Abstract
Background and Purpose: Identifying the thromboembolism source in ischemic stroke is crucial for treatment and secondary prevention yet is often undetermined. This study describes a self-supervised deep learning approach in digital pathology of emboli for classifying ischemic stroke clot origin from histopathological images. Methods: The dataset included whole slide images (WSI) from the STRIP AI Kaggle challenge, consisting of retrieved clots from ischemic stroke patients following mechanical thrombectomy. Transformer-based deep learning models were developed using transfer learning and self-supervised pretraining for classifying WSI. Customizations included an attention pooling layer, weighted loss function, and threshold optimization. Various model architectures were tested and compared, and model performances were primarily evaluated using weighted logarithmic loss. Results: The model achieved a logloss score of 0.662 in cross-validation and 0.659 on the test set. Different model backbones were compared, with the swin_large_patch4_window12_384 showed higher performance. Thresholding techniques for clot origin classification were employed to balance false positives and negatives. Conclusion: The study demonstrates the extent of efficacy of transformer-based deep learning models in identifying ischemic stroke clot origins from histopathological images and emphasizes the need for refined modeling techniques specifically adapted to thrombi WSI. Further research is needed to improve model performance, interpretability, validate its effectiveness. Future enhancement could include integrating larger patient cohorts, advanced preprocessing strategies, and exploring ensemble multimodal methods for enhanced diagnostic accuracy.
Abstract (translated)
背景和目的：确定动脉粥样硬化性中风血栓的来源对于治疗和二次预防至关重要，但通常很难确定。这项研究描述了一种自监督的深度学习方法，用于病理图像中的血栓分类，以确定动脉粥样硬化性中风血栓的来源。 方法：数据集包括从STRIP AI Kaggle挑战中获取的整张图像（WSI），这些 WSI 是来自接受机械取栓治疗的患者。使用迁移学习和自监督预训练的Transformer-based深度学习模型进行分类。自定义包括注意力池化层、加权损失函数和阈值优化。各种模型架构都被测试和比较，主要通过加权对数损失进行评估来评估模型性能。 结果：在交叉验证中，模型实现了logloss分数为0.662，在测试集中为0.659。对不同的模型骨干进行了比较， swin_large_patch4_window12_384 显示了更高的性能。采用阈值技术平衡 false positives 和 negatives。 结论：本研究证明了Transformer-based深度学习模型在从病理图像中确定动脉粥样硬化性中风血栓来源方面的有效性。强调了需要专门针对血栓 WSI 的精细建模技术。还需要进一步研究提高模型性能、可解释性和验证其有效性。未来的增强可以包括纳入更大的患者队列、采用更先进的预处理策略和探索集成多模态方法以提高诊断准确性。
URL
https://arxiv.org/abs/2405.00908
