CellPilot: A unified approach to automatic and interactive segmentation in histopathology

Philipp Endres; Valentin Koch; Julia A. Schnabel; Carsten Marr

CellPilot: A unified approach to automatic and interactive segmentation in histopathology

Philipp Endres, Valentin Koch, Julia A. Schnabel, Carsten Marr

Biomedical Computing

Research output: Working paper/Preprint › Preprint

23 Downloads (Pure)

Abstract

Histopathology, the microscopic study of diseased tissue, is increasingly digitized, enabling improved visualization and streamlined workflows. An important task in histopathology is the segmentation of cells and glands, essential for determining shape and frequencies that can serve as indicators of disease. Deep learning tools are widely used in histopathology. However, variability in tissue appearance and cell morphology presents challenges for achieving reliable segmentation, often requiring manual correction to improve accuracy. This work introduces CellPilot, a framework that bridges the gap between automatic and interactive segmentation by providing initial automatic segmentation as well as guided interactive refinement. Our model was trained on over 675,000 masks of nine diverse cell and gland segmentation datasets, spanning 16 organs. CellPilot demonstrates superior performance compared to other interactive tools on three held-out histopathological datasets while enabling automatic segmentation. We make the model and a graphical user interface designed to assist practitioners in creating large-scale annotated datasets available as open-source, fostering the development of more robust and generalized diagnostic models.

Original language	Undefined/Unknown
Publication status	Published - 23 Nov 2024

Keywords

cs.CV

Access to Document

2411.15514v1

Cite this

@techreport{701ff472c1c441759f033a3150fd6bb0,

title = "CellPilot: A unified approach to automatic and interactive segmentation in histopathology",

abstract = "Histopathology, the microscopic study of diseased tissue, is increasingly digitized, enabling improved visualization and streamlined workflows. An important task in histopathology is the segmentation of cells and glands, essential for determining shape and frequencies that can serve as indicators of disease. Deep learning tools are widely used in histopathology. However, variability in tissue appearance and cell morphology presents challenges for achieving reliable segmentation, often requiring manual correction to improve accuracy. This work introduces CellPilot, a framework that bridges the gap between automatic and interactive segmentation by providing initial automatic segmentation as well as guided interactive refinement. Our model was trained on over 675,000 masks of nine diverse cell and gland segmentation datasets, spanning 16 organs. CellPilot demonstrates superior performance compared to other interactive tools on three held-out histopathological datasets while enabling automatic segmentation. We make the model and a graphical user interface designed to assist practitioners in creating large-scale annotated datasets available as open-source, fostering the development of more robust and generalized diagnostic models.",

keywords = "cs.CV",

author = "Philipp Endres and Valentin Koch and Schnabel, {Julia A.} and Carsten Marr",

year = "2024",

month = nov,

day = "23",

language = "Undefined/Unknown",

type = "WorkingPaper",

}

TY - UNPB

T1 - CellPilot

T2 - A unified approach to automatic and interactive segmentation in histopathology

AU - Endres, Philipp

AU - Koch, Valentin

AU - Schnabel, Julia A.

AU - Marr, Carsten

PY - 2024/11/23

Y1 - 2024/11/23

N2 - Histopathology, the microscopic study of diseased tissue, is increasingly digitized, enabling improved visualization and streamlined workflows. An important task in histopathology is the segmentation of cells and glands, essential for determining shape and frequencies that can serve as indicators of disease. Deep learning tools are widely used in histopathology. However, variability in tissue appearance and cell morphology presents challenges for achieving reliable segmentation, often requiring manual correction to improve accuracy. This work introduces CellPilot, a framework that bridges the gap between automatic and interactive segmentation by providing initial automatic segmentation as well as guided interactive refinement. Our model was trained on over 675,000 masks of nine diverse cell and gland segmentation datasets, spanning 16 organs. CellPilot demonstrates superior performance compared to other interactive tools on three held-out histopathological datasets while enabling automatic segmentation. We make the model and a graphical user interface designed to assist practitioners in creating large-scale annotated datasets available as open-source, fostering the development of more robust and generalized diagnostic models.

AB - Histopathology, the microscopic study of diseased tissue, is increasingly digitized, enabling improved visualization and streamlined workflows. An important task in histopathology is the segmentation of cells and glands, essential for determining shape and frequencies that can serve as indicators of disease. Deep learning tools are widely used in histopathology. However, variability in tissue appearance and cell morphology presents challenges for achieving reliable segmentation, often requiring manual correction to improve accuracy. This work introduces CellPilot, a framework that bridges the gap between automatic and interactive segmentation by providing initial automatic segmentation as well as guided interactive refinement. Our model was trained on over 675,000 masks of nine diverse cell and gland segmentation datasets, spanning 16 organs. CellPilot demonstrates superior performance compared to other interactive tools on three held-out histopathological datasets while enabling automatic segmentation. We make the model and a graphical user interface designed to assist practitioners in creating large-scale annotated datasets available as open-source, fostering the development of more robust and generalized diagnostic models.

KW - cs.CV

M3 - Preprint

BT - CellPilot

ER -