Jenna L Mueller1, Jennifer E Gallagher2, Rhea Chitalia3, Marlee Krieger3, Alaattin Erkanli4, Rebecca M Willett5, Joseph Geradts6, Nimmi Ramanujam3. 1. Department of Biomedical Engineering, Duke University, 136 Hudson Hall Box 90281, Durham, NC, 27708, USA. jenna.mueller@duke.edu. 2. Department of Surgery, Duke University Medical Center, 30 Medicine Drive White Zone, 3rd Floor, Suite 3570, Durham, NC, 27710, USA. 3. Department of Biomedical Engineering, Duke University, 136 Hudson Hall Box 90281, Durham, NC, 27708, USA. 4. Department of Biostatistics and Bioinformatics, Duke University, Brightleaf Square Suite 22B, 905 West Main Street, Durham, NC, 27701, USA. 5. Department of Electrical and Computer Engineering, University of Wisconsin - Madison, 1415 Engineering Drive, Madison, WI, 53706, USA. 6. Department of Pathology, Brigham and Women's Hospital, 75 Francis Street, Boston, MA, 02115, USA.
Abstract
PURPOSE: Histopathology is the clinical standard for tissue diagnosis; however, it requires tissue processing, laboratory personnel and infrastructure, and a highly trained pathologist to diagnose the tissue. Optical microscopy can provide real-time diagnosis, which could be used to inform the management of breast cancer. The goal of this work is to obtain images of tissue morphology through fluorescence microscopy and vital fluorescent stains and to develop a strategy to segment and quantify breast tissue features in order to enable automated tissue diagnosis. METHODS: We combined acriflavine staining, fluorescence microscopy, and a technique called sparse component analysis to segment nuclei and nucleoli, which are collectively referred to as acriflavine positive features (APFs). A series of variables, which included the density, area fraction, diameter, and spacing of APFs, were quantified from images taken from clinical core needle breast biopsies and used to create a multivariate classification model. The model was developed using a training data set and validated using an independent testing data set. RESULTS: The top performing classification model included the density and area fraction of smaller APFs (those less than 7 µm in diameter, which likely correspond to stained nucleoli).When applied to the independent testing set composed of 25 biopsy panels, the model achieved a sensitivity of 82 %, a specificity of 79 %, and an overall accuracy of 80 %. CONCLUSIONS: These results indicate that our quantitative microscopy toolbox is a potentially viable approach for detecting the presence of malignancy in clinical core needle breast biopsies.
PURPOSE: Histopathology is the clinical standard for tissue diagnosis; however, it requires tissue processing, laboratory personnel and infrastructure, and a highly trained pathologist to diagnose the tissue. Optical microscopy can provide real-time diagnosis, which could be used to inform the management of breast cancer. The goal of this work is to obtain images of tissue morphology through fluorescence microscopy and vital fluorescent stains and to develop a strategy to segment and quantify breast tissue features in order to enable automated tissue diagnosis. METHODS: We combined acriflavine staining, fluorescence microscopy, and a technique called sparse component analysis to segment nuclei and nucleoli, which are collectively referred to as acriflavine positive features (APFs). A series of variables, which included the density, area fraction, diameter, and spacing of APFs, were quantified from images taken from clinical core needle breast biopsies and used to create a multivariate classification model. The model was developed using a training data set and validated using an independent testing data set. RESULTS: The top performing classification model included the density and area fraction of smaller APFs (those less than 7 µm in diameter, which likely correspond to stained nucleoli).When applied to the independent testing set composed of 25 biopsy panels, the model achieved a sensitivity of 82 %, a specificity of 79 %, and an overall accuracy of 80 %. CONCLUSIONS: These results indicate that our quantitative microscopy toolbox is a potentially viable approach for detecting the presence of malignancy in clinical core needle breast biopsies.
Entities:
Keywords:
Breast cancer; Image analysis; Logistic models; Optical fluorescence imaging
Authors: Anthony A Tanbakuchi; Joshua A Udovich; Andrew R Rouse; Kenneth D Hatch; Arthur F Gmitro Journal: Am J Obstet Gynecol Date: 2009-10-03 Impact factor: 8.661
Authors: Anthony A Tanbakuchi; Andrew R Rouse; Joshua A Udovich; Kenneth D Hatch; Arthur F Gmitro Journal: J Biomed Opt Date: 2009 Jul-Aug Impact factor: 3.170
Authors: Timothy J Muldoon; Mark C Pierce; Dawn L Nida; Michelle D Williams; Ann Gillenwater; Rebecca Richards-Kortum Journal: Opt Express Date: 2007-12-10 Impact factor: 3.894
Authors: Jenna L Mueller; Zachary T Harmany; Jeffrey K Mito; Stephanie A Kennedy; Yongbaek Kim; Leslie Dodd; Joseph Geradts; David G Kirsch; Rebecca M Willett; J Quincy Brown; Nimmi Ramanujam Journal: PLoS One Date: 2013-06-18 Impact factor: 3.240
Authors: Michael G Giacomelli; Tadayuki Yoshitake; Lucas C Cahill; Hilde Vardeh; Liza M Quintana; Beverly E Faulkner-Jones; Jeff Brooker; James L Connolly; James G Fujimoto Journal: Biomed Opt Express Date: 2018-04-30 Impact factor: 3.732
Authors: Alexandra G Liddane; Chelsea A McNamara; Mallory C Campbell; Isabelle Mercier; James M Holaska Journal: Mol Cancer Res Date: 2021-03-26 Impact factor: 5.852
Authors: Brian T Crouch; Jennifer Gallagher; Roujia Wang; Joy Duer; Allison Hall; Mary Scott Soo; Philip Hughes; Timothy Haystead; Nirmala Ramanujam Journal: Sci Rep Date: 2019-03-05 Impact factor: 4.379