# Practical AI-based cell extraction and spatial statistics for large 3D bone marrow tissue images

## TL;DR

This paper introduces PACESS, a new AI-based tool for analyzing 3D bone marrow images to better understand how different cell types are spatially organized.

## Contribution

PACESS is a novel computational workflow combining AI and spatial statistics to analyze complex 3D bone marrow tissue data.

## Key findings

- PACESS identifies distinct leukemic clusters and reveals diverse cell neighborhoods in bone marrow.
- The method classifies hundreds of thousands of cells in 3D and evaluates multicellular interactions.
- PACESS detects local heterogeneities within acute myeloid leukemia cell clusters.

## Abstract

Although the molecular regulation of hematopoiesis is well characterized, the spatial organization of hematopoietic cells within bone marrow (BM) remains unclear. Advances in microscopy have produced increasingly detailed images of murine BM, yet accurate and scalable methods to extract and analyze these complex datasets are limited. We present PACESS, a computational workflow for BM analysis that combines convolutional neural networks for 2D cell detection and classification with an automated method to extrapolate into 3D, spatial statistical analyses to define tissue regions based on local cell-type densities, and logistic regression to assess whether the relative abundances of cell types reflect reciprocal dependencies. Using PACESS, we investigate the spatial organization of T cells, megakaryocytes, and leukemic cells, revealing that distinct leukemic clusters generate diverse, previously unrecognized neighborhoods within the same BM cavity. PACESS, thus, provides a powerful tool to dissect BM architecture.

•We develop PACESS, a method for generating and analyzing large 3D bone marrow samples•PACESS provides the location of cells of interest and classifies cell types•PACESS contains a statistical framework for identifying cell clusters•PACESS identifies local heterogeneities within acute myeloid leukemia cell clusters

We develop PACESS, a method for generating and analyzing large 3D bone marrow samples

PACESS provides the location of cells of interest and classifies cell types

PACESS contains a statistical framework for identifying cell clusters

PACESS identifies local heterogeneities within acute myeloid leukemia cell clusters

The high cellular density of bone marrow (BM) complicates image segmentation, and current spatial analyses are often restricted to pairwise comparisons, unsuitable for investigating interactions between more than two cell types simultaneously. To overcome these limitations, we developed PACESS, a readily applicable neural network-based framework that classifies hundreds of thousands of cells in 3D BM samples and applies spatial statistical methods to evaluate multicellular interactions.

Adams et al. develop PACESS, a method for generating and analyzing large 3D bone marrow sections. PACESS uses neural networks and spatial statistical approaches to classify hundreds of thousands of cells and reveal cellular relationships in bone marrow.