| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Science to Practice |
Molecular Imaging Program,
National Cancer Institute,
10 Center Dr, MSC 1182,
Bldg 10, Room 1B40,
Bethesda, MD 20892-1088,
pchoyke@nih.gov
SUMMARY
Quantitative fluorescence molecular imaging can be performed by compensating for variations in the light intensity caused by a change in the position of the endoscope. By accounting for these differences, quantitation of fluorescence can be used to assess the biologic importance of a lesion that might not otherwise be detected with white light endoscopy. This area of inquiry represents an important new step toward the detection of disease in its earliest phases when it can be treated with minimally invasive surgery or ablation.
THE SETTING
In traditional endoscopy, white light (WL) is transmitted through an optical fiber to illuminate a body surface. Reflected light is then transmitted back through a different part of the fiber and detected with a video camera. Other "physical channels" include an insufflation channel that is used to distend and clear the field and a working channel through which instruments are inserted. However, new developments in optical imaging permit the use of "optical channels," such as fluorescence imaging, to detect light emitted at a specific wavelength by a fluorophore (ie, a molecule that emits light of one wavelength when excited by light of another wavelength). If the fluorophore is conjugated to a targeting molecule and it accumulates within a particular pathologic entity (eg, a tumor), a highly specific molecular diagnosis can be made without the need for biopsy. Moreover, if the fluorescence is generated in the near-infrared range, it can be detected up to several centimeters below the surface being examined. Thus, this optical channel not only opens the possibility for accurate real-time diagnosis but also extends the field of optical endoscopy below the mucosa. However, owing to motion, the intensity of the detected light varies with the proximity and angle of the endoscope relative to the surface. Physicians can intuitively adjust to rapidly changing WL conditions, and the effects of these changes can be reduced with automatic exposure control; however, physicians are less able to adjust for alterations in fluorescent image intensity. Therefore, the degree of fluorescence can be greatly over- or underestimated depending on the lighting conditions. In this issue of Radiology, Upadhyay et al (1) provide a solution to this limitation of fluorescence endoscopy.
|
Like many good ideas, once the solution to a problem is known, one is struck by its simple elegance. In brief, Upadhyay et al (1) designed a dual-head camera with which WL and near-infrared fluorescent images were acquired simultaneously. The fluorescent images were corrected in real time by dividing the fluorescence pixel intensity by the corresponding WL pixel intensity. For example, if the intensity of the image was inordinately high because the endoscope was close to the surface, the fluorescent image was adjusted by dividing each pixel by the WL pixel intensity. Thus, the fluorescent image maintained uniform signal intensity (±10%) even while the signal intensity of the WL image varied as the endoscope was moved. By considering the exposure time, fluorescence intensity (measured in counts per second) can be quantitated in a manner that enables estimation of the concentration of the fluorophore at the target. In two animal models (peritoneal cancer and colon cancer), Upadhyay et al (1) used a fluorescent molecule (Prosense; VisEn Medical, Woburn, Mass) as a "smart" optical near-infrared range probe that became fluorescent in the presence of specific proteases, such as cathepsin B, to depict cancer sites (2). Low concentrations of the fluorescent molecule were detectable despite adverse lighting conditions.
THE PRACTICE
Clinical use: The medical uses of optical imaging are revolutionizing medicine; optical imaging is undergoing explosive growth fueled by advances in high-sensitivity detectors, improved optics, and developments in molecular biology. For example, it is possible that future screening colonoscopy will involve the use of not only current WL imaging but also fluorescent imaging aided by target fluorescent probes directed at colon cancers. This may enable the detection of early polyps that might otherwise be missed and the avoidance of lesions that might otherwise be examined with biopsy. With use of fluorescent probes, colonoscopy studies may be easier to interpret and colonoscopy examinations may require less bowel preparation, thus lowering barriers to screening (3).
Future opportunities and challenges: Fluorescence imaging is notoriously difficult to quantify; however, the method described by Upadhyay et al (1) represents an important first step by normalizing light intensity variations caused by the position of the endoscope. It may be possible to determine threshold values of fluorescence that will be used to distinguish benign from malignant processes. Given the broad variations in tumor grade, this may require more than one fluorescent label or the addition of other optical classifiers to correctly categorize endoscopic lesions by their "optical signature" (4).
FOOTNOTES
See also the article by Upadhyay et al in this issue.
References
Related Article
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
