HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text Full Text (PDF) Erratum (v240,p304) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wolbarst, A. B. Articles by Hendee, W. R. Search for Related Content PubMed PubMed Citation Articles by Wolbarst, A. B. Articles by Hendee, W. R.

Evolving and Experimental Technologies in Medical Imaging¹

¹ From the Department of Radiation Medicine, Georgetown University Medical School, Washington, DC (A.B.W.); and Medical College of Wisconsin, 8701 Watertown Plank Rd, Milwaukee, WI 53226 (W.R.H.). Received September 20, 2004; revision requested November 15; revision received December 9; accepted January 14, 2005; updated July 14; final version accepted August 5. Supported in part by grants from the National Institutes of Health (RO1 CA80490, P01 CA 87634) and GE Medical Systems. Address correspondence to W.R.H. (e-mail: whendee{at}mcw.edu).

View larger version (62K): [in a new window]   Figure 1a: Digital tomosynthesis of human lung. (a) Posteroanterior radiograph shows region of interest. (b) Digital tomosynthesis section demonstrates 15-mm pulmonary nodule (arrow) that was not visualized in a. Note also improved clarity of vascular detail in b, which was reconstructed with a matrix inversion tomosynthesis technique that used 61 projection images acquired in 10 seconds over a total tube swing angle of 16°. Fifty-nine sections were generated with 3-mm spacing. Total subject entrance exposure was approximately the same as that for a screen-film lateral chest radiograph. (Image courtesy of James T. Dobbins III, PhD, H. Page McAdams, MD, and Devon J. Godfrey, Duke University Medical Center.)

View larger version (68K): [in a new window]   Figure 1b: Digital tomosynthesis of human lung. (a) Posteroanterior radiograph shows region of interest. (b) Digital tomosynthesis section demonstrates 15-mm pulmonary nodule (arrow) that was not visualized in a. Note also improved clarity of vascular detail in b, which was reconstructed with a matrix inversion tomosynthesis technique that used 61 projection images acquired in 10 seconds over a total tube swing angle of 16°. Fifty-nine sections were generated with 3-mm spacing. Total subject entrance exposure was approximately the same as that for a screen-film lateral chest radiograph. (Image courtesy of James T. Dobbins III, PhD, H. Page McAdams, MD, and Devon J. Godfrey, Duke University Medical Center.)

View larger version (77K): [in a new window]   Figure 2: Dependence of relative amplitude of x-ray scatter (in this case, through 180°) on atomic number. Truck is being irradiated from the side, and backscattered radiation is imaged by an image receptor on the same side as the radiation source. Organic and other materials with a lower atomic number, such as drugs, explosives, plastic weapons, or people, scatter x-rays much more readily than do high-atomic-number items, such as trucks and guns. The amount of back-scatter from an object also depends, of course, on its physical density. (Image courtesy of American Science and Engineering, Billerica, Mass.)

View larger version (121K): [in a new window]   Figure 3a: Slip rings used to bring power to x-ray tube on rotating gantry of a helical CT machine and, for some designs, to acquire information from the detector array. (a) The shiny metal strips carry electric signals that are swept off by special brushes. (b) The brushes are not in the form of bristles but rather of metal blocks (in this case a silver alloy). The five pairs of larger brushes provide the voltage required by the x-ray tube, and the three pairs of smaller ones transfer signals from the gantry controller.

View larger version (89K): [in a new window]   Figure 3b: Slip rings used to bring power to x-ray tube on rotating gantry of a helical CT machine and, for some designs, to acquire information from the detector array. (a) The shiny metal strips carry electric signals that are swept off by special brushes. (b) The brushes are not in the form of bristles but rather of metal blocks (in this case a silver alloy). The five pairs of larger brushes provide the voltage required by the x-ray tube, and the three pairs of smaller ones transfer signals from the gantry controller.

View larger version (55K): [in a new window]   Figure 4a: Electron-beam CT, also known as fifth-generation CT. (a) Diagram shows electron-beam CT scanner architecture. Target is a long continuous tungsten strip that makes a 210° arc around the patient. Electron beam and focal spot traverse the entire arc in 50–100 msec. Electron beam and target reside in a single funnel-shaped evacuated chamber (not shown). Detector array does not move. (b) Transverse electron-beam CT image in a patient with obstructive coronary artery disease who had undergone bypass surgery. Calcium is visible in left main coronary artery (lower left arrow), left anterior descending coronary artery (lower right arrow), and aorta (upper left arrow). Surgical clip (upper right arrow) from bypass surgery is also visible.

