-
公开(公告)号:US20140314301A1
公开(公告)日:2014-10-23
申请号:US14356661
申请日:2012-11-09
CPC分类号: G06T7/0012 , G06K9/0014 , G06K9/4652 , G06T7/90 , G06T2207/10024 , G06T2207/10056 , G06T2207/30024
摘要: The method according to the invention generates a color decomposition of histological tissue image data into density maps, where each density map corresponds to the portion of the original image data that contains one stain/tissue combination. After a microscope captures histological tissue image data from a tissue sample that has been stained with at least one stain, the said stain or stains being light absorbent, the method derives from the histological tissue image data a color mixing matrix with the columns of the matrix being color reference vectors. From this mixing matrix the method derives a density map, when the mixing matrix is well-conditioned, by applying a pseudo-inverse of the mixing matrix to the optical density data, and when the mixing matrix is medially-conditioned, by applying a piece-wise pseudo-inverse of the mixing matrix to the optical density data. The method according to the invention also generates stain/tissue combination efficacy measures.
摘要翻译: 根据本发明的方法将组织学组织图像数据的颜色分解生成密度图,其中每个密度图对应于包含一个染色/组织组合的原始图像数据的部分。 在显微镜拍摄来自已经用至少一个染色剂染色的组织样品的组织学组织图像数据之后,所述染色剂或染色剂是吸光剂,该方法从组织学组织图像数据得到颜色混合矩阵与基质列 作为颜色参考矢量。 从该混合矩阵中,该方法得到密度图,当混合矩阵条件良好时,通过将混合矩阵的伪反相应用于光密度数据,并且当混合矩阵被中间调节时,通过应用一个 将混合矩阵的逆逆向到光密度数据。 根据本发明的方法还产生染色/组织组合功效测量。
-
2.
公开(公告)号:US09779503B2
公开(公告)日:2017-10-03
申请号:US15444941
申请日:2017-02-28
申请人: CADESS Medical AB , Christer Busch
CPC分类号: G06T7/0012 , G06K9/0014 , G06K9/4652 , G06T7/90 , G06T2207/10024 , G06T2207/10056 , G06T2207/30024
摘要: A method measures the efficacy of a stain/tissue combination for histological tissue image data comprising a set of pixels. The measures of efficacy comprise a statistical measure of clustering performance of the different stain/tissue combinations in the Maxwellian Chromaticity plane, and a representative density of the stain/tissue combinations in the density maps. The method captures histological tissue image data using a light absorbent stain; transforms the histological tissue image data to optical density data; projects the optical density data to the Maxwellian Chromaticity plane; identifies at least one reference color using a statistical technique; inverts the reference color to form a color mixing matrix; derives a density map; derives a statistical measure for the clustering performance of the different stain/tissue combinations in the Maxwellian Chromaticity plane; and finds a representative density of the stain/tissue combinations in the density maps.
-