, histopathological, and molecular diagnostic data from patients who underwent neurosurgical tumor resection or biopsy between January 1, 2010 and December 31, 2025 will be retrospectively analyzed.
The study aims to train and evaluate weakly supervised multiple-instance learning models using pathology foundation models and conventional convolutional neural network feature extractors to predict tumor category, tumor family, terminal WHO 2021 CNS tumor diagnosis, and selected molecular alterations directly from routine H&E slides. Internal model validation will be performed using patient-level training, validation, and hold-out test datasets. Secondary analyses include comparison of model architectures, virtual molecular profiling, interpretability analyses using attention heatmaps, and comparison of AI-assisted versus pathologist-only diagnostic performance on selected internal test cases.
详细描述
Central nervous system tumors comprise a highly heterogeneous group of neoplasms with substantial diagnostic complexity. The WHO 2021 Classification of Tumors of the Central Nervous System integrates histology with molecular biomarkers, making accurate diagnosis increasingly dependent on molecular features such as IDH mutation, 1p/19q codeletion, H3 alterations, TERT promoter mutation, and other genomic or epigenomic markers. However, broad implementation of comprehensive molecular testing remains limited in many settings because of cost, turnaround time, technical complexity, and tissue constraints.
This retrospective study will use archived formalin-fixed paraffin-embedded H&E glass slides or existing digital WSIs from approximately 20,000 patients with primary or secondary CNS tumors treated at Huashan Hospital, Fudan University and Shandong Provincial Hospital. Slides will be digitized when necessary, de-identified, quality controlled, segmented for tissue regions, and divided into image patches. Patch-level features will be extracted using pretrained image encoders, including ResNet50, UNI, and CONCH, followed by weakly supervised multiple-instance learning aggregation methods such as attention-based MIL and CLAM.
The primary objective is to develop and internally validate an AI model capable of hierarchical CNS tumor classification, including tumor category, tumor family, and terminal WHO 2021 diagnosis. Secondary objectives are to compare alternative model architectures, evaluate prediction performance for key molecular markers, assess model interpretability with attention mapping, and compare AI-only, pathologist-only, and AI-assisted diagnosis on an internal test subset.
No intervention will be delivered to participants, and no clinical treatment decisions will be based on model outputs during this research stage. All data processing and model development will be conducted on secure in-hospital servers using de-identified data in accordance with institutional ethics approval and data protection procedures. ClinicalTrials.gov defines observational studies as studies in which investigators assess outcomes without assigning interventions, which matches this study design.
研究设计
- 研究类型
- Observational
- 观察模型
- Cohort
- 时间视角
- Retrospective
入排标准
- 年龄范围
- 9 Years 至 —（Child, Adult, Older Adult）
- 性别
- All
- 接受健康志愿者
- 否
入选标准
- •Patients who underwent brain or spinal tumor resection or biopsy at Huashan Hospital Fudan University and Shandong Provincial Hospital.
- •Postoperative pathology diagnosis consistent with a primary or secondary central nervous system tumor.
- •Availability of archived routine H&E-stained glass slides or existing digital whole-slide image files of adequate quality for analysis.
- •Availability of essential de-identified clinical and pathological information, including age, sex, tumor location, and key surgical/pathology records.
- •Use of archived data and samples permitted under institutional ethics approval, including waiver of informed consent where applicable.
排除标准
- •Severe slide preparation or scanning artifacts that preclude meaningful computational analysis, including extensive tissue folding, severe bubbles, severe detachment, markedly uneven staining/fading, or severe out-of-focus scanning.
- •Insufficient viable tumor tissue or insufficient analyzable tumor area for patch extraction.
- •Missing or uncertain pathological diagnosis that cannot be reliably reassigned according to the WHO 2021 CNS tumor classification using available records.
- •Cases lacking sufficient clinical, pathological, or molecular information required for core study analyses.
- •Other cases determined by the investigators to be unsuitable for algorithm training or evaluation after quality control review.
结局指标
主要结局
Hierarchical CNS tumor classification performance on the internal hold-out test set
时间窗: Assessed at model evaluation after completion of training, up to Jul 2029
Diagnostic performance of the final AI model for hierarchical classification of CNS tumors at the tumor category, tumor family, and terminal WHO 2021 diagnosis levels using de-identified H\&E whole-slide images. Performance metrics will include macro- and/or micro-area under the receiver operating characteristic curve (AUC), balanced accuracy, weighted F1 score, and Matthews correlation coefficient (MCC).
次要结局
- Comparative performance of alternative feature extractors and MIL aggregation methods(Up to Jul 2029)
- Prediction performance for selected molecular biomarkers(Up to Jul 2029)
- Agreement between AI attention maps and neuropathologist-identified diagnostic regions(Up to Jul 2029)
- Human versus AI versus AI-assisted diagnostic performance(Up to Jul 2029)