Clinical Imaging

Combined CT, PET and ultrasound images could help diagnose gynecological cancers

A method of automatically combining CT, PET, and ultrasound scans into one image may help clinicians diagnose gynecological cancers. Together, the three modalities provide a clearer picture of indeterminate solid masses in the pelvic area.
Ultrasound is the gold standard for examining abnormal pelvic masses to differentiate between a cyst and a solid mass. It cannot differentiate, however, between a benign solid mass and a malignant one. PET scans showing metabolism and blood flow within an area can provide more information about malignancy, but localizing the pathology can be a problem without CT. Putting PET/CT and ultrasound scans together yields an image with the benefits of both.

Fusion of PET, CT, and ultrasound images for improved diagnosis of gynecological cancer. Top: PET, CT, and ultrasound images of same patient. Bottom: Three-D fusion of three imaging modalities. Combined PET/CT provides both anatomic (bone in white) and pathologic (tumor in red) information. At right, ultrasound image registered with PET/CT demonstrates malignancy of mass in pelvic area.

Dr. Baowei Fei and colleagues described their method of adding ultrasound to an already-fused PET/CT image in a paper presented at the 2007 American Institute of Ultrasound in Medicine meeting last month in New York City.

By registering ultrasound with CT, the researchers automatically registered the ultrasound with the preregistered PET scan as well, providing a level of functional imaging over the two sources of anatomic imaging. The researchers used a slice-to-volume registration method previously applied to MR images.

“As each image may have thousands or millions of pixels, the intensity values of the pixels represent the information of the image. We use the rich information from the intensity values to compute the mutual information. If two images are registered, their mutual information value is maximized. … This method is fully automatic and does not need landmarks for the registration,” Fei said.
The computations take only seconds on a desktop computer, and Fei thinks real-time registration could be done with powerful enough hardware.

The researchers tested their approach in 100 simulated trials, using real clinical data with variations in noise levels and image contrast. They had a success rate of 98% with a mean error of less than 0.1 mm for translation and 0.1° rotation. They also tested the technique on a cervical cancer patient, and visual inspection indicated excellent registration. Since publishing the paper, the researchers have tried their procedure on additional patient data and anatomic targets such as the prostate.
“The method will provide a powerful tool for clinical applications,” Fei said. “Combining ultrasound and PET/CT images has great potential to detect cancer at an early stage, improve the sensitivity and specificity, and better diagnose normal and malignant tissues.”

He also looked forward to its use with image-guided radiation therapy and image-guided radiofrequency ablation.
Fei’s colleagues contributing to the study included Drs. Nami Azar, Peter Faulhaber, Paul Rochon and Dean Nakamoto.

Article by Wendy Despain, Diagnostic Imaging

Selected Publications

Yogananda CG, Shah BR, Yu FF, Pinho MC, Nalawade SS, Murugesan GK, Wagner BC, Mickey B, Patel TR, Fei BW, Madhuranthakam AJ. A novel fully automated MRI-based deep-learning method for classification of 1p/19q co-deletion status in brain gliomas. Neuro-oncology advances. 2020 Jan; 2(1):vdaa066.
[PDF]
Yogananda CG, Shah BR, Vejdani-Jahromi M, Nalawade SS, Murugesan GK, Yu FF, Pinho MC, Wagner BC, Emblem KE, Bjørnerud A, Fei BW (Corresponding author). A fully automated deep learning network for brain tumor segmentation. Tomography. 2020 Jun; 6(2):186.
[PDF]

Yogananda CG, Shah BR, Nalawade S, Murugesan GK, Frank FY, Pinho MC, Wagner BC, Mickey B, Patel TR, Fei BW, Madhuranthakam AJ, Maldjian JA. MRI-based deep learning method for determining methylation status of the 06 methylguanne DNA methyltransferase promoter outperforms tissue based methods in brain gliomas. bioRxiv 2020.

[PDF]
Nalawade S, Fang FY, Yogananda CG, Murugesan GK, Shah BR, Pinho MC, Wagner BC, Mickey B, Patel TR, Fei BW, Madhuranthakam AJ, Maldjian JA. Brain tumor IDH, 1p/19q, and MGMT molecular classification using MRI-based deep learning: effect of motion and motion correction. bioRxiv 2020.
[PDF]
Zhou X, Ma L, Halicek M, Dormer J, Fei BW (Corresponding author). Development of a new polarized hyperspectral imaging microscope. Imaging, Therapeutics, and Advanced Technology in Head and Neck Surgery and Otolaryngology 2020; 11213(1121308). International Society for Optics and Photonics,
[PDF]
Edwards K, Chhabra A, Dormer J, Jones P, Boutin RD, Lenchik L, Fei BW (Corresponding author). Abdominal muscle segmentation from CT using a convolutional neural network. Medical Imaging 2020: Biomedical Applications in Molecular, Structural, and Functional Imaging; 11317(113170L). International Society for Optics and Photonics,
[PDF]
Ma L, Liu X, Fei BW. A multi-level similarity measure for the retrieval of the common CT imaging signs of lung diseases. Medical & Biological Engineering & Computing; 2020:1-15.
[PDF]