View larger version (145K): [in a new window]   Figure 4b: Electron-beam CT, also known as fifth-generation CT. (a) Diagram shows electron-beam CT scanner architecture. Target is a long continuous tungsten strip that makes a 210° arc around the patient. Electron beam and focal spot traverse the entire arc in 50–100 msec. Electron beam and target reside in a single funnel-shaped evacuated chamber (not shown). Detector array does not move. (b) Transverse electron-beam CT image in a patient with obstructive coronary artery disease who had undergone bypass surgery. Calcium is visible in left main coronary artery (lower left arrow), left anterior descending coronary artery (lower right arrow), and aorta (upper left arrow). Surgical clip (upper right arrow) from bypass surgery is also visible.

View larger version (114K): [in a new window]   Figure 5: PET/CT fusion imaging. Coronal CT (left), PET (middle), and fused PET/CT (right) images show distribution of positron-emitting 18F-fluorodeoxyglucose superimposed on CT display of anatomy. (Image courtesy of Robert Hellman, MD, Medical College of Wisconsin, Milwaukee, Wis.)

View larger version (76K): [in a new window]   Figure 6a: Transverse MR/CT fusion imaging for radiation therapy treatment planning. Brain tumor is barely visible on (a) CT scan but is clearly evident on (b) MR image (repetition time msec/echo time msec, 10 000/138) and (c) fused MR/CT image. At present, nearly all treatment-planning systems used to generate isodose maps for radiation therapy can work with CT or with fused MR/CT images (as in d) but not with MR images alone. This situation is likely to improve soon. (Image courtesy of Allen Li, PhD, Medical College of Wisconsin, Milwaukee, Wis.)

View larger version (117K): [in a new window]   Figure 6b: Transverse MR/CT fusion imaging for radiation therapy treatment planning. Brain tumor is barely visible on (a) CT scan but is clearly evident on (b) MR image (repetition time msec/echo time msec, 10 000/138) and (c) fused MR/CT image. At present, nearly all treatment-planning systems used to generate isodose maps for radiation therapy can work with CT or with fused MR/CT images (as in d) but not with MR images alone. This situation is likely to improve soon. (Image courtesy of Allen Li, PhD, Medical College of Wisconsin, Milwaukee, Wis.)

View larger version (106K): [in a new window]   Figure 6c: Transverse MR/CT fusion imaging for radiation therapy treatment planning. Brain tumor is barely visible on (a) CT scan but is clearly evident on (b) MR image (repetition time msec/echo time msec, 10 000/138) and (c) fused MR/CT image. At present, nearly all treatment-planning systems used to generate isodose maps for radiation therapy can work with CT or with fused MR/CT images (as in d) but not with MR images alone. This situation is likely to improve soon. (Image courtesy of Allen Li, PhD, Medical College of Wisconsin, Milwaukee, Wis.)

View larger version (95K): [in a new window]   Figure 6d: Transverse MR/CT fusion imaging for radiation therapy treatment planning. Brain tumor is barely visible on (a) CT scan but is clearly evident on (b) MR image (repetition time msec/echo time msec, 10 000/138) and (c) fused MR/CT image. At present, nearly all treatment-planning systems used to generate isodose maps for radiation therapy can work with CT or with fused MR/CT images (as in d) but not with MR images alone. This situation is likely to improve soon. (Image courtesy of Allen Li, PhD, Medical College of Wisconsin, Milwaukee, Wis.)

View larger version (45K): [in a new window]   Figure 7: Chemical shift MR/MR spectroscopic imaging (1000/18; bandwidth, 1000 Hz; 512 points with two signals acquired). Transverse short-echo-time chemical shift image (left) acquired at 0.5 T in a presymptomatic patient with Huntington disease shows strong elevation of glutamate (right: upper spectrum) in the head of the putamen. Unaffected thalamus (right: lower spectrum) is shown for comparison. (Image courtesy of Robert Prost, PhD, Medical College of Wisconsin, Milwaukee, Wis.)

View larger version (103K): [in a new window]   Figure 8a: Mammography and MR imaging of the breast. (a) Left craniocaudal and (b) left mediolateral oblique mammograms demonstrate an irregular high-density mass (*). Overall, breast tissue density is heterogeneous, which is consistent with patient's age (35 years). Biopsy results showed the mass to be grade III invasive ductal carcinoma. (c, d) Transverse three-dimensional fast spoiled gradient-echo MR imaging (21.4/4.2) was performed to assess remaining breast tissue. (c) Enhancing mass corresponds to the known biopsy-proved breast cancer. (d) Incidental mass (arrow) was found posteriorly in the same breast. Biopsy was performed later with US guidance, and the mass found to be grade I invasive ductal carcinoma. This mass was not evident on the mammogram, either initially or in retrospect. (Image courtesy of Lonie Salkowski, MD, Medical College of Wisconsin, Milwaukee, Wis.)

