A Novel Nucleus-Based Classifier for Discrimination of Osteoclasts and Mesenchymal Precursor Cells in Mouse Bone Marrow Cultures
Bone remodeling occurs by the balanced action of
bone resorbing osteoclasts (OC) and bone-building osteoblasts.
Increased bone resorption by excessive OC activity contributes
to malignant and non-malignant diseases including osteoporosis.
To study OC differentiation and function, OC formed in
in vitro cultures are currently counted manually, a tedious
procedure which is prone to inter-observer differences. Aiming
for an automated OC-quantification system, classification of
OC and precursor cells was done on fluorescence microscope
images based on the distinct appearance of fluorescent nuclei.
Following ellipse fitting to nuclei, a combination of eight
features enabled clustering of OC and precursor cell nuclei.
After evaluating different machine-learning techniques, LOGREG
achieved 74% correctly classified OC and precursor cell
nuclei, outperforming human experts (best expert: 55%). In
combination with the automated detection of total cell areas,
this system allows to measure various cell parameters and most
importantly to quantify proteins involved in osteoclastogenesis.
[1] J. Edwards and G. Mundy, "Advances in osteoclast biology:
old findings and new insights from mouse model," Nat Rev
Rheumatol., pp. 235-43, 2011.
[2] M. Bethel, E. Srour, and M. Kacena, "Hematopoietic cell regulation
of osteoblast proliferation and differentiation," Curr
Osteoporos Rep., pp. 96-102, 2011.
[3] J. Edwards, M. Weivoda, and M. Kacena, "Osteoclasts: malefactors
of disease and targets for treatment." Discov Med., pp.
201-210, 2012.
[4] A. Ortiz, S. Lin, and M. Kacena, "Osteolytic and osteoblastic
bone metastases: two extremes of the same spectrum?" Recent
Results Cancer Res., pp. 192-225, 2012.
[5] M. Broadhead, J. Clark, P. Choong, and D. Myers, "Therapeutic
targeting of osteoclast function and pathways." Expert Opin
Ther Targets., pp. 169-181, 2011.
[6] E. Bradley and M. Oursler, "Osteoclast culture and resorption
assays." Methods MolBiol., pp. 19-35, 2008.
[7] G. Andersson and S. J. Marks, "Tartrate-resistant acid atpase as
a cytochemical marker for osteoclasts." J Histochem Cytochem.,
pp. 115-117, 1989.
[8] K. Henriksen, L. Tanko, P. Qvist, P. Delmas, C. Christiansen,
and M. Karsdal, "Assessment of osteoclast number and function:
application in the development of new and improved
treatment modalities for bone diseases." Osteoporos Int., pp.
681-685, 2007.
[9] T. Suzuki, T. Matsuzaki, H. Hagiwara, T. Aoki, and K. Takata,
"Recent advances in fluorescent labeling techniques for fluorescence
microscopy. acta histochem cytochem." Acta Histochem
Cytochem., pp. 131-137, 2007.
[10] A. Heindl, M. Schepelmann, R. Ecker, P. Pietschmann,
I. Ellinger, A. Seewald, and T. Thalhammer, Principles of
Osteoimmunology: Molecular Mechanisms and Clinical Applications.
Springer, 2012.
[11] B. Filas, P. Bayly, and L. Taber, "Mechanical stress as a regulator
of cytoskeletal contractility and nuclear shape in embryonic
epithelia." Ann Biomed Eng., pp. 443-454, 2010.
[12] E. Glazer, P. Bartels, A. Prasad, M. Yozwiak, H. Bartels,
D. Einspahr, JG. Alberts, and R. Krouse, "Nuclear morphometry
identifies a distinct aggressive cellular phenotype in cutaneous
squamous cell carcinoma." Cancer Prev Res (Phila)., pp.
1770-1777, 2011.
[13] C. de Andrea, A. Petrilli, R. Jesus-Garcia, L. Bleggi-Torres, and
M. Alves, "Large and round tumor nuclei in osteosarcoma: good
clinical outcome." Int J Clin Exp Pathol., pp. 169-174, 2011.
