Industries / Retail, hospitality and personal services / Beauty salons and spas / Tattoo studios
Removal
This work takes an unwanted tattoo out of the skin, today mostly with lasers whose very short pulses break the ink into pieces the body can carry away.
7
Milestones achieved, from 1979 to 2015.
2
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1979
Salt, abrasion and burning
Before lasers aimed at ink, a tattoo came out with the skin that held it. It was cut out, sanded off by dermabrasion with a motorized brush (W. Clabaugh described a superficial method in 1968), rubbed away with table salt, which is called salabrasion (A. M. Strong and I. T. Jackson, 1979, after papers by an author named Manchester in 1973 and 1974), or burned off with a carbon dioxide (CO2) laser, a method S. C. Brady and colleagues developed at Vancouver General Hospital in 1979. In a series of 51 patients treated with a CO2 laser, reported in 1989, 24 of the 44 followed up had excellent or good results after six months and three had raised (hypertrophic) scars.
LessonThe FDA says these methods wound the skin rather than remove the pigment, give highly variable results and carry a greater chance of scarring.
Credit
W. Clabaugh; A. M. Strong and I. T. Jackson; S. C. Brady and colleagues
1990
Short pulses clear black ink
A Q-switched laser stores its energy and releases it in pulses a few billionths of a second long, too short for the heat to spread far from the ink. At Massachusetts General Hospital, C. R. Taylor, R. Rox Anderson and colleagues treated 57 blue-black tattoos with a Q-switched ruby laser (694 nanometers, pulses of 40 to 80 nanoseconds) about every three weeks, at doses from 1.5 to 8.0 J/cm². Of the tattoos that finished the protocol, 18 of 23 amateur tattoos and 3 of 13 professional ones lightened substantially or cleared. One scarred, and small pale spots that did not go away were frequent.
Credit
C. R. Taylor, R. Rox Anderson and colleagues, Massachusetts General Hospital
1993
A longer wave for black
Melanin, the skin's own pigment, also absorbs ruby light, and pale spots were common after ruby treatment. S. L. Kilmer, R. Rox Anderson and colleagues at Harvard tested the Q-switched Nd:YAG laser, whose 1,064-nanometer light melanin absorbs less, in a blinded trial of 39 tattoos in 25 patients, each tattoo divided into quarters given different doses. After four sessions, 77% of black tattoos lost more than 75% of their ink, and no scarring or changes in skin color were seen. Bright colors barely responded.
LessonEach ink color absorbs its own wavelengths, so the FDA notes that multicolored tattoos may need several lasers.
Credit
S. L. Kilmer, R. Rox Anderson and colleagues
1993
Some inks turn black
R. Rox Anderson, Roy Geronemus, S. L. Kilmer and colleagues reported five tattoos, cosmetic, white, skin-colored and pink-red, that darkened the moment they were treated with Q-switched or pulsed lasers. Two could not be cleared afterwards and were cut out. Red cosmetic ink and ferric oxide, the rust-red pigment common in permanent makeup, blackened the same way in the laboratory, probably because the laser turns ferric oxide into black ferrous oxide.
LessonThe authors advised testing a small spot before treating such inks; Dutch dermal therapists surveyed in 2026 still reported this darkening in about 0.2% of sessions.
Credit
R. Rox Anderson, Roy Geronemus, S. L. Kilmer, W. Farinelli and R. E. Fitzpatrick
2009
A score predicts sessions
W. Kirby of Dr. Tattoff Inc. in Beverly Hills, A. Desai and colleagues reviewed the charts of 100 patients and proposed a score built from six features of a tattoo: the client's skin type, the tattoo's location, colors, amount of ink, scarring and layering. The total correlated with the number of sessions needed (correlation coefficient 0.757). Clinics in India (22 patients, 2019) and Israel (11 patients, 2024) later needed fewer sessions than the score predicted, 5.09 on average against 9.9 in the Israeli series, and a 2026 study says the score overestimates sessions with picosecond lasers.
LessonA score fitted to one generation of lasers drifts when the lasers change.
Credit
W. Kirby, A. Desai and colleagues
2012
Four passes in one visit
The first laser pulse turns treated skin white for a while, and the whitening blocks further pulses. T. Kossida, R. Rox Anderson and colleagues in Athens split 18 tattoos on 12 adults in half: one half got one pass of a Q-switched alexandrite laser, the other four passes 20 minutes apart, which they called the R20 method. Three months later, blinded reviewers of photographs favored the R20 half in every tattoo. Doctors in Düsseldorf (one case, 2014) and Baghdad (40 split tattoos, 2021) reported the same advantage from repeated passes, though both treated over three sessions. In a trial funded by ON Light Sciences, a patch soaked in perfluorodecalin, a clear liquid that reduces the whitening, let doctors fire 3.7 passes in five minutes against 1.4 without it.