View larger version (106K): [in a new window]   Figure 8b: Mammography and MR imaging of the breast. (a) Left craniocaudal and (b) left mediolateral oblique mammograms demonstrate an irregular high-density mass (*). Overall, breast tissue density is heterogeneous, which is consistent with patient's age (35 years). Biopsy results showed the mass to be grade III invasive ductal carcinoma. (c, d) Transverse three-dimensional fast spoiled gradient-echo MR imaging (21.4/4.2) was performed to assess remaining breast tissue. (c) Enhancing mass corresponds to the known biopsy-proved breast cancer. (d) Incidental mass (arrow) was found posteriorly in the same breast. Biopsy was performed later with US guidance, and the mass found to be grade I invasive ductal carcinoma. This mass was not evident on the mammogram, either initially or in retrospect. (Image courtesy of Lonie Salkowski, MD, Medical College of Wisconsin, Milwaukee, Wis.)

View larger version (147K): [in a new window]   Figure 8c: Mammography and MR imaging of the breast. (a) Left craniocaudal and (b) left mediolateral oblique mammograms demonstrate an irregular high-density mass (*). Overall, breast tissue density is heterogeneous, which is consistent with patient's age (35 years). Biopsy results showed the mass to be grade III invasive ductal carcinoma. (c, d) Transverse three-dimensional fast spoiled gradient-echo MR imaging (21.4/4.2) was performed to assess remaining breast tissue. (c) Enhancing mass corresponds to the known biopsy-proved breast cancer. (d) Incidental mass (arrow) was found posteriorly in the same breast. Biopsy was performed later with US guidance, and the mass found to be grade I invasive ductal carcinoma. This mass was not evident on the mammogram, either initially or in retrospect. (Image courtesy of Lonie Salkowski, MD, Medical College of Wisconsin, Milwaukee, Wis.)

View larger version (139K): [in a new window]   Figure 8d: Mammography and MR imaging of the breast. (a) Left craniocaudal and (b) left mediolateral oblique mammograms demonstrate an irregular high-density mass (*). Overall, breast tissue density is heterogeneous, which is consistent with patient's age (35 years). Biopsy results showed the mass to be grade III invasive ductal carcinoma. (c, d) Transverse three-dimensional fast spoiled gradient-echo MR imaging (21.4/4.2) was performed to assess remaining breast tissue. (c) Enhancing mass corresponds to the known biopsy-proved breast cancer. (d) Incidental mass (arrow) was found posteriorly in the same breast. Biopsy was performed later with US guidance, and the mass found to be grade I invasive ductal carcinoma. This mass was not evident on the mammogram, either initially or in retrospect. (Image courtesy of Lonie Salkowski, MD, Medical College of Wisconsin, Milwaukee, Wis.)

View larger version (158K): [in a new window]   Figure 9a: Transverse intermolecular double-quantum coherence (IDQC) MR images (1500/60; matrix, 128 x 128; four signals acquired) reveal enhanced contrast in healthy human brain. (a) IDQC-encode gradient was applied along direction of the magnetic field (B0). (b) IDQC-encode gradient was applied along "magic angle," where signals are minimized. (c) Conventional T2-weighted single-quantum coherence image. Images in a and b are displayed with same window setting but are different from that for c. (Image courtesy of Jianhui Zhong, PhD, University of Rochester, Rochester, NY.)

View larger version (135K): [in a new window]   Figure 9b: Transverse intermolecular double-quantum coherence (IDQC) MR images (1500/60; matrix, 128 x 128; four signals acquired) reveal enhanced contrast in healthy human brain. (a) IDQC-encode gradient was applied along direction of the magnetic field (B0). (b) IDQC-encode gradient was applied along "magic angle," where signals are minimized. (c) Conventional T2-weighted single-quantum coherence image. Images in a and b are displayed with same window setting but are different from that for c. (Image courtesy of Jianhui Zhong, PhD, University of Rochester, Rochester, NY.)

View larger version (147K): [in a new window]   Figure 9c: Transverse intermolecular double-quantum coherence (IDQC) MR images (1500/60; matrix, 128 x 128; four signals acquired) reveal enhanced contrast in healthy human brain. (a) IDQC-encode gradient was applied along direction of the magnetic field (B0). (b) IDQC-encode gradient was applied along "magic angle," where signals are minimized. (c) Conventional T2-weighted single-quantum coherence image. Images in a and b are displayed with same window setting but are different from that for c. (Image courtesy of Jianhui Zhong, PhD, University of Rochester, Rochester, NY.)

