Prior Information Guided Regularized Deep Learning for Cell Nucleus Detection

Mohammad Tofighi, Tiantong Guo, Jairam K.P. Vanamala, Vishal Monga

Research output: Contribution to journalArticle

Abstract

Cell nuclei detection is a challenging research topic because of limitations in cellular image quality and diversity of nuclear morphology, i.e., varying nuclei shapes, sizes, and overlaps between multiple cell nuclei. This has been a topic of enduring interest with promising recent success shown by deep learning methods. These methods train convolutional neural networks (CNNs) with a training set of input images and known, labeled nuclei locations. Many such methods are supplemented by spatial or morphological processing. Using a set of canonical cell nuclei shapes, prepared with the help of a domain expert, we develop a new approach that we call shape priors (SPs) with CNNs (SPs-CNN). We further extend the network to introduce an SP layer and then allowing it to become trainable (i.e., optimizable). We call this network as tunable SP-CNN (TSP-CNN). In summary, we present new network structures that can incorporate "expected behavior" of nucleus shapes via two components: learnable layers that perform the nucleus detection and a fixed processing part that guides the learning with prior information. Analytically, we formulate two new regularization terms that are targeted at: 1) learning the shapes and 2) reducing false positives while simultaneously encouraging detection inside the cell nucleus boundary. Experimental results on two challenging datasets reveal that the proposed SP-CNN and TSP-CNN can outperform the state-of-the-art alternatives.

Original languageEnglish (US)
Pages (from-to)2047-2058
Number of pages12
JournalIEEE transactions on medical imaging
Volume38
Issue number9
DOIs
StatePublished - Sep 1 2019

Fingerprint

Cell Nucleus
Cells
Learning
Neural networks
Cell Nucleus Shape
Processing
Image quality
Research
Deep learning

All Science Journal Classification (ASJC) codes

  • Software
  • Radiological and Ultrasound Technology
  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

Tofighi, Mohammad ; Guo, Tiantong ; Vanamala, Jairam K.P. ; Monga, Vishal. / Prior Information Guided Regularized Deep Learning for Cell Nucleus Detection. In: IEEE transactions on medical imaging. 2019 ; Vol. 38, No. 9. pp. 2047-2058.
@article{620b8a34e89a4a0e865379907a0fd939,
title = "Prior Information Guided Regularized Deep Learning for Cell Nucleus Detection",
abstract = "Cell nuclei detection is a challenging research topic because of limitations in cellular image quality and diversity of nuclear morphology, i.e., varying nuclei shapes, sizes, and overlaps between multiple cell nuclei. This has been a topic of enduring interest with promising recent success shown by deep learning methods. These methods train convolutional neural networks (CNNs) with a training set of input images and known, labeled nuclei locations. Many such methods are supplemented by spatial or morphological processing. Using a set of canonical cell nuclei shapes, prepared with the help of a domain expert, we develop a new approach that we call shape priors (SPs) with CNNs (SPs-CNN). We further extend the network to introduce an SP layer and then allowing it to become trainable (i.e., optimizable). We call this network as tunable SP-CNN (TSP-CNN). In summary, we present new network structures that can incorporate {"}expected behavior{"} of nucleus shapes via two components: learnable layers that perform the nucleus detection and a fixed processing part that guides the learning with prior information. Analytically, we formulate two new regularization terms that are targeted at: 1) learning the shapes and 2) reducing false positives while simultaneously encouraging detection inside the cell nucleus boundary. Experimental results on two challenging datasets reveal that the proposed SP-CNN and TSP-CNN can outperform the state-of-the-art alternatives.",
author = "Mohammad Tofighi and Tiantong Guo and Vanamala, {Jairam K.P.} and Vishal Monga",
year = "2019",
month = "9",
day = "1",
doi = "10.1109/TMI.2019.2895318",
language = "English (US)",
volume = "38",
pages = "2047--2058",
journal = "IEEE Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "9",

}

Prior Information Guided Regularized Deep Learning for Cell Nucleus Detection. / Tofighi, Mohammad; Guo, Tiantong; Vanamala, Jairam K.P.; Monga, Vishal.

In: IEEE transactions on medical imaging, Vol. 38, No. 9, 01.09.2019, p. 2047-2058.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Prior Information Guided Regularized Deep Learning for Cell Nucleus Detection

AU - Tofighi, Mohammad

AU - Guo, Tiantong

AU - Vanamala, Jairam K.P.

AU - Monga, Vishal

PY - 2019/9/1

Y1 - 2019/9/1

N2 - Cell nuclei detection is a challenging research topic because of limitations in cellular image quality and diversity of nuclear morphology, i.e., varying nuclei shapes, sizes, and overlaps between multiple cell nuclei. This has been a topic of enduring interest with promising recent success shown by deep learning methods. These methods train convolutional neural networks (CNNs) with a training set of input images and known, labeled nuclei locations. Many such methods are supplemented by spatial or morphological processing. Using a set of canonical cell nuclei shapes, prepared with the help of a domain expert, we develop a new approach that we call shape priors (SPs) with CNNs (SPs-CNN). We further extend the network to introduce an SP layer and then allowing it to become trainable (i.e., optimizable). We call this network as tunable SP-CNN (TSP-CNN). In summary, we present new network structures that can incorporate "expected behavior" of nucleus shapes via two components: learnable layers that perform the nucleus detection and a fixed processing part that guides the learning with prior information. Analytically, we formulate two new regularization terms that are targeted at: 1) learning the shapes and 2) reducing false positives while simultaneously encouraging detection inside the cell nucleus boundary. Experimental results on two challenging datasets reveal that the proposed SP-CNN and TSP-CNN can outperform the state-of-the-art alternatives.

AB - Cell nuclei detection is a challenging research topic because of limitations in cellular image quality and diversity of nuclear morphology, i.e., varying nuclei shapes, sizes, and overlaps between multiple cell nuclei. This has been a topic of enduring interest with promising recent success shown by deep learning methods. These methods train convolutional neural networks (CNNs) with a training set of input images and known, labeled nuclei locations. Many such methods are supplemented by spatial or morphological processing. Using a set of canonical cell nuclei shapes, prepared with the help of a domain expert, we develop a new approach that we call shape priors (SPs) with CNNs (SPs-CNN). We further extend the network to introduce an SP layer and then allowing it to become trainable (i.e., optimizable). We call this network as tunable SP-CNN (TSP-CNN). In summary, we present new network structures that can incorporate "expected behavior" of nucleus shapes via two components: learnable layers that perform the nucleus detection and a fixed processing part that guides the learning with prior information. Analytically, we formulate two new regularization terms that are targeted at: 1) learning the shapes and 2) reducing false positives while simultaneously encouraging detection inside the cell nucleus boundary. Experimental results on two challenging datasets reveal that the proposed SP-CNN and TSP-CNN can outperform the state-of-the-art alternatives.

UR - http://www.scopus.com/inward/record.url?scp=85071739351&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071739351&partnerID=8YFLogxK

U2 - 10.1109/TMI.2019.2895318

DO - 10.1109/TMI.2019.2895318

M3 - Article

C2 - 30703016

AN - SCOPUS:85071739351

VL - 38

SP - 2047

EP - 2058

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

SN - 0278-0062

IS - 9

ER -