Bangalore Yogananda CG, Shah BR, Vejdani-Jahromi M, Nalawade SS, Murugesan GK, Yu FF, Pinho MC, Wagner BC, Mickey B, Patel TR, Fei BW (Corresponding author). A novel fully automated MRI-based deep-learning method for classification of IDH mutation status in brain gliomas. Neuro-oncology; 22(3):402-11.

[PDF]

Shahedi M, Dormer JD, TT AD, Do QN, Xi Y, Lewis MA, Madhuranthakam AJ, Twickler DM, Fei BW (Corresponding author). Segmentation of uterus and placenta in MR images using a fully convolutional neural network. Medical Imaging 2020: Computer-Aided Diagnosis; 11314(113141R). International Society for Optics and Photonics,

[PDF]

Ortega S, Halicek M, Fabelo H, Guerra R, Lopez C, Lejeune M, Godtliebsen F, Callico GM, Fei BW (Corresponding author). Hyperspectral imaging and deep learning for the detection of breast cancer cells in digitized histological images. Medical Imaging 2020: Digital Pathology; 11320(113200V). International Society for Optics and Photonics,

[PDF]

Dormer J, Bhuiyan M, Rahman N, Deaton N, Sheng J, Padala M, Desai J, Fei BW (Corresponding author). Image guided mitral valve replacement: registration of 3D ultrasound and 2D x-ray images. Medical Imaging 2020: Image-Guided Procedures, Robotic Interventions, and Modeling; 11315(113150Z). International Society for Optics and Photonics,

[PDF]

Pfefferle M, Shahub S, Shahedi M, Gahan J, Johnson B, Le P, Vargas J, Judson B, Alshara Y, Fei BW (Corresponding author). Renal biopsy under augmented reality guidance. Medical Imaging 2020: Image-Guided Procedures, Robotic Interventions, and Modeling; 11315(113152W). International Society for Optics and Photonics,

[PDF]

Selim M, Zhang J, Fei BW, Zhang GQ, Chen J. STAN-CT: Standardizing CT Image using Generative Adversarial Network. arXiv preprint; arXiv: 2004.01307.

[PDF]

Ortega S, Halicek M, Fabelo H, Callico GM, Fei BW (Corresponding author). Hyperspectral and multispectral imaging in digital and computational pathology: a systematic review. Biomedical Optics Express; 11(6): 3195-3233.

[PDF]

Nalawade SS, Muregesan GK, Vejdani-Jahromi M, Fisicaro RA, Yogananda CGB, Wagner B, Mickey B, Maher E, Pinho MC, Fei BW. Classification of Brain Tumor IDH Status using MRI and Deep Learning. bioRxiv (2019):757344.

[PDF]

Brown K, Dormer JD, Fei BW, Hoyt K, Ruiter NV, Byram BC. Deep 3D convolutional neural networks for fast super-resolution ultrasound imaging. Proceedings of SPIE Medical Imaging 2019: Ultrasonic Imaging and Tomography.

[PDF]

Qin X, Fei BW. “Cardiac Fiber Imaging with 3D Ultrasound and MR Diffusion Tensor Imaging.” In Cardiovascular Imaging: An Engineering and Clinical Perspective, edited by Ayman El-Baz. Boca Raton, FL: CRC Press, Taylor & Francis Group,

Dormer JD, Halicek M, Ma L, Reilly CM, Schreibmann E, Fei BW (Corresponding author). Convolutional neural networks for the detection of diseased hearts using CT images and left atrium patches. Proceedings of SPIE: The International Society for Optical Engineering;10575.

[PDF]

Dormer JD, Ma L, Halicek M, Reilly CM, Schreibmann E, Fei BW (Corresponding author). Heart chamber segmentation from CT using convolutional neural networks. Proceedings of SPIE: The International Society for Optical Engineering;10578.

[PDF]

Mei KQ, Hu B, Fei BW, Qin BJ. Phase asymmetry guided adaptive fractional-order total variation and diffusion for feature-preserving ultrasound despeckling. arXiv:1810.12538.