View larger version (142K): [in a new window]   Figure 10: Four-dimensional US images of breast in a 39-year-old woman with stage T2 cancer (arrow) in the left breast. Left: Cancer initially measured 24 x 12 x 19 mm. Patient was examined with combined three-dimensional US and digital mammography research system being developed at the University of Michigan (Ann Arbor, Mich) in partnership with GE Global Research. Right: Scans obtained after completion of four cycles of neoadjuvant chemotherapy. By viewing nearly whole breast image volumes, it is easier to localize any posttreatment remains of tumor and to judge treatment success, hopefully with automatic volume change measurements. Such four-dimensional imaging may revolutionize three-dimensional US screening for breast cancer, allowing precise section-by-section or pixel-by-pixel visual comparison. Ant = anterior, Lat = lateral, Med = medial, Post = posterior. (Image courtesy of Charles Meyer, PhD, and Paul Carson, PhD, University of Michigan, Ann Arbor, Mich)

View larger version (90K): [in a new window]   Figure 11: Confocal scanning laser tomographic image of posterior segment of the eye shows topography of optic nerve head and of surrounding retinal surface in three dimensions. In less than 2 seconds, the 670-nm laser performed a sequence of 64 scans of the retina over a 15° x 15° field, creating 64 386 x 386-pixel planar images out of light reflected from different depths. The shape of the indentation edge, emphasized by the drawn green line, indicates nerve fiber layer defect at the rim of the optic nerve head. (Image courtesy of Heidelberg Engineering, Dossenheim, Germany.)

View larger version (105K): [in a new window]   Figure 12a: Terahertz imaging. (a) In vivo terahertz image of volunteer's forearm acquired by using a handheld terahertz imaging system. The image shows a 15 x 15-mm region; a scar running from left to right can be seen in the top half of the image. The image was generated by plotting the electric field value reflected from beneath the skin surface. Dark circular regions are hair follicles of normal skin, which are not present at the scar. (b) Axial terahertz image (b scan) of edge of volunteer's hand. Gray scale indicates signal amplitude, which is plotted against optical delay (y-axis) and position across the scanned area (x-axis). Decrease in stratum corneum thickness across the x-axis, from the palm-side (30 mm on x-axis) to the backside (70 mm) of the hand, is evident. (Image courtesy of Vincent Wallace, PhD, TeraView, Cambridge, England)

View larger version (65K): [in a new window]   Figure 12b: Terahertz imaging. (a) In vivo terahertz image of volunteer's forearm acquired by using a handheld terahertz imaging system. The image shows a 15 x 15-mm region; a scar running from left to right can be seen in the top half of the image. The image was generated by plotting the electric field value reflected from beneath the skin surface. Dark circular regions are hair follicles of normal skin, which are not present at the scar. (b) Axial terahertz image (b scan) of edge of volunteer's hand. Gray scale indicates signal amplitude, which is plotted against optical delay (y-axis) and position across the scanned area (x-axis). Decrease in stratum corneum thickness across the x-axis, from the palm-side (30 mm on x-axis) to the backside (70 mm) of the hand, is evident. (Image courtesy of Vincent Wallace, PhD, TeraView, Cambridge, England)

View larger version (35K): [in a new window]   Figure 13a: Proton electron double resonance imaging (PEDRI; also known as Overhauser imaging). (a) Time course of PEDRI study of myocardial uptake of free radical probe TEMPONE (4-oxo-2,2,6,6-tetramethylpiperidine-1-oxyl) by isolated perfused rat heart. TEMPONE was infused through a side arm proximal to the perfusion cannula at final concentration of about 3 mol/L. Two-dimensional PEDRI sections were then sequentially acquired every 30 seconds, with each scan taking 27 seconds. At low field strength of 0.02 T (201 G), the electron-spin resonance and nuclear MR frequencies are 567 MHz and 856 KHz, respectively, and an Overhauser enhancement of –13 was achieved. (b) Three-dimensional gradient-echo PEDRI images of isolated beating rat heart infused with 3 mol/L TEMPONE. Top left image shows complete three-dimensional surface-rendered image; the other images are cutaways to show internal structure. The image took 4 minutes 30 seconds to acquire at 0.02 T. (Images courtesy of Haihong Li and Jay Zweier, MD, Davis Heart and Lung Research Institute, Ohio State University, Columbus, Ohio.)

