基于 DNA 超分子水凝胶的保护性 NK 细胞储库用于增强三阴性乳腺癌治疗
Protective NK Cell Reservoir Based on DNA Supramolecular Hydrogel for Enhanced Triple-Negative Breast Cancer Therapy.
CELL INTELLIGENCE · 肿瘤细胞治疗研究
肿瘤细胞治疗研究
采用自监督学习提升乳腺癌 TIL（肿瘤浸润淋巴细胞）评分预测
英文原题：Improving Tumor-Infiltrating Lymphocytes Score Prediction in Breast Cancer with Self-Supervised Learning.
查看英文原题
Improving Tumor-Infiltrating Lymphocytes Score Prediction in Breast Cancer with Self-Supervised Learning.
分数与星级只用于站内排序 —— 不代表疗效、安全性或个人适用性。
中文摘要
肿瘤微环境（TME）在免疫肿瘤学中起着关键作用，该领域研究肿瘤与人体免疫系统之间复杂的相互作用。具体而言，TIL（肿瘤浸润淋巴细胞）（TILs）是评估乳腺癌患者预后的重要生物标志物，并具有优化免疫治疗精准度和准确识别特定癌症类型中肿瘤细胞的潜力。在本研究中，我们开展了组织分割和淋巴细胞检测任务，通过采用能够解决有限标注数据问题的自监督学习（SSL）模型方法来预测TIL评分。我们的实验表明，与ImageNet预训练模型相比，组织分割提高了1.9%，淋巴细胞检测提高了2%。使用这些基于SSL的模型，我们实现了0.718的TIL评分，提高了4.4%。特别是，当仅使用整个数据集的10%进行训练时，SwAV预训练模型表现出优于其他模型的性能。我们的工作突出了使用SSL模型在更少标注数据下进行TIL评分预测时组织分割和淋巴细胞检测的改进。
展开英文摘要原文
Tumor microenvironment (TME) plays a pivotal role in immuno-oncology, which investigates the intricate interactions between tumors and the human immune system. Specifically, tumor-infiltrating lymphocytes (TILs) are crucial biomarkers for evaluating the prognosis of breast cancer patients and have the potential to refine immunotherapy precision and accurately identify tumor cells in specific cancer types.
In this study, we conducted tissue segmentation and lymphocyte detection tasks to predict TIL scores by employing self-supervised learning (SSL) model-based approaches capable of addressing limited labeling data issues.
Our experiments showed a 1. 9% improvement in tissue segmentation and a 2% improvement in lymphocyte detection over the ImageNet pre-training model. Using these SSL-based models, we achieved a TIL score of 0. 718 with a 4. 4% improvement. In particular, when trained with only 10% of the entire dataset, the SwAV pre-trained model exhibited a superior performance over other models.
Our work highlights improved tissue segmentation and lymphocyte detection using the SSL model with less labeled data for TIL score prediction.
论文信息
- 作者
- Kim S、Rakib Hasan K、Ando Y、Ko S、Lee D、Park NJ、Cho J
- 第一作者单位
- Department of Biomedical Science, Kyungpook National University, Daegu 41566, Republic of Korea.South Korea
- 通讯作者单位
- Clinical Omics Institute, Kyungpook National University, Daegu 41405, Republic of Korea.South Korea
- 期刊
- Life (Basel, Switzerland)2024 Jan 5
- 原文标识
- PubMed 38255705 · DOI 10.3390/life14010090
在 PubMed 查看 → 出版商原文（DOI） 全文 PDF（PMC）· 可下载 治疗专题与资料阅读指南 资料来源与翻译说明 报告译文或资料问题 →