page-title

Machine learning & Image analysis

Early work
Our earliest work in machine learning began in the early 1980s, involving image recognition research sponsored by the Army Research Office, including papers on recognition of objects in photon-counting images. In 1991, Dr. Wernick proposed a precursor of today’s support vector machine (SVM).

Survey article
Since the early 1990s, we have been applying machine learning to problems in medical imaging, as surveyed in an article in a special issue of IEEE Signal Processing Magazine, for which Dr. Wernick was lead editor.

Content-based image retrieval (CBIR) in mammography
In 2002, we proposed a concept of CBIR for mammography in which an algorithm recalls mammograms from a database that are visually similar to a given patient’s image. This method uses a learned concept of similarity, trained using expert-provided similarity scores, to identify the most relevant mammograms in a database.

Computer-aided diagnosis (CAD) in mammography
Led by Dr. Yang, our group has published widely cited papers on the application of kernel machines (SVM and RVM) to detect microcalcifications in mammograms, as well as numerous other computer-aided diagnostic methods described in the list of publications below.

Learning-based model observers for image quality assessment
Led by Dr. Brankov, we have developed algorithms for automated assessment of the quality of medical images, by using machine learning to predict human readers’ ability to perform tasks using these images, such as detecting perfusion or motion defects in cardiac SPECT. Further details are provided here.

Machine learning in functional neuroimaging
We have been active since the early 1990s in the use of machine learning for functional brain mapping and pharmaceutical evaluation. Dr. Wernick has co-founded two companies to pursue this area commercially, including a company that has developed automated algorithms for Alzheimer’s disease detection.

Crime prediction / Predictive policing
In partnership with the Chicago Police Department (CPD), we won a nationwide competition for a grant from the National Institute of Justice to demonstrate the potential for using machine learning to predict crime. Our algorithms are currently in use at CPD. Further details are provided here.

Vector-space projection methods
We have extensively studied a class of signal processing algorithms based on vector-space projections, about which Dr. Yang authored a book with Dr. Henry Stark.

Deblocking of compressed images and video
Dr. Yang is known for his highly cited pioneering work in recovery from compressed images, which has since inspired several generations of research on this topic.

Automated detection of prostate cancer
Led by Dr. Samil Yetik, we have developed several machine learning techniques for automated diagnosis of prostate cancer from multispectral MRI images.

Selected publications

Fundamentals of kernel machines (SVM and RVM)

Miles N. Wernick, “Pattern classification by convex analysis,” Journal of the Optical Society of America A, vol. 8, pp. 1874-1880, 1991.

D. G. Tzikas, L. Wei, A. Likas, Y. Yang, and N. P. Galatsanos, “A Tutorial on RVM for Regression and Classification with Applications,” EURASIP News Letter, vol. 17, no. 2, pp. 4-23, 2006.

Content-based image retrieval (CBIR) in mammography

J. Wang, H. Jing, M. N. Wernick, R. M. Nishikawa, and Y. Yang, “Analysis of perceived similarity between pairs of microcalcification clusters in mammograms,” Medical Physics, 2014.

H. Jing, Y. Yang, and R. M. Nishikawa, “Retrieval boosted computer-aided diagnosis of clustered microcalcifications for breast cancer,” Medical Physics, vol. 39, no. 2, pp.676-85, 2012.

H. Jing, Y. Yang, and R. M. Nishikawa, “Regularization in retrieval-driven classification of clustered microcalcifications for breast cancer,” International Journal of Biomedical Imaging, vol. 2012, 2012.

J. H. Oh, Y. Yang, and I. El Naqa, “Adaptive learning for relevance feedback: application to digital mammography,” Medical Physics, vol. 37, no. 8, pp.4432-4444, 2010.

L. Wei, Y. Yang, M. N. Wernick, and R. M. Nishikawa, “Learning of perceptual similarity from expert readers for mammogram retrieval,” IEEE Journal of Selected Topics in Signal Processing, vol. 3, pp. 53-61, 2009.

L. Wei, Y. Yang, and R. M. Nishikawa, “Microcalcification classification assisted by content-based image retrieval for breast cancer diagnosis,” Pattern Recognition, vol. 42, pp. 1126-1132, 2009.

I. El-Naqa, Y. Yang, N. P. Galatsanos, and M. Wernick, “A similarity learning approach to content based image retrieval: application to digital mammography,” IEEE Trans. on Medical Imaging, vol. 23, pp. 1233-1244, 2004.

Computer-aided diagnosis (CAD) in mammography

Y. Lia, H. Chen, Y. Yang, and N. Yang, “Pectoral muscle segmentation in mammograms based on homogenous texture and intensity deviation,” Pattern Recognition, vol. 46, no. 3, pp. 681-691, 2013.

H. Jing, Y. Yang, M. N. Wernick, L. M. Yarusso, and R. M. Nishikawa, “A comparison study of image features between FFDM and film mammogram images,” Medical Physics, vol. 39, no. 7, 2012.

H. Jing, Y. Yang, and R. M. Nishikawa, “Detection of clustered microcalcifications using spatial point process modeling,” Phys. Med. Biol., vol. 56, no.1, pp.1-17, 2011.