View larger version (87K): [in a new window]   Figure 13b: Proton electron double resonance imaging (PEDRI; also known as Overhauser imaging). (a) Time course of PEDRI study of myocardial uptake of free radical probe TEMPONE (4-oxo-2,2,6,6-tetramethylpiperidine-1-oxyl) by isolated perfused rat heart. TEMPONE was infused through a side arm proximal to the perfusion cannula at final concentration of about 3 mol/L. Two-dimensional PEDRI sections were then sequentially acquired every 30 seconds, with each scan taking 27 seconds. At low field strength of 0.02 T (201 G), the electron-spin resonance and nuclear MR frequencies are 567 MHz and 856 KHz, respectively, and an Overhauser enhancement of –13 was achieved. (b) Three-dimensional gradient-echo PEDRI images of isolated beating rat heart infused with 3 mol/L TEMPONE. Top left image shows complete three-dimensional surface-rendered image; the other images are cutaways to show internal structure. The image took 4 minutes 30 seconds to acquire at 0.02 T. (Images courtesy of Haihong Li and Jay Zweier, MD, Davis Heart and Lung Research Institute, Ohio State University, Columbus, Ohio.)

View larger version (80K): [in a new window]   Figure 14: Thermogram of horse that had recently received an intramuscular injection in its neck. Skin in the region of injection is about 2°C warmer than elsewhere and shows up as a red area of increased brightness. (Image courtesy of Martin Furr, DVM, PhD, Virginia-Maryland Regional College of Veterinary Medicine.)

View larger version (64K): [in a new window]   Figure 15: Magnetoencephalographic spatial information superimposed on transverse T1-weighted MR sections (13/4.47). Magnetoencephalographic spike activity () in a 10-year-old patient with Landau-Kleffner syndrome. (Image courtesy of M. Funke, MD, PhD, Center for Advanced Medical Technologies, University of Utah, Salt Lake City, Utah.)

View larger version (62K): [in a new window]   Figure 16a: Biomolecular imaging. (a) Optical bioluminescence imaging of cardiac gene delivery shown on charge-coupled device optical images of three rats injected (directly into the lateral wall) with control adenovirus (Ad) (left), adenovirus carrying firefly luciferase (Fluc) driven by cytomegalovirus (CMV) promoter (middle), and adenovirus carrying luciferase driven by myosin light chain (MLC) promoter (right). Middle: Image shows expression in the myocardium and liver due to leakage of adenovirus into bloodstream from the injection site. Right: Image shows only cardiac expression of luciferase due to use of myosin light chain promoter, which is not markedly active in hepatocytes. Left: Control image shows background low activity. All images show light image of mouse onto which bioluminescence image is superimposed, in relative light units (RLU) per minute. (Image courtesy of Sanjiv Gambhir, MD, PhD, Stanford University, Palo Alto, Calif.) (b) Serial transverse PET images of reporter gene expression in a rat as function of time in days after dose injection. Top row: Animal was injected with 13N-labeled ammonia. Bottom row: Injected agent is 18F-labeled HBG gene. %ID = percent of injected dose. (Image courtesy of Heinrich Schelbert, MD, PhD, David Geffen School of Medicine, University of California at Los Angeles.)

View larger version (46K): [in a new window]   Figure 16b: Biomolecular imaging. (a) Optical bioluminescence imaging of cardiac gene delivery shown on charge-coupled device optical images of three rats injected (directly into the lateral wall) with control adenovirus (Ad) (left), adenovirus carrying firefly luciferase (Fluc) driven by cytomegalovirus (CMV) promoter (middle), and adenovirus carrying luciferase driven by myosin light chain (MLC) promoter (right). Middle: Image shows expression in the myocardium and liver due to leakage of adenovirus into bloodstream from the injection site. Right: Image shows only cardiac expression of luciferase due to use of myosin light chain promoter, which is not markedly active in hepatocytes. Left: Control image shows background low activity. All images show light image of mouse onto which bioluminescence image is superimposed, in relative light units (RLU) per minute. (Image courtesy of Sanjiv Gambhir, MD, PhD, Stanford University, Palo Alto, Calif.) (b) Serial transverse PET images of reporter gene expression in a rat as function of time in days after dose injection. Top row: Animal was injected with 13N-labeled ammonia. Bottom row: Injected agent is 18F-labeled HBG gene. %ID = percent of injected dose. (Image courtesy of Heinrich Schelbert, MD, PhD, David Geffen School of Medicine, University of California at Los Angeles.)