hyacinth0906 / MedicalImageAI

5 stars 3 forks source link

第二版框架更有实操性 #3

Open hyacinth0906 opened 5 years ago

hyacinth0906 commented 5 years ago

影像AI的发展和技术

行业分析及独角兽分析

放射科医生未来走向

目前能做的,针对目前AI缺陷与不足

  1. 结构化报告
  2. 诊断标准制定
  3. 客观评估一项产品
  4. 接入更多临床资料

备选,是否加入目前放射科就医处理流程,推测以后AI可能参与部分

hyacinth0906 commented 5 years ago

影像AI的发展和技术

Radiomics- Images Are More than Pictures, They Are Data.pdf Radiomics- Extracting more information from medical images using advanced feature analysis.pdf

  1. 放射基因学 snip20181114_117 snip20181114_118 ACEMAP上一共只有23篇文献 snip20181114_119 snip20181114_120 snip20181114_126

Radiogenomics- Creating a link between molecular diagnostics and diagnostic imaging.pdf

  1. 无监督学习 snip20181114_128 snip20181114_123 Convolutional Deep Belief Networks For Scalable Unsupervised Learning Of Hierarchical Representations.pdf snip20181114_129 Greedy Layer-Wise Training Of Deep Networks.pdf

snip20181114_124 snip20181114_127

Hastie T., Tibshirani R., Friedman J. (2009) Unsupervised Learning. In: The Elements of Statistical Learning. Springer Series in Statistics. Springer, New York, NY 978-0-387-84858-7

hyacinth0906 commented 5 years ago

行业分析及独角兽分析

hyacinth0906 commented 5 years ago

放射科医生未来走向

hyacinth0906 commented 5 years ago

目前能做的,针对目前AI缺陷与不足

  1. 结构化报告
  2. 诊断标准制定
  3. 客观评估一项产品
  4. 接入更多临床资料

备选,是否加入目前放射科就医处理流程,推测以后AI可能参与部分

hyacinth0906 commented 5 years ago

ACEMAP 相关子话题的中心文献,因为这是交叉学科

snip20181114_79 snip20181114_80 snip20181114_81 snip20181114_82 snip20181114_83 snip20181114_84 snip20181114_85 snip20181114_86 snip20181114_90 snip20181114_87 snip20181114_91 snip20181114_88 snip20181114_92

hyacinth0906 commented 5 years ago

snip20181115_155