[PDF]

Abiodun-Ojo OA, Fei BW Nieh PT, Master VA, Akintayo A, Tade F, Akin-Akintayo O, Alemozaffar M, Osunkoya AO, Goodman MM, Schuster DM. The role of fluciclovine (18F) PET/CT directed, 3D ultrasound-guided fusion targeted biopsy in the detection of biochemically recurrent prostate cancer. Journal of Nuclear Medicine;59(s1):1481.

Ma L, Guo R, Zhang G, Schuster, DM, Fei BW (Corresponding author). “A Combined Learning Algorithm for Prostate Segmentation on 3D CT Images,” Medical Physics; 44(11): 5768-5781.

[PDF]

Ma L, Guo R, Zhang G, Fei BW (Corresponding author). “A random walk‐based segmentation framework for 3D ultrasound images of the prostate,” Medical Physics, 2017 Jun 5. PubMed PMID:28582803.

[PDF] [DOI]

Tian, ZQ., Liu, LZ., Zhang, ZF., Xue, JR., Fei, BW.(2017). “A supervoxel-based segmentation method for prostate MR images.”Medical Physics 44(2): 558-569.

[PDF]

Tian Z, Liu L, Fei BW, Deep convolutional neural network for prostate MR segmentation, Proc. SPIE 10135, Medical Imaging 2017: Image-Guided Procedures, Robotic Interventions, and Modeling, 101351L, March 3, 2017, Orlando, FL.

[PDF]

Ma L, Liu X, Fei BW (Corresponding author). Learning with distribution of optimized features for recognizing common CT imaging signs of lung diseases, Physics in Medicine and Biology 2017, 62:612-632.

[PDF]

Ma L, Guo R, Tian Z, Venkataraman R, Sarkar S, Liu X, Nieh PT, Master VV, Schuster DM, Fei BW (Corresponding author). Random walk based segmentation for the prostate on 3D transrectal ultrasound images. Proceedings of SPIE – The International Society for Optical Engineering 2016; 9786:978607. PubMed PMID:27660383.

[PDF]

Tian Z, Liu L, Zhang Z, Fei BW (Corresponding author). “Superpixel-based segmentation for 3D prostate MR images.” IEEE Transactions on Medical Imaging; 35(3):791-801.

[PDF]

Tian Z, Liu L, Fei BW (Corresponding author). A fully automatic multi-atlas based segmentation method for prostate MR images. Proceedings of SPIE – The International Society for Optical Engineering 2015; 9413:941340. PubMed PMID:26798187.

[PDF]

Tian Z, Liu L, Fei BW (Corresponding author). A supervoxel-based segmentation method for prostate MR images. Proceedings of SPIE – The International Society for Optical Engineering 2015; 9413:941318. PubMed PMID:26848206.

[PDF]

Qin X, Fei BW (Corresponding author). DTI template-based estimation of cardiac fiber orientations from 3D ultrasound. Medical Physics 2015; 42:2915.

[PDF]

Qin X, Fei BW (Corresponding author). Measuring myofiber orientations from high-frequency ultrasound images using multiscale decompositions. Physics in Medicine and Biology 2014; 59:3907-24.

[PDF]

Fei BW, Nieh PT, Schuster DM, Master VA, Multimodality molecular imaging for targeted biopsy of prostate cancer, The 2nd International Conference of Biomedical Engineering, Beijing, China, June 13-15, 2014.

Akbari H, Fei BW (Corresponding author). Automatic 3D segmentation of the kidney in MR images using wavelet feature extraction and probability shape model. Proceedings of SPIE – The International Society for Optical Engineering 2013; 8314:83143D. PubMed PMID:24027620.

[PDF]

Qin X, Cong Z, Jiang R, Shen M, Wagner MB, Kishbom P, Fei BW (Corresponding author). Extracting cardiac myofiber orientations from high frequency ultrasound images. Proceedings of SPIE – The International Society for Optical Engineering 2013; 8675. PubMed PMID:24392208.

[PDF]

Yang X, Fei BW (Corresponding author). Multiscale segmentation of the skull in MR images for MRI-based attenuation correction of combined MR/PET. Journal of the American Medical Informatics Association 2013; 20: 1037-45.

[PDF]

Fei BW (Corresponding author), Nieh PT, Schuster DM, Master VA. PET directed, 3D ultrasound-guided prostate biopsy, Diagnostic Imaging Europe 2013, 12-15 (Invited Paper, Featured by the journal cover).