[14] S. Petushi, F. Garcia, M. Haber, C. Katsinis, and A. Tozeren,
"Large-scale computations on histology images reveal gradedifferentiating
parameters for breast cancer." BMC Med Imaging.,
2006.
[15] V. Nandakumar, L. Kelbauskas, K. Hernandez, K. Lintecum,
P. Senechal, K. Bussey, P. Davies, R. Johnson, and D. Meldrum,
"Isotropic 3d nuclear morphometry of normal, fibrocystic and
malignant breast epithelial cells reveals new structural alterations."
PLoS One., 2012.
[16] C. Chow and T. Kaneko, "Automatic boundary detection of the
left ventricle from cineangiograms." Comp. Biomed. Res., pp.
388-410, 1972.
[17] C. Yap and H. Lee, "Identification of cell nucleus using a
mumford-shah ellipse detector." Advances in Visual Computing
Lecture Notes in Computer Science, pp. 582-593, 2008.
[18] C. van Rijsbergen, Information Retrieval (2nd ed.). Butterworth,
1979.
[19] S. Kothari, Q. Chaudry, and W. MD., "A nonlinear mapping
for data structure analysis," IEEE Transactions on Computers,
pp. 401-409, 1969.
[20] M. Hall, E. Frank, G. Holmes, B. Pfahringer, P. Reutemann,
and I. Witten, "The weka data mining software: An update."
SIGKDD Explorations, 2009.
[21] D. Brllmann, A. Pabst, K. Lehmann, T. Ziebart, M. Klein, and
B. d-Hoedt, "Counting touching cell nuclei using fast ellipse
detection to assess in vitro cell characteristics: a feasibility
study." Clin Oral Investig., pp. 33-38, 2010.
[22] S. Kothari, Q. Chaudry, and W. MD., "Extraction of informative
cell features by segmentation of densely clustered tissue
images." Conf Proc IEEE Eng Med Biol Soc., pp. 6706-6709,
2009.
[1] J. Edwards and G. Mundy, "Advances in osteoclast biology:
old findings and new insights from mouse model," Nat Rev
Rheumatol., pp. 235-43, 2011.
[2] M. Bethel, E. Srour, and M. Kacena, "Hematopoietic cell regulation
of osteoblast proliferation and differentiation," Curr
Osteoporos Rep., pp. 96-102, 2011.
[3] J. Edwards, M. Weivoda, and M. Kacena, "Osteoclasts: malefactors
of disease and targets for treatment." Discov Med., pp.
201-210, 2012.
[4] A. Ortiz, S. Lin, and M. Kacena, "Osteolytic and osteoblastic
bone metastases: two extremes of the same spectrum?" Recent
Results Cancer Res., pp. 192-225, 2012.
[5] M. Broadhead, J. Clark, P. Choong, and D. Myers, "Therapeutic
targeting of osteoclast function and pathways." Expert Opin
Ther Targets., pp. 169-181, 2011.
[6] E. Bradley and M. Oursler, "Osteoclast culture and resorption
assays." Methods MolBiol., pp. 19-35, 2008.
[7] G. Andersson and S. J. Marks, "Tartrate-resistant acid atpase as
a cytochemical marker for osteoclasts." J Histochem Cytochem.,
pp. 115-117, 1989.
[8] K. Henriksen, L. Tanko, P. Qvist, P. Delmas, C. Christiansen,
and M. Karsdal, "Assessment of osteoclast number and function:
application in the development of new and improved
treatment modalities for bone diseases." Osteoporos Int., pp.
681-685, 2007.
[9] T. Suzuki, T. Matsuzaki, H. Hagiwara, T. Aoki, and K. Takata,
"Recent advances in fluorescent labeling techniques for fluorescence
microscopy. acta histochem cytochem." Acta Histochem
Cytochem., pp. 131-137, 2007.
[10] A. Heindl, M. Schepelmann, R. Ecker, P. Pietschmann,
I. Ellinger, A. Seewald, and T. Thalhammer, Principles of
Osteoimmunology: Molecular Mechanisms and Clinical Applications.
Springer, 2012.
[11] B. Filas, P. Bayly, and L. Taber, "Mechanical stress as a regulator
of cytoskeletal contractility and nuclear shape in embryonic
epithelia." Ann Biomed Eng., pp. 443-454, 2010.
