Jove
Visualize
联系我们
JoVE
x logofacebook logolinkedin logoyoutube logo
关于 JoVE
概览领导团队博客JoVE 帮助中心
作者
出版流程编辑委员会范围与政策同行评审常见问题投稿
图书馆员
用户评价订阅访问资源图书馆顾问委员会常见问题
研究
JoVE JournalMethods CollectionsJoVE Encyclopedia of Experiments存档
教育
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab Manual教师资源中心教师网站
使用条款与条件
隐私政策
政策

相关概念视频

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序
Same author

A machine learning approach to identify active polysorbate 20 degrading hydrolases in biopharmaceutical formulations.

Journal of pharmaceutical sciences·2026
Same author

Phosphatidylcholine-Polysorbate 20-Based Mixed Micelles: A New Option to Prevent Protein Aggregation?

Pharmaceutics·2026
Same author

singIST: An integrative method for comparative single-cell transcriptomics between disease models and humans.

PLoS computational biology·2026
Same author

Do we worry too much about polysorbate degradation? An industry-wide perspective with real-life case studies.

Journal of pharmaceutical sciences·2026
Same author

Specific polysorbate fingerprints of CHO hydrolases and implications for indirect assays.

Journal of pharmaceutical sciences·2025
Same author

Extraction of the polysorbate 20 and 80 fingerprint via generative modeling.

International journal of pharmaceutics: X·2025

相关实验视频

Updated: Jul 11, 2025

Methods for Characterizing the Co-development of Biofilm and Habitat Heterogeneity
09:21

Methods for Characterizing the Co-development of Biofilm and Habitat Heterogeneity

Published on: March 11, 2015

10.0K

通过机器学习方法对流成像显微镜进行次可见粒子分类和标签一致性分析.

Angela Lopez-Del Rio1, Anabel Pacios-Michelena2, Sergio Picart-Armada3

  • 1Pharmaceutical Development Biologicals, Boehringer Ingelheim Pharma GmbH & Co. KG, Biberach an der Riss 88397, Federal Republic of Germany.

Journal of pharmaceutical sciences
|November 4, 2023
PubMed
概括

机器学习有效地使用流图识别识别药品中的次可见颗粒. 无监督学习有助于粒子分类,但专家在标签上达成一致,特别是对于小粒子,仍然是一个挑战.

关键词:
流成像显微镜的流量成像显微镜.图像分析 图像分析机器学习是机器学习.粒子的表征 粒子的表征治疗解决方案 治疗解决方案

更多相关视频

Simultaneous Measurement of Turbulence and Particle Kinematics Using Flow Imaging Techniques
10:53

Simultaneous Measurement of Turbulence and Particle Kinematics Using Flow Imaging Techniques

Published on: March 12, 2019

7.1K
Characterization of Aquatic Biofilms with Flow Cytometry
08:30

Characterization of Aquatic Biofilms with Flow Cytometry

Published on: June 6, 2018

9.1K

相关实验视频

Last Updated: Jul 11, 2025

Methods for Characterizing the Co-development of Biofilm and Habitat Heterogeneity
09:21

Methods for Characterizing the Co-development of Biofilm and Habitat Heterogeneity

Published on: March 11, 2015

10.0K
Simultaneous Measurement of Turbulence and Particle Kinematics Using Flow Imaging Techniques
10:53

Simultaneous Measurement of Turbulence and Particle Kinematics Using Flow Imaging Techniques

Published on: March 12, 2019

7.1K
Characterization of Aquatic Biofilms with Flow Cytometry
08:30

Characterization of Aquatic Biofilms with Flow Cytometry

Published on: June 6, 2018

9.1K

科学领域:

  • 制药分析 制药分析
  • 生物制药品质量控制的质量控制
  • 粒子的表征 粒子的表征

背景情况:

  • 微可见颗粒是关键的质量属性在肠道药品.
  • 流成像显微镜产生粒子图像,需要进行表征.
  • 机器学习 (ML) 用于粒子检测和分类,通常需要人工专家标签.

研究的目的:

  • 开发和评估ML技术来表征生物制药中的次可见颗粒.
  • 评估无监督学习和单类分类器对粒子分析的有用性.
  • 调查基于专家的颗粒标签的一致性.

主要方法:

  • 生成的人工数据集,模仿现实世界的粒子群 (油,蛋白质,玻璃).
  • 应用无监督学习来描述样本中的粒子组成.
  • 训练有素的独立一级分类器来检测特定的粒子类型 (油,玻璃).
  • 评估模型性能与专家标记的数据对比,评估专家间的协议.

主要成果:

  • 无监督学习有效地描述了粒子组成.
  • 一类分类器显示适用于异质流量成像数据的适用性.
  • 观察到专家之间很少达成一致,特别是对于小于8微米的粒子.
  • 无监督学习可以协助粒子标记过程.

结论:

  • 一类分类器是分析复杂的流量成像显微镜数据的可行方法.
  • 无监督学习有可能简化粒子识别工作流程.
  • 专家标签的主观性和不一致性,特别是对于小颗粒,需要进一步调查和报告标签的可靠性.