MIBIscope™ Image Analysis

The MIBIscope™ generates high-resolution, multiplexed tissue images that enable researchers to gain new insights into the cellular structure and morphology within intact tissue. The system generates multi-layer TIFF images that are supported by Bio-Formats and can be analyzed with common image analysis tools such as ImageJ, MATLAB, CellProfiler, QuPath, Fiji, HALO® and Visiopharm®.

Examples of the types of analysis that can be performed to characterize the cellular behaviors and cell-cell interactions include: colocalization, single-cell analysis, and spatial analysis.

IONpath also offers data analysis services performed by our expert MIBI bioinformatics team as part of our IONpath Research Services.


MIBIscope data can be used to simultaneously identify a multitude of cell subsets, such as  T cells, NK cells, macrophages, other immune cell types, and cancer cells, all in a single image.

Simultaneous imaging of CD68+ macrophages; CD3, CD4 and CD8 T cells with PD-1 and PD-L1 quantification.

IONpath - MIBI - colocalization of PD-L1 CD3 CD68
PD-L1 (cyan), CD3 (yellow), CD68 (magenta).
PD-L1 is present on CD68+ macrophages.
IONpath - MIBI - colocalization of CD4 CD8 PD-1
CD4 (cyan), CD8 (yellow), PD-1 (magenta).
PD-1 is present on a subset of CD8 T cells (red) and a subset of CD4 T cells (blue).
IONpath - MIBI - colocalization of PD-L1 CD3 PD-1
PD-L1 (cyan), CD3 (yellow), PD-1 (magenta).
CD3 and PD-1 colocalized (red).

Single-cell Identification and Quantification

The true sub-cellular resolution that can be obtained with MIBIscope enables precise cell segmentation and enumeration of tissue immune cells with the latest imaging analysis algorithms.

In a recent paper published by Keren et al., single-cell analysis was performed with MIBI™ imaging data for 41 TNBC samples, illustrating a possible chronology of immune cell recruitment into the tumor.

Possible chornology of immune cell recruitement to tumors
Analysis of the co-occurrence of immune populations across 41 patients.

Each immune population was classified as either present or absent in the sample, and their co-occurrence was evaluated using a chi-square test. The authors found striking interdependence between the immune populations across patients. For example, all patients that had B cells also had CD4+ T cells and CD8+ T cells (X2 p < 0.005, p < 0.0001 respectively).

Spatial analysis: Phenotype Proximities & Immune Cell Mixing 

Much of biology happens on the local level such that a cell’s nearest neighbors having large influences on cellular activities. This is especially true within the tumor microenvironment (TME) where the cellular composition may provide clinically actionable information to help guide patient treatment and disease classification.

Examples below show the distances between PD-L1 expressing cells and PD-1+ cells as well as other immune cell subsets. Images show that most of the PD-1+ cells are also CD8+ cytotoxic T cells, a feature of a suppressed immune environment within the TME.

MIBI Data Analysis
MIBIscope permits characterization of a wide range of cells within the TME and their spatial relationships, including the degree of immune infiltration within a tumor. Two tumor samples are visualized below. In Core B, keratin+ cancer cells and CD45+ immune cells are compartmentalized. In Core A, these populations are mixed. This data can be represented as a histogram plot based on the cells quantified in the image.
2 and 3 animation v2
Core A                                                 Core B
IONpath-MIBI-nearest neighbor immune tumor distances
Contact Us

Let us know how we can help you.

    Immuno-oncology/CancerNeuroscienceInfectious DiseaseImmunologyStem CellOther


    Contact Us