在数字病理学中基于原型的可解释性线粒分裂检测的自适应性示例选择
Mita Banik1, Ken Kreutz-Delgado1,2, Ishan Mohanty1
1Pattern Computer, Inc., Redmond, WA, United States.
Scientific reports
|February 18, 2026
概括
适应性示例选择 (AES) 提高了用于癌症诊断的AI解释性. 这种可解释的人工智能方法使用原型图像来澄清深度学习决定在线检测,提高信任和采用.
科学领域:
- 人工智能的人工智能
- 计算病理学计算病理学
- 医学图像分析 医学图像分析
背景情况:
- 黑盒子神经网络在关键医疗应用中对人工智能安全构成挑战,例如在组织病理学中.
- 对临床医生来说,解释性至关重要,他们可以信任并有效地利用癌症诊断中的AI工具.
研究的目的:
- 引入自适应示例选择 (AES),这是一个新的可解释AI框架,用于提高深度学习模型在线索检测中的可解释性.
- 让临床医生可视化AI推理,评估不确定性,并进行对比分析以提高诊断信心.
主要方法:
- 开发了一个基于原型的可解释AI框架 (AES),可以检索支持和矛盾的图像原型.
- 集成的AES带有更快的R-CNN检测器,用于强大的线粒分裂检测和跨瘤性能评估.
- AES在本地近似模型的置信面,以生成与可解释的示例相关的具体案例解释.
主要成果:
- 更快的R-CNN检测器实现了强大的跨瘤性能,在犬类皮肤乳腺细胞瘤数据集上F1得分为0.84.
- AES提供了简洁的解释,准确地捕捉了当地决策边界,并将预测与相关原型联系起来.
- 展示了与线粒体和非线粒体原型的相似性如何影响分级的信心,提供了超越离散预测的细微观点.
结论:
- AES显著提高了人工智能辅助的真菌分裂检测系统的透明度和可信度.
- 该框架促进了AI在癌症诊断中的实际应用,使临床医生能够理解和验证模型预测.
- AES代表了一步向前,使AI决策在组织病理学更容易获得和可靠的临床使用.
相关概念视频
Rapid Identification of Pathogens
MALDI-TOF MS has transformed clinical microbiology by offering a rapid and reliable method for pathogen identification. The traditional approach to microbial identification typically involves time-consuming culture techniques and biochemical tests, which can delay the initiation of appropriate antimicrobial therapy. MALDI-TOF MS avoids these delays by using characteristic ribosomal protein mass patterns of microbial cells, enabling accurate species-level identification within minutes.Principle...
Automated Microbial Diagnostics
Automated diagnostic analyzers have transformed clinical microbiology by providing rapid and reliable methods for pathogen identification and antibiotic susceptibility testing. Among these systems, the Vitek 2 is widely used because it automates the traditionally labor-intensive processes of microbial identification (ID) and antibiotic susceptibility testing (AST), delivering standardized and timely results that are essential for effective patient care.Microbial Identification with ID CardsThe...


