> For the complete documentation index, see [llms.txt](https://docs.owkin.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.owkin.com/connect-and-integrate/pathology-explorer-mcp-ai-powered-tissue-analysis/understanding-pathology-explorers-analysis-capabilities.md).

# Understanding Pathology Explorer's Analysis Capabilities

Pathology Explorer exposes slide-level histomics features derived from TCGA H\&E slides.

It is built on Owkin's pan-cancer H\&E cell detection model, trained on more than 100,000 annotated nuclei across 6 indications and 13 cell types. Compared with standard histomics, which focuses on cell densities, Pathology Explorer adds nuclear morphology, spatial interaction scores, finer region-specific densities, and improved detection of rare immune and stromal cell types.

This page documents the feature families and naming conventions.

### How Pathology Explorer compares with standard histomics <a href="#how-pathology-explorer-compares-with-standard-histomics" id="how-pathology-explorer-compares-with-standard-histomics"></a>

| Feature                                 | Standard histomics | Pathology Explorer                      |
| --------------------------------------- | ------------------ | --------------------------------------- |
| Cell counts and global densities        | Yes                | Yes                                     |
| Region-specific densities               | Tumor only         | Tumor, tumor core, stroma in tumor core |
| Region areas                            | No                 | Yes                                     |
| Nuclear morphology per cell type        | Not by default     | Area, circularity, perimeter            |
| TIL diffusivity                         | Not by default     | Yes                                     |
| Cell–cell co-occurrence                 | No                 | 9 cell-type pairs, 20 µm radius         |
| Tertiary lymphoid structure (TLS) score | Yes                | No                                      |

The two are complementary. The TLS score is only available from histomics data in K-Pro (see [Analysis methods](/explore-and-analyse-data/understand-your-data/analysis-methods.md)), not from Pathology Explorer.

### Cell quantification and distribution <a href="#cell-quantification-and-distribution" id="cell-quantification-and-distribution"></a>

Pathology Explorer supports six cell types: cancer cells, lymphocytes, neutrophils, plasmocytes, fibroblasts, and eosinophils. For each `cell_type`, the model provides:

* `count_{cell_type}`: Total number of detected cells in the slide.
* `global_density_{cell_type}`: Cell density per unit tissue area.

The model was designed to improve detection of understudied cell types, including across different tissue types and staining conditions:

* **Neutrophils**: tumor-associated neutrophils are increasingly linked to immunosuppression in the tumor microenvironment.
* **Plasmocytes** (plasma cells): key effectors of humoral anti-tumor immunity.
* **Eosinophils**: emerging roles in anti-tumor responses.
* **Fibroblasts**: central to cancer-associated stromal remodeling.

{% hint style="info" %}
The exact `cell_type` tokens are returned by the **List available cell types** tool.
{% endhint %}

### Nuclear morphology metrics <a href="#nuclear-morphology-metrics" id="nuclear-morphology-metrics"></a>

For each `cell_type`, the model provides:

* `mean_area_{cell_type}`: Mean nuclear area.
* `mean_circularity_{cell_type}`: Mean nuclear circularity.
* `mean_perimeter_{cell_type}`: Mean nuclear perimeter.

These metrics quantify changes in nuclear shape associated with malignancy or immune activation.

### Spatial organization and tissue architecture <a href="#spatial-organization-and-tissue-architecture" id="spatial-organization-and-tissue-architecture"></a>

Pathology Explorer also measures how cells are distributed across tissue regions.

#### Regional density analysis <a href="#regional-density-analysis" id="regional-density-analysis"></a>

For three region types (tumor, tumor core, and stroma in tumor core), the model provides:

* `density_{cell_type}_in_{region}`: Cell density for a given cell type in a given region.

Densities are available for all six cell types in every region. This lets you compare immune infiltration in the tumor core with infiltration in the surrounding stroma.

#### Regional area measurements <a href="#regional-area-measurements" id="regional-area-measurements"></a>

For each region, the model provides:

* `area_{region}`: Region area within the slide (`area_tumor`, `area_tumor_core`, `area_stroma_in_tumor_core`).

#### Cell–cell interaction analysis <a href="#cell-cell-interaction-analysis" id="cell-cell-interaction-analysis"></a>

For selected pairs of cell types, the model provides:

* `average_co_occurrence_{cell_type}_{cell_type2}_rad_20.0um`

This feature answers:

*How many `cell_type2` nuclei are found, on average, within 20 µm of each `cell_type` nucleus?*

Interpretation:

* **0** means no local co-occurrence at 20 µm.
* Larger values mean denser local neighborhoods of `cell_type2` around `cell_type`.

<details>

<summary>Available cell-type pairs</summary>

| `cell_type`  | `cell_type2` |
| ------------ | ------------ |
| cancer\_cell | lymphocytes  |
| cancer\_cell | neutrophils  |
| cancer\_cell | plasmocytes  |
| cancer\_cell | fibroblasts  |
| lymphocytes  | lymphocytes  |
| neutrophils  | neutrophils  |
| plasmocytes  | lymphocytes  |
| plasmocytes  | plasmocytes  |
| plasmocytes  | eosinophils  |

</details>

#### Tumor-infiltrating lymphocyte assessment <a href="#tumor-infiltrating-lymphocyte-assessment" id="tumor-infiltrating-lymphocyte-assessment"></a>

The model also computes:

* `tils_diffusivity`: a metric that quantifies how diffusely TILs are distributed in the slide.

It adds spatial context that TIL density alone does not capture: two slides with the same TIL density can have very different TIL distributions.

### Supported TCGA cohorts <a href="#supported-tcga-cohorts" id="supported-tcga-cohorts"></a>

Histomics features are available for the following TCGA cohorts:

* TCGA\_ACC
* TCGA\_BLCA
* TCGA\_BRCA
* TCGA\_CESC
* TCGA\_CHOL
* TCGA\_COAD
* TCGA\_DLBC
* TCGA\_ESCA
* TCGA\_HNSC
* TCGA\_KICH
* TCGA\_KIRC
* TCGA\_KIRP
* TCGA\_LIHC
* TCGA\_LUAD
* TCGA\_LUSC
* TCGA\_MESO
* TCGA\_OV
* TCGA\_PAAD
* TCGA\_PRAD
* TCGA\_READ
* TCGA\_SARC
* TCGA\_STAD
* TCGA\_THCA
* TCGA\_THYM
* TCGA\_UCEC
* TCGA\_UCS


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.owkin.com/connect-and-integrate/pathology-explorer-mcp-ai-powered-tissue-analysis/understanding-pathology-explorers-analysis-capabilities.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.