J. Tang, R. Rangayyan, J. Yao, and Y. Yang, “Digital image processing and pattern recognition techniques for the detection of cancer,” Pattern Recognition, vol. 42, pp. 1015-1016, 2009.

J. Tang, R. M. Rangayyan, J. Xu, I. El Naqa, and Y. Yang, “Computer-aided detection and diagnosis of breast cancer with mammography: recent advances,” IEEE Trans. Information Technology in Biomedicine, vol. 13, pp. 236-251, 2009.

J. Tang, R. Rangayyan, J. Yao, and Y. Yang, “Introduction to the issue on digital image processing techniques for oncology,” IEEE Journal of Selected Topics in Signal Processing, vol. 3, pp. 1-3, 2009.

L. Wei, Y. Yang, R. M. Nishikawa, M. N. Wernick, and Alexandra Edwards, “Relevance vector machine for automatic detection of clustered microcalcifications,” IEEE Trans. on Medical Imaging, vol. 24, pp. 1278-1285, 2005.

L. Wei, Y. Yang, R. M. Nishikawa, and Y. Jiang, “A study on several machine-learning methods for classification of malignant and benign clustered microcalcifications,” IEEE Trans. on Medical Imaging, vol. 24, pp. 371-380, 2005.

I. El-Naqa, Y. Yang, M. Wernick, N. P. Galatsanos, and R. Nishikawa, “A support vector machine approach for detection of microcalcifications,” IEEE Trans. on Medical Imaging, vol. 21, no. 12, pp. 1552-1563, 2002.

Learning-based model observers

J. G. Brankov, “Evaluation of channelized Hotelling observer with internal-noise model in a train-test paradigm for cardiac SPECT defect detection,” accepted with minor revisions, Physics in Medicine and Biology, 2013.

M. M. Kalayeh, T. Marin, and J. G. Brankov, “Generalization evaluation of machine learning numerical observers for image quality,” IEEE Transactions on Nuclear Science, vol. 60, no. 3, pp. 1609-1618, 2013.

J. G. Brankov, Y. Yang, L. Wei, I. El Naqa, and M. N. Wernick, “Learning a nonlinear channelized observer for image quality assessment,” IEEE Transactions on Medical Imaging, vol. 28, no. 7, pp. 991-999, 2009.

Machine learning in functional neuroimaging

S. Strother, D. Matthews, A. Lukic, R. Andrews, and M. Wernick, “Superior performance of a multi-stage PET classifier for the Alzheimer’s disease cascade,” Annual Meeting of the Organization on Human Brain Mapping, 2011.

Strother SC, Lukic AS, Andrews RD, Wernick MN, Matthews DC, and ADNI. Detecting progression from MCI towards probable Alzheimer’s disease using longitudinal FDG-PET scans from the ADNI database.” The Annual Cognitive Neuroscience Conference: Mild Cognitive Impairment (MCI), 2012.

Matthews DC, Andrews RD, Lukic AS, Mosconi L, Schmidt ME, Wernick MN, Strother SC and ADNI. (2012) A multivariate glucose metabolism metric for patient stratification and prediction of cognitive decline in Alzheimer’s Disease clinical trials. AICC, Vancouver, IC-P-073 in Alzheimer’s and Dementia, 8(4, suppl 2) July 2012.

Schmidt ME, Gregg K, Margolin R, Lukic AS, Andrews RD, Matthews DC, Wernick MN, Strother SC, Brashear R, Liu E, “Preliminary analysis of baseline FDG PET in the PET substudies of the Phase 3 i.v. bapineuzumab trials in mild to moderate Alzheimer’s disease: Patterns and severity of regional brain hypometabolism and relationship to fibrillar amyloid burden measured by 11C-PiB PET and clinical outcomes,” Clinical Trials on Alzheimer’s Disease, San Diego, 2013.

Grigori Yourganov, Xu Chen, Ana S. Lukic, Cheryl L. Grady, Steven L. Small, Miles N. Wernick, and Stephen C. Strother, “Dimensionality estimation for optimal detection of functional networks in BOLD fMRI data,” Neuroimage, vol. 56 no. 2, pp. 531-543, 2011.

A. S. Lukic, M. N. Wernick, Y. Yang, L. K. Hansen, K. Arfanakis, and S. C. Strother, “Effect of spatial alignment transformations in PCA and ICA of functional neuroimages,” IEEE Trans. on Medical Imaging, vol. 26, pp. 1058-1068, 2007.

A. S. Lukic, M. N. Wernick, D. G. Tzikas, X. Chen, A. Likas, N. P. Galatsanos, Y. Yang, F. Zhao, and S. C. Strother, “Bayesian kernel methods for analysis of functional neuroimages,” IEEE Trans. on Medical Imaging, vol. 26, pp. 1613-1624, 2007.

Ana Lukic, Miles N. Wernick, and Stephen C. Strother, “Evaluation of methods for detection of brain activations from functional neuroimages,” Artificial Intelligence in Medicine (invited for special issue), vol. 25, pp. 69-88, 2002.