[12] E. Glazer, P. Bartels, A. Prasad, M. Yozwiak, H. Bartels,
D. Einspahr, JG. Alberts, and R. Krouse, "Nuclear morphometry
identifies a distinct aggressive cellular phenotype in cutaneous
squamous cell carcinoma." Cancer Prev Res (Phila)., pp.
1770-1777, 2011.
[13] C. de Andrea, A. Petrilli, R. Jesus-Garcia, L. Bleggi-Torres, and
M. Alves, "Large and round tumor nuclei in osteosarcoma: good
clinical outcome." Int J Clin Exp Pathol., pp. 169-174, 2011.
[14] S. Petushi, F. Garcia, M. Haber, C. Katsinis, and A. Tozeren,
"Large-scale computations on histology images reveal gradedifferentiating
parameters for breast cancer." BMC Med Imaging.,
2006.
[15] V. Nandakumar, L. Kelbauskas, K. Hernandez, K. Lintecum,
P. Senechal, K. Bussey, P. Davies, R. Johnson, and D. Meldrum,
"Isotropic 3d nuclear morphometry of normal, fibrocystic and
malignant breast epithelial cells reveals new structural alterations."
PLoS One., 2012.
[16] C. Chow and T. Kaneko, "Automatic boundary detection of the
left ventricle from cineangiograms." Comp. Biomed. Res., pp.
388-410, 1972.
[17] C. Yap and H. Lee, "Identification of cell nucleus using a
mumford-shah ellipse detector." Advances in Visual Computing
Lecture Notes in Computer Science, pp. 582-593, 2008.
[18] C. van Rijsbergen, Information Retrieval (2nd ed.). Butterworth,
1979.
[19] S. Kothari, Q. Chaudry, and W. MD., "A nonlinear mapping
for data structure analysis," IEEE Transactions on Computers,
pp. 401-409, 1969.
[20] M. Hall, E. Frank, G. Holmes, B. Pfahringer, P. Reutemann,
and I. Witten, "The weka data mining software: An update."
SIGKDD Explorations, 2009.
[21] D. Brllmann, A. Pabst, K. Lehmann, T. Ziebart, M. Klein, and
B. d-Hoedt, "Counting touching cell nuclei using fast ellipse
detection to assess in vitro cell characteristics: a feasibility
study." Clin Oral Investig., pp. 33-38, 2010.
[22] S. Kothari, Q. Chaudry, and W. MD., "Extraction of informative
cell features by segmentation of densely clustered tissue
images." Conf Proc IEEE Eng Med Biol Soc., pp. 6706-6709,
2009.
@article{"International Journal of Medical, Medicine and Health Sciences:49227", author = "Andreas Heindl and Alexander K. Seewald and Martin Schepelmann and Radu Rogojanu and Giovanna Bises and Theresia Thalhammer and Isabella Ellinger", title = "A Novel Nucleus-Based Classifier for Discrimination of Osteoclasts and Mesenchymal Precursor Cells in Mouse Bone Marrow Cultures", abstract = "Bone remodeling occurs by the balanced action of
bone resorbing osteoclasts (OC) and bone-building osteoblasts.
Increased bone resorption by excessive OC activity contributes
to malignant and non-malignant diseases including osteoporosis.
To study OC differentiation and function, OC formed in
in vitro cultures are currently counted manually, a tedious
procedure which is prone to inter-observer differences. Aiming
for an automated OC-quantification system, classification of
OC and precursor cells was done on fluorescence microscope
images based on the distinct appearance of fluorescent nuclei.
Following ellipse fitting to nuclei, a combination of eight
features enabled clustering of OC and precursor cell nuclei.
After evaluating different machine-learning techniques, LOGREG
achieved 74% correctly classified OC and precursor cell
nuclei, outperforming human experts (best expert: 55%). In
combination with the automated detection of total cell areas,
this system allows to measure various cell parameters and most
importantly to quantify proteins involved in osteoclastogenesis.", keywords = "osteoclasts, machine learning, ellipse fitting.", volume = "6", number = "7", pages = "286-6", }
