Pseudo-spectral angle mapping for automated pixel-level analysis of highly multiplexed tissue image data

This article has been Reviewed by the following groups

Read the full article See related articles

Listed in

Log in to save this article


The rapid development of highly multiplexed microscopy systems has enabled the study of cells embedded within their native tissue, which is providing exciting insights into the spatial features of human disease [1]. However, computational methods for analyzing these high-content images are still emerging, and there is a need for more robust and generalizable tools for evaluating the cellular constituents and underlying stroma captured by high-plex imaging [2]. To address this need, we have adapted spectral angle mapping – an algorithm used widely in hyperspectral image analysis – to compress the channel dimension of high-plex immunofluorescence images. As many high-plex immunofluorescence imaging experiments probe unique sets of protein markers, existing cell and pixel classification models do not typically generalize well. Pseudospectral angle mapping (pSAM) uses reference pseudospectra – or pixel vectors – to assign each pixel in an image a similarity score to several cell class reference vectors, which are defined by each unique staining panel. Here, we demonstrate that the class maps provided by pSAM can directly provide insight into the prevalence of each class defined by reference pseudospectra. In a dataset of high-plex images of colon biopsies from patients with gut autoimmune conditions, sixteen pSAM class representation maps were combined with instance segmentation of cells to provide cell class predictions. Finally, pSAM detected a diverse set of structure and immune cells when applied to a novel dataset of kidney biopsies imaged with a 43-marker panel. In summary, pSAM provides a powerful and readily generalizable method for evaluating high-plex immunofluorescence image data.

Significance Statement

Understanding the cellular constituents captured by highly multiplexed tissue imaging is a major limitation affecting the usability of these novel imaging methods. Many imaging experiments have uniquely designed staining panels, reducing the generalizability of cell classification models to new datasets. We present pseudospectral angle mapping (pSAM), which can compress high-dimensional image data into class representations. We demonstrate that the class representations generated by pSAM can be used to interpret high-plex image data and guide cell classification. Importantly, we also demonstrate that pSAM can generalize to new image datasets—collected with a different staining panel in samples from different tissues—without manual image annotation, subjective intensity gating, or re-training an algorithm.

Article activity feed

  1. The predicted classes agreed well with the class representations and the fluorescence images themselves

    Can this claim be quantified without deferring to a figure?

  2. ideal

    You provide a clear explanation of how you generate the pseudospectra in the methods, but I question the use of "ideal" in describing them. I get that you mean idealized, but it sounds as if they are perfect and in no way subjective which is misleading.

  3. In a test set of 50 image tiles per dataset

    Could you please clarify if this test was done by using manual segmentation as ground truth? If not, what was the source of ground truth data used to compute F1-scores?

    An F1-score of 0.7 is not generally seen as very high. Could emphasize more strongly in your discussion the power of your approach (combining the instance segmentation with class maps for classification) given that this seems to be a non-trivial segmentation problem.

  4. Figure 2.

    I found panels A-C confusing and have questions that may help you clarify to the reader. Is the grayscale image to the right of the four panels a merge of the 4 markers shown in C? If so, I might write merge on it. I was trying to connect the panels in C to the red line in A because I thought it was an example of how you might classify CD4+ T cells based on the high signal for CD3, CD 45, CD4, and CD8. But then I saw the image to the right of them. Regardless, I would switch the top two panels in C so that they read in the order of the markers in A and B (CD3 and then CD45).

  5. In the 43-plex kidney images, pSAM detected a wide range of cell classes with highly variable prevalences, suggesting that it can accurately detect rare cell types.

    This tool seems quite robust for classifying many cell types based on multiple markers. I'm wondering if you see a useful application of the tool as identifying unusual combinations of markers that deviate from the training data?