Vector-space projection methods

B. Marendic, Y. Yang, and H. Stark, “Phase unwrapping of two-dimensional phase functions using vector space projection methods,” J. Opt. Soc. Am. A, vol. 23, no. 8, pp. 1846-1855, 2006.

L. Liu, Y. Yang, and H. Stark, “Processing halftone color images by vector space methods,” J. Opt. Soc. Am. A, vol. 23, no. 2, pp. 247-257, 2006.

J. Gu, H. Stark, and Y. Yang, “Design of tapped-delay line antenna array using vector space projections,” IEEE Trans. on Antennas and Propagation, vol. 53, pp. 4178-4182, 2005.

L. Liu, Y. Yang, and H. Stark, “Spatial processing in color reproduction,” J. Opt. Soc. Am. A, vol. 22, no. 8, pp. 1482-1491, 2005.

J. Gu, H. Stark, and Y. Yang, “Wideband smart antenna design using vector space projection methods,” vol. 52, pp. 3228-3236, IEEE Trans. on Antennas and Propagation, 2004.

Y. Yang and H. Stark, “Design of self-healing arrays using vector space projections,” IEEE Trans. on Antennas and Propagation, vol. 49, no. 4, pp. 526-534, 2001.

Y. Yang, H. Stark, D. Gurkan, C. L. Lawson, and R. W. Cohen, “High-diffraction-efficiency pseudorandom encoding,” J. Opt. Soc. Am. A, vol. 17, no. 2, pp. 285-293, 2000.

H. Stark, Y. Yang, and D. Gurkan, “Factors affecting convergence in the design of diffractive optics by iterative vector-space methods,” J. Opt. Soc. Am. A, vol. 16, no. 1, pp. 149-159, 1999.

Y. Yang and N. P. Galatsanos, “Removal of compression artifacts using projections onto convex sets and line process modeling,” IEEE Trans. on Image Processing, vol. 6, no. 10, pp. 1345-1357, 1997.

Y. Yang, N. Galatsanos, and A. Katsaggelos, “Projection-based spatially-adaptive reconstruction of block transform compressed images,” IEEE Trans. on Image Processing, vol. 4, no. 7, pp. 896-908, 1995.

H. Stark, J. L. Wurster, and Y. Yang, “Restoration of quantum-limited images by convex projections,” J. Opt. Soc. Am. A, vol. 12, pp. 2586-2592, 1995.

Y. Yang and H. Stark, “Solutions of several color matching problems using projection theory,” J. Opt. Soc. Am. A, vol. 11, no. 1, pp. 89-96, 1994.

Y. Yang, N. P. Galatsanos, and H. Stark, “Projection-based blind deconvolution,” J. Opt. Soc. Am. A, vol. 11, no. 9, pp. 2401-2409, 1994.

Deblocking of compressed images and video

M. Choi, Y. Yang, and N. P. Galatsanos, “Multichannel regularized recovery of compressed video sequences,” IEEE Trans. on Circuits and Systems II, vol. 48, no. 4, pp. 376-387, 2001.

Y. Yang, N. P. Galatsanos, and A. K. Katsaggelos, “Regularized reconstruction to reduce blocking artifacts of block discrete cosine transform compressed images,” IEEE Trans. on Circuits and Systems for Video Tech., vol. 3, no. 6, pp. 421-432, 1993.

Object recognition in photon-limited (night vision) images

Ahmad Abu Naser, Nikolas P. Galatsanos, and Miles N. Wernick, “Methods to detect objects in photon-limited images,” Journal of the Optical Society of America A, vol. 23, pp. 272-278, 2006.

Ahmad Abu-Naser, Nikolas P. Galatsanos, Miles N. Wernick, and Dan Schonfeld, “Object recognition based on impulse restoration with use of the expectation-maximization algorithm,” Journal of the Optical Society of America A, vol. 15, pp. 2327-2340, 1998.

G. Michael Morris, Miles N. Wernick, and Thomas A. Isberg, “Image correlation at low light levels,” Optics Letters, vol. 10, pp. 315-317, 1985.

Miles N. Wernick and G. Michael Morris, “Image classification at low light levels,” Journal of the Optical Society of America A, vol. 3, pp. 2179-2187, 1986.

Automated detection of prostate cancer

Y. Artan, M. A. Haider, D. L. Langer, Y. Yang, M. N. Wernick, I. S. Yetik, “Prostate cancer localization with multispectral MRI using cost-sensitive support vector machines and conditional random fields,” IEEE Trans. on Image Processing, vol. 19, no. 9, pp.2444-2455, 2010.

S. Ozer, D. L. Langer, X. Liu, M. A. Haider, T. H. van der Kwast, A. J. Evans, Y. Yang, M. N. Wernick, I. S. Yetik, “Supervised and unsupervised methods for prostate cancer localization with multispectral MRI,” Medical Physics, vol. 37, no.4, pp.1873-83, 2010.

X. Liu, I. S. Yetik, D. L. Langer, M. A. Haider, Y. Yang, and Miles N. Wernick, “Prostate cancer segmentation with simultaneous estimation of Markov random field parameters and classes,” IEEE Trans. on Medical. Imaging, vol. 28, pp. 906-915, 2009.