Near-IR imager detects early dental decay

Researchers from the University of California San Francisco (UCSF) have developed a near-IR imaging system that detects dental decay earlier and more accurately than x-rays or other transillumination sources.

Nov 1st, 2003
Th 137444

Researchers from the University of California San Francisco (UCSF) have developed a near-IR imaging system that detects dental decay earlier and more accurately than x-rays or other transillumination sources. Funded by a National Institutes of Health grant, Daniel Fried and colleagues at UCSF are exploring near-IR imaging of tooth structure for the early detection of dental caries using near-IR transillumination and polarization-sensitive optical-coherence tomography. Through this work they have identified the 1310-nm wavelength as ideal for diagnostic imaging of hard dental tissue.

"What is special about 1310 nm is that it is where you have the maximum transmission through dental enamel," said Fried, associate professor in the department of preventive and restorative dental sciences at UCSF. "We have measured the transmission of light through dental hard tissue, and enamel is almost completely transparent at this wavelength because of the decrease in light scattering. To see early decay, you only need to be able to see through the enamel."

Fried says the near-IR approach has several advantages over other imaging techniques, including conventional x-rays and visible-imaging devices. Dental decay begins on the surface of the tooth enamel, but x-rays are not sensitive enough to pick it up before it progresses to the dentin (in the interior of the tooth). Prior to the development of x-rays, dentists used simple transillumination techniques to search for caries lesions; with the development of high-intensity fiberoptic illumination sources, the dental community is once again looking to transillumination to give them a more effective tool for identifying tooth decay at an earlier stage. The U.S. Food and Drug Administration (FDA) recently granted marketing clearance to the DIFOTI (Digital Imaging Fiberoptic Transillumination) system from Electro-Optical Sciences (Irvington, NY), which uses visible light for transillumination. While this is a step in the right direction, Fried says visible wavelengths have their own restrictions when trying to image hard tissue.

"Ideally, you want the light to go through the tooth, which is the advantage of the near-IR," Fried said. "Visible light is attenuated by the light scattering; in the visible range, the scattering coefficient is around 80 to 100 cm-1. At 1310 nm, the attenuation coefficient drops to 2 to 3 cm-1."

Experimental results

In experiments involving extracted human teeth containing simulated lesions, Fried and his colleagues demonstrated the efficacy of a 1310-nm imaging system for detecting and imaging interproximal caries lesions. The experimental system comprises a fiber-coupled broadband light source (either a 150-W halogen lamp or superluminescent diodes), two crossed linear polarizers for polarization gating, an indium gallium arsenide focal-plane array, and a near-IR camera. A 50-nm-bandpass filter centered at 1310 nm was used to remove all light outside the spectral region of interest. A tooth section of minimal sample thickness (3 mm) was chosen for comparison of the near-IR transillumination system with visible-light fiberoptic transillumination and x-rays; a hole was drilled into each tooth section and filled with hydroxyapatite powder to simulate decay. The near-IR results showed significant contrast between the lesion and the enamel (greater than 0.35) and a spatial line profile that clearly resolves the lesion (see figure). The researchers also worked with thicker samples, up to 6.75 mm, with similar results.1

Near-IR (top), x-ray (center), and visible (bottom) images show a simulated lesion (yellow box) placed in a 3-mm-thick tooth section. The lesion cannot be seen using visible-light transillumination but is clearly visible with near-IR transillumination. The x-ray image shows poor contrast between the lesion and surrounding enamel.
Click here to enlarge image

One drawback of the near-IR system is that, despite a fairly straightforward design, much of the technology is still expensive; while superluminescent diodes have dropped to less than $1000, the near-IR camera costs $25,000, Fried points out. He and his colleagues are now working to refine the setup and to do additional experiments on teeth with natural lesions. The next step beyond that will be in-vivo experiments. A near-IR dental imager could be on the market in three to five years, depending on investment from a commercial partner willing to do the necessary product development and FDA trials, according to Fried.

"FDA clearance shouldn't take that long because this is a safe technique with eye-safe light-intensity levels," he said.


  1. R. Jones, G. Huynh, G. Jones, D. Fried, Optics Express 11(18; Sept. 8, 2003).

More in Detectors & Imaging