Figures 2, 2f: Virtual endodontics, periapical healing and four restorations.

Figure 2f shows a re-edited Figure 2 where virtual endodontic treatment was performed, the periapical radiolucency partially healed (smaller size) and four restorations placed. Nothing was really done except for computer simulations.

Figures 3, 3f: Computer-created crown and three carious lesions (molars).

Figure 3 shows distal decay in the upper second premolar and no other decay. Figure 3f demonstrates a virtual crown that “restored” distal caries in the upper second premolar; also note the virtual decay on the upper second (distal) and third (mesial) molars and mesial decay on the lower second molar.

Figures 4, 4f: Virtual removal of root, bone healing and creation of a 3-unit fixed bridge.

Figure 4f presents virtual treatment of Figure 4 where a root appears to have been extracted, partial healing of the socket accomplished and a 3-unit fixed bridge made — all within a few minutes of computer work.

Figures 5f: Computer-generated endodontics and crown on first premolar.

Figure 5f shows the addition of an extra endodontic filling and crown on the first bicuspid of the perfectly natural tooth in Figure 5.

Figures 6, 6f: Virtual tooth whitening.

Figure 6f shows whiter teeth than what actually appears in Figure 6. This can be used as proof of the power of “Virtual Product X,” which bleached the patient’s teeth in a few seconds — except the active ingredient came from the computer.

Sophisticated photos of dental products and techniques are appearing in dental literature in the form of advertisements or research treatment results. Many of these images are computer-generated and retouched.

Sophisticated photos of dental products and techniques are appearing in dental literature in the form of advertisements or research treatment results. Many of these images are computer-generated and retouched.

The discriminating practitioner must be able to identify the virtual from the real so that he or she can view the material presented on a sound, accurate scientific basis, rather than a virtual basis. There are a number of classic techniques one can use to detect computer enhancements in digitally produced photos and radiographs.

• If it looks too good to be true, it probably is. Wild claims (supported by impressive photos) that defy your own experience can indicate potential computer editing. Perfect margins, shading, alveolar bone regrowth and other signs of perfection should raise questions. Whether it is seen in a lecture on clinical dentistry or in a magazine or journal touting whitening, etc., one must be cautious. In the real world, nothing is perfect. Any photo that is becomes suspect.^5,6,7,8

If it looks too good to be true, it probably is. Wild claims (supported by impressive photos) that defy your own experience can indicate potential computer editing. Perfect margins, shading, alveolar bone regrowth and other signs of perfection should raise questions.

• Computers are made to make straight lines and perfect angles. This seldom occurs in real life and is a dead giveaway for deception. The straight edges and perfect angle of the virtual molar mesial occlusal restoration in Figure 1f is a very good example. Compare the irregular shape of real restorations in Figure 3 with the straight, even artifacts in Figure 1f. The four virtual restorations in Figure 2f are another example. They are too perfectly rounded and of similar size and shape — a rarity in real life. Close examination of the margin of the virtual crown in Figure 3f reveals a jagged border. This is an unrealistic attempt to eliminate a straight line, and thus detection that the image has been modified.^2,5,6
• Many digitally enhanced DR restorations appear as one solid color. Most natural restorations will show various degrees of radiopaqueness (density), as the restoration becomes thinner or thicker when placed on or in a real tooth. Actual restorations are seldom all one color or shade. Compare the real crown in Figure 5 or the restorations in Figure 3 to the restoration images of Figures 1f, 2f and 3f. There are subtle differences in each real restoration’s peripheral density.
• Look for “Frankenstein” images where, in many photos, the head of a subject appears to be stitched onto the body or a virtual tooth is inserted into a jaw. Changes in shading, angles and seams can identify this kind of manipulation. Relative uniform lighting requires that shadows appear in the same direction and on all areas of the photo. If they do not, then editing should be suspect. Look at the reflections of light on the skin of the patient in Figure 6 as compared to virtual Figure 6f. In Figure 6f, there are more highlights on the skin around the lips, indicating that the shade of the entire photo was lightened in order to make the teeth appear whiter.^{1,2,3,4,5,6,7,8}

Look at the reflections of light on the skin of the patient in Figure 6 as compared to virtual Figure 6f. In Figure 6f, there are more highlights on the skin around the lips, indicating that the shade of the entire photo was lightened in order to make the teeth appear whiter.

• Look for imperfections in geometry and whether the vanishing point is “off.” Check if the quality of the image is the same throughout the photo. If it is not, then something may have been added. Look for reoccurring patterns and objects, as well as discontinuities in the background. A lazy image editor may use standard shapes to fake restorations rather than spend the time to draw each restoration individually. Figure 2f shows virtual restorations of similar size and perfect shape. These were digitally enhanced.^5,6
• Often, enhancements to images can become apparent if you use “false color image” editing. Radically changing the contrast and brightness can also be telltale signs. Using your computer to radically change the colors (false coloring) of the suspected photo will serve to identify otherwise subtle changes. This process can be done in just a few seconds with most photo editing software.^5,6,7,8
• Look for artifacts of the editing process. These may include tracks, brush marks, cavities and unnatural compression. Rough texture transitions and uneven pixilation (the small dots that make up the image) across the image is strong evidence of editing.⁶
• Experienced image editors can detect metadata, short digital fingerprints formed on every digital image. In addition to storing information on the exact date of creation and the modification of a file, metadata can link the photo to the camera used, ownership (e.g., watermarking) and a variety of manipulations that may have been used to modify the raw image. Unfortunately, an image editor bent on deception can remove this data and thus play an infinite cat-and-mouse game of detecting and hiding or scrubbing evidence of image manipulation. Unless you wish to spend considerable sums of time and money for this service, this level of detective work is seldom practical in dentistry.^5,6,8
• Look at the size (megabytes) of the image file. Enhanced images will have a considerably larger file size than those that have no changes.

Because digital imaging has become so popular in dentistry, images of photos and radiographs may be edited to show different conditions than what is seen in reality. Photos can be enhanced to show whiter and straighter teeth, smoother skin or a well-blended margin on a crown. Digital radiographs can be changed to show additional pathology or treatment. There are numerous signs of this manipulation, which an informed practitioner can detect.

The authors wish to thank Andent Inc. for permission to republish the photos in this article.

Dr. Ellis Neiburger is a general practitioner in Waukegan, Illinois. Contact him at 847-244-0292 or eneiburger@comcast.net.

Yehonatan Frandzel is a forensic photography consultant based in Haifa, Israel.