[PDF]

Yang X, Fei BW,3D prostate segmentation of ultrasound images combining longitudinal image registration and machine learning, Edited by David R. Holmes III, Kenneth H. Wong, Proceedings of SPIE – The International Society for Optical Engineering 2012; 8316, 83162O-1~9.

[PDF]

Fei BW, Schuster DM, Nieh P, Akbari H, Fenster A, Master V, A molecular image-directed, 3D ultrasound-guided biopsy system for the prostate, Edited by David R. Holmes III, Kenneth H. Wong, Proceedings of SPIE – The International Society for Optical Engineering 2012, 8316, 831613-1~8.

[PDF]

Feng S, Bliznakova K, Qin X, Fei BW, Sechopoulos I, Characterization of the homogeneous breast tissue mixture approximation for breast image dosimetry, The Annual Meeting of the American Association of Physics in Medicine, Charlotte, NC, July 29-August 2, 2012.

Sechopoulos I, Bliznakova K, Qin X, Fei BW, Feng SS. Characterization of the homogeneous tissue mixture approximation in breast imaging dosimetry, Medical Physics 2012; 39:5050-5059. (2012 Best Paper Award, The Southeast Chapter of AAPM).

[PDF]

Yang X, Wu S, Sechopoulos I, Fei BW (Corresponding author). Cupping artifact correction and automated classification for high-resolution dedicated breast CT images. Medical Physics 2012; 39:6397-6406.

[PDF]

Fei BW, Schuster DM, Master V, Nieh P, Incorporating PET/CT images into 3D ultrasound-guided biopsy of the prostate. The Annual Meeting of the American Association of Physics in Medicine, Charlotte, NC, July 29-August 2, 2012.

[PDF]

Fei BW (Corresponding author), Yang X, Nye J, Aarovold J, Cervo M, Stark R, Meltzer CC, and Votaw J. MR/PET quantification tools: registration, segmentation, classification, and MR-based attenuation correction, Medical Physics 2012, 39:6443-6454.

[PDF]

Yang X, Ghafourian P, Sharma P, Salman K, Martin D, Fei BW (Corresponding author). Nonrigid registration and classification of the kidneys in 3D dynamic contrast enhanced (DCE) MR images. Proceedings of SPIE – The International Society for Optical Engineering 2012; 8314:83140B. PubMed PMID:22468206.

[PDF]

Yang X, Akbari H, Halig L, Fei BW (Corresponding author). 3D non-rigid registration using surface and local salient features for transrectal ultrasound image-guided prostate biopsy. Proceedings of SPIE – The International Society for Optical Engineering 2011; 7964:79642V. PubMed PMID:24027609.

[PDF]

Fei BW, Master V, Nieh P, Akbari H, Yang X, Fenster A, Schuster D. A PET/CT directed, 3D ultrasound-guided biopsy system for prostate cancer.Workshop on Prostate Cancer Imaging at the Annual Meeting of the Society of Medical Imaging Computing and Image Assisted Interventions (MICCAI), Toronto, Canada, September 18-22, 2011.

[PDF]

Yang X, Sechopoulos I, Fei BW (Corresponding author). Automatic tissue classification for high-resolution breast CT images based on bilateral filtering. Proceedings of SPIE – The International Society for Optical Engineering 2011; 7962:79623H. PubMed PMID:24027608.

[PDF]

Yang X, Fei BW, A skull segmentation method for brain MR images based on multiscale bilateral filtering scheme. SPIE Medical Imaging: Image Processing, Edited by Benoit M. Dawant, David R. Haynor, Proceedings of SPIE – The International Society for Optical Engineering2010;7623:76233K-1~8.

[PDF]

Li K, Fei BW (Corresponding author). A deformable model-based minimal path segmentation method for kidney MR images. Proceedings of SPIE – The International Society for Optical Engineering 2008; 6914. PubMed PMID:24386528.

[PDF]

Li K, Fei BW, A New 3D Model-based minimal path segmentation method for kidney MR images. The 2nd International Conference on Bioinformatics and Biomedical Engineering, 2342-2344, May 16-18, 2008.

[PDF]

Chen X, Gilkeson RC, Fei BW (Corresponding author). Automatic 3D-to-2D registration for CT and dual-energy digital radiography for calcification detection. Medical Physics 2007; 34:4934-4943.

[PDF]

Fei BW, Azar N, Greer M, Rochon PJ, Faulhaber PP. Automatic registration and fusion of ultrasound imaging and positron emission tomography (PET) for improved diagnosis of gynecologic cancer. The American Institute of Ultrasound in Medicine 2007 Convention, New York, NY, March 15-18, 2007.

