Radiographic fog

Radiographic fog is unwanted density spread across a radiograph that raises the overall darkness and lowers contrast, leaving the image flat and grey. Scattered and secondary radiation fog any receptor. On a phosphor plate the digital-era fault runs the opposite way — ambient light erases the latent image, and that area reads whiter and noisier.

Synonyms: film fog; chemical fog; scatter fog; light leak; image fog

What is radiographic fog?

Fog is density that carries no information. It sits on top of the real image and closes the gap between light and dark, which is what contrast is. The term is film-era, and it was measured there: fog built up in the emulsion and was recorded as base plus fog, in optical density units. Faster film built it faster, with a matching fall in image contrast [4].

The mechanism that still fogs a modern receptor is scatter. Professional guidance from the FDA and ADA puts it plainly — a benefit of rectangular collimation is "an improvement in contrast as a result of a reduction in fogging caused by secondary and scattered radiation," and the same document asks for collimation of the beam to the size of the receptor whenever feasible [5]. Scatter does not care what is behind the collimator. The same guidance names the receptors in use — conventional film, CCD and CMOS solid-state detectors, and phosphor plates — and states the collimation benefit without qualifying it to any of them [5].

So fog belongs in the current list of radiographic artifacts, not in a history chapter. What has changed is that a second fault now sits beside it, looks similar at a glance, and runs in the opposite direction.

What does fog look like on a radiograph?

Flat. Every structure moves a little closer to every other in density, so enamel stops standing away from dentine and the trabecular pattern goes soft. Scatter fog adds density, so the whole image is a shade more radiolucent than it should be while the anatomy stays in the right places.

Ambient light on a phosphor plate looks different, and this is the appearance most clinicians are actually meeting. In a retrospective review of 5,000 intraoral periapical images from a phosphor plate system, whitening was the second most common error recorded, after operator errors [1]. It has three tells. Signal loss starts at the periphery, because the photons reaching the edge of the plate had lower intensity than those reaching the centre [1]. The highest-attenuation areas go strange — metallic restorations and fixed prostheses "lose photons and have a shiny appearance when exposed to ambient light" [1]. And light reaching the plate through something patterned prints that pattern: when a plate was exposed through a sheet of paper with writing on it, more light passed through the unwritten areas and the text appeared on the latent image [1].

Expect it to be common. In a clinic audit of 11,443 intraoral phosphor plate images, 2,344 (20.4%) carried at least one artifact, and non-uniform image brightness was the single most frequent artifact in adult patients [8].

What causes it on digital receptors?

Two mechanisms, and keeping them apart is most of the diagnosis.

Scatter fogs any receptor. It is the receptor-agnostic one, it is reduced by collimating to receptor size, and reducing it is what improves contrast [5].

Ambient light damages a phosphor plate, and only in one window. A phosphor plate holds a latent image between exposure and read-out. During that window it is light-sensitive: "If the scanning of an exposed PSP is delayed without protection from the ambient light, latent electrons in the image are released spontaneously. These electrons cause noisy and whiter image views" [1]. In practice the cause is usually mundane — the artifacts were mainly traced to taking the plate out of its protective cover after exposure, in order to disinfect it [1].

The window is short. One 2023 study radiographed six regions of a dry human skull with two phosphor plate systems, VistaScan and Express, and exposed the plates to ambient light for 0, 5, 10, 30, 60 or 90 seconds before read-out. Six observers rated the 288 resulting radiographs, rejection rates rose mostly from 10 seconds, and the authors proposed a safe limit of 5 seconds for the two systems they tested [2]. That is a laboratory figure for two named systems under one lab's lighting, not a number to quote as a clinical rule.

A solid-state sensor has no equivalent window — it reads out immediately rather than carrying a latent image to a scanner [5]. Nothing retrieved for this entry describes a CCD or CMOS intraoral sensor fogging from room light, and the mechanism above is specific to the waiting period. On a solid-state sensor, the fog to think about is scatter.

How do you tell fog from underexposure?

Start with the direction of the shift, then check whether it is even.

On film, fog raises density — the compensations tested in the classic experiment were needed precisely because fog produced "the increased density caused by fog" [4]. Underexposure does the reverse. So on film the direction alone separates them.

On a phosphor plate it does not, because ambient light takes signal away rather than adding it, and a light-damaged plate reads whiter, like an underexposed one [1]. Three things separate them there. Distribution: light damage begins at the periphery [1]. Metal: restorations take on a shiny appearance under light damage [1]. History: the plate came out of its cover, or sat before scanning.

Recoverability is the fourth test, and it is worth knowing before you start dragging sliders. Phosphor plates have a wide dynamic range that gives real latitude to correct under- and overexposed images [1]. Light damage is a different problem. In a step-wedge study, digital enhancement restored 23.48% of brightness for the VistaScan system but under 1% for the Express; it restored 53.25% of contrast for the VistaScan and 5.79% for the Express, and 40.71% and 35% of contrast-to-noise ratio respectively [3].

What it means clinically

A foggy radiograph is not automatically a non-diagnostic one, and the most useful evidence here is old and counter-intuitive. In a 1990 study, standardised fog was simulated on E-speed film imaging extracted teeth with and without approximal caries, and nine dentists scored them. Fog up to a base plus fog of 0.6 optical density units had no influence on diagnostic accuracy when nothing was done to compensate. When exposure was lowered to hold density constant, accuracy fell significantly. Observers ranked the foggier radiographs as lower quality, but no significant correlation was found between that ranking and the accuracy they actually achieved [4]. This is in vitro, on extracted teeth and film-era receptors, so read the numbers as being about film. Read the direction as a warning: turning the exposure down to make a flat image look normal is what cost accuracy.

Light damage on a plate is a different story, because it removes information rather than laying a veil over it. Both groups studying it reach the same practical point — plates exposed to light drive retakes and additional patient exposure [2][3]. Across 50,000 intraoral phosphor plate radiographs taken over two years, 3,550 (7.1%) were retaken because of artifacts, and delayed scanning was the most common scanning error, at 145 images [6]. If a set of bitewings comes back flat and pale at the edges, the decision to retake is a dose decision. That is the argument for preventing it rather than managing it.

How do you prevent it?

• Collimate to the size of the receptor. It reduces the fogging caused by secondary and scattered radiation, and improves contrast as a result [5] — and it is the dose-limiting move anyway.
• Keep the plate covered until it enters the scanner. Manufacturers recommend opening the adhesive side of the protective cover and sliding the plate out of it directly into the scanner input unit, with the light-protection cover in place, specifically to prevent handling and ambient-light artifacts [1].
• Scan promptly. The exposed plate is the vulnerable object, and delay under room light is what damages it [1][2].
• Don't disinfect an exposed plate at the chairside. Uncovering it to clean it was the main traced cause of ambient-light artifacts in a 5,000-image review [1].
• Don't reduce exposure to correct a flat image, and don't assume enhancement will rescue a light-damaged one [3][4].

Common errors

Reading a flat, pale plate image as underexposure and pushing the exposure up on the retake, when the plate was simply left out of its cover. Treating the 5-second figure as a rule rather than as one lab's finding on two named systems [2]. Trusting the enhancement button — on the Express system it restored under 1% of brightness [3].