[PDF]

Fei BW (Corresponding author), Chen X, Wang H, Sabol JM, DuPont E, Gilkeson RC. Automatic registration of CT volumes and dual-energy digital radiography for detection of cardiac and lung diseases. Proceedings of IEEE Engineering in Medicine and Biology Society 2006; 1:1976-9. PubMed PMID:17945687.

[PDF]

Suri JS, Pappu V, Salvado O, Fei BW, Laxminarayan S, Zhang S, Lewin JS, Duerk JL, Wilson DL. “Accurate lumen identification, detection, and quantification in MR plaque volumes.” In Handbook of Biomedical Image Analysis: Volume II: Segmentation Models, edited by Jasjit S. Suri, David L. Wilson, Swamy Laxminarayan, 451-530. New York, NY: Kluwer Academic/Plenum Publishers,

Fei BW (Corresponding author), Duerk JL, Sodee DB, Wilson DL. Semiautomatic nonrigid registration for the prostate and pelvic MR volumes.Academic Radiology 2005; 12:815-824.

[PDF]

Fei BW, Suri JS, Wilson DL. “Three-dimensional volume registration of carotid MR images.” In Plaque Imaging: Pixel to Molecular Level, edited by Jasjit S. Suri, Chun Yuan, David L. Wilson, Swamy Laxminarayan, 294-411. Fairfax, VA: IOS Press, Inc.,

Suri JS, Pappu V, Salvado O, Fei BW, Zhang S; Lewin JS, Duerk, JL, Wilson DL. Rotational effect on ROI’s for accurate lumen quantification in bifurcated MR plaque volumes. Proceedings of 17th IEEE Symposium on Computer-Based Medical Systems, Proceedings of IEEE 2004; 414- 418.

[PDF]

Fei BW, Kemper C, Wilson DL. A comparative study of warping and rigid body registration for the prostate and pelvic MR volumes. Computerized Medical Imaging and Graphics 2003; 27:267-281.

[PDF]

Fei BW, Lee Z, Boll DT, Duerk JL, Lewin JS, Wilson DL. Image registration and fusion for interventional MRI-guided thermal ablation of the prostate cancer. The Sixth Annual International Conference on Medical Imaging Computing & Computer Assisted Intervention. Lecture Notes in Computer Science (LNCS) 2003;2879:364-372, Springer-Verlag Berlin Heidelberg.

[PDF]

Fei BW, Lee Z, Duerk JL, Wilson DL. Image registration for interventional MRI-guided procedures: similarity measurements, interpolation methods, and applications to the prostate. The Second International Workshop on Biomedical Image Registration, Philadelphia, PA, June 23-24, 2003, Lecture Notes in Computer Science (LNCS) 2003; 2717:321-329, Springer-Verlag Berlin Heidelberg.

[PDF]

Fei BW, Frinkley K, Wilson DL. Registration algorithms for interventional MRI-guided treatment of the prostate. SPIE Medical Imaging: Visualization, Display, and Image-Guided Procedures, Edited by Robert L. Galloway, Proceedings of SPIE – The International Society for Optical Engineering 2003;5029:192-201.

[PDF]

Fei BW, Duerk JL, Wilson DL. Automatic 3D registration for interventional MRI-guided treatment of prostate cancer. Computer Aided Surgery 2002; 7:257-267.

[PDF]

Fei BW, Boll DT, Duerk JL, Wilson DL. Image registration for interventional MRI-guided minimally invasive treatment of prostate cancer. The Second Joint Conference of the Annual Fall Meeting of the Biomedical Engineering Society and the IEEE EMBS/BMES Conference, Proceedings of IEEE 2002; 2:1185.

[PDF]

Fei BW, Wheaton A, Lee Z, Nagano K, Duerk JL, Wilson DL. Robust registration method for interventional MRI-guided thermal ablation of prostate cancer. SPIE Medical Imaging 2001: Visualization, Display, and Image-Guided Procedures, Edited by Seong Kim Mum, Proceedings of SPIE – The International Society for Optical Engineering 2001;4319:53-60.

[PDF]

Fei BW, Zhuang T, Hu J, Zhou F. Frameless stereotactic localization and multimodal image registration using DSA/CT/MRI. The 20thAnnual International Conference of the IEEE Engineering in Medicine and Biology Society, Proceedings of IEEE 1998;2:683-685.

[PDF]

Fei BW, Wang B, Bian Z, Cheng J. Numerical coding of phased-array in intelligent phased-array ultrasonic tomography. Journal of Biomedical Engineering 1993; 10:336-340.