Credit
T. Kossida, R. Rox Anderson and colleagues
2015
Laser light makes cyanide
BfR chemists fired a ruby laser at copper phthalocyanine blue, the common blue tattoo pigment, in liquid. It broke down into 1,2-benzenedicarbonitrile, benzonitrile, benzene and hydrogen cyanide, more as the dose rose, and the cyanide released from strong suspensions harmed skin cells. In 2018 the same institute reported that laser light splits organic tattoo pigments into carcinogens, among them 3,3'-dichlorobenzidine, which broke DNA strands in human skin cells. A group at Italy's National Research Council and the University of Rome Tor Vergata found in 2025 that femtosecond lasers break Pigment Green 7 into similar fragments and into rods and needles shaped like harmful fibers.
LessonHeating a pigment is no stand-in for a laser: in a 2021 test of Pigment Green 36, heating and laser light produced different fragments.
Credit
I. Schreiver, A. Luch and colleagues, BfR; H. Hering and colleagues, BfR
Watch for
A score for picosecond lasers
Researchers at Laserplast SrL in Milan and the Politecnico di Milano built a new score, the PT scale, from 545 patients treated at one center between 2019 and 2024 with one picosecond laser, the Discovery Pico made by Quanta System. Their best model predicted the number of sessions with an average error of 0.58 sessions and beat models built on the Kirby-Desai score. Two of the authors work for El.En. Group, and the data are available only on request.
Achieved whena clinic group with no tie to the authors or the laser maker scores the PT scale on its own patients and reports an average error under one session.
Claimed by
V. Pedrelli, E. Caiani and colleagues, July 2026
Lasers that fire on ink
A Purdue University chemistry group built SPOT, a system in which a separate guidance laser reads how much light the surface reflects and switches a nanosecond treatment laser on, within microseconds, only over dark targets. It removed printed letters from paper and lightened black ink tattooed into synthetic practice skin and into sections of pig skin, at a dose of about 270 mJ/cm², below the 1 to 10 J/cm² of clinical machines. The authors say only dead tissue was tested.
Achieved whenanother group repeats it on tattoos in living tissue at clinical doses and measures both clearance and damage to the skin around the ink.
Claimed by
C. D. Appel, C. Zhang and colleagues, Purdue University, September 2026
Open
01
Session estimates, tested elsewhere
An outside test of the two published scores that predict how many laser sessions a tattoo will need, the Kirby-Desai score and the 2026 PT scale, on de-identified records of finished removals from clinics that took part in neither study. It reads existing records only; nobody is treated or contacted for it.
The briefCredit · unclaimed02
Home removal lasers, catalogued
A public table of the tattoo removal lasers and "picosecond pens" offered to consumers on online marketplaces, giving what each listing claims (wavelength, pulse length, energy, intended user) against its FDA clearance or EU device registration, if it has one. The work reads listings and public databases only; nobody buys, fires or wears a device.
The briefCredit · unclaimed
Open 01 · the brief
Session estimates, tested elsewhere
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
An outside test of the two published scores that predict how many laser sessions a tattoo will need, the Kirby-Desai score and the 2026 PT scale, on de-identified records of finished removals from clinics that took part in neither study. It reads existing records only; nobody is treated or contacted for it.
Why it is still open
The Kirby-Desai score came from 100 charts at one practice, and its outside checks are small, 22 patients in India and 11 in Israel, the Israeli one co-written by an author from Alma Lasers Ltd. The PT scale comes from one center using one laser, with data on request. No test across several clinics and laser makers was found.
Built with what exists
A fixed form for scoring records, filled in blind to the outcome; REDCap or a shared spreadsheet; Python with scikit-learn to compute errors; an ethics committee approval for de-identified chart review at each clinic.
First step
Score 30 completed cases at one clinic on both scales from their intake photographs and notes, with the session counts hidden.
Done looks like
At least 300 completed removals from at least three clinics using lasers from at least two makers, each scored on both scales without sight of the session count, with the average error in sessions and the share predicted within one session reported for each clinic and each laser.
Open 02 · the brief
Home removal lasers, catalogued
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public table of the tattoo removal lasers and "picosecond pens" offered to consumers on online marketplaces, giving what each listing claims (wavelength, pulse length, energy, intended user) against its FDA clearance or EU device registration, if it has one. The work reads listings and public databases only; nobody buys, fires or wears a device.
Why it is still open
On 19 November 2026 a search of Amazon's US site for "picosecond laser pen tattoo removal" returned 64 results, among them a Q-switched Nd:YAG machine at 1,064 and 532 nanometers offered for salon or home use. The FDA says the lasers it has cleared for tattoo removal are for use by or under the supervision of a health care professional, and that it has approved no do-it-yourself removal kits. In a 2026 survey of 173 Dutch dermal therapists, 31% had seen clients who had tried other removal methods, and 56.7% of those had seen complications. No count of such devices by listing was found.
Built with what exists
Saved copies of marketplace search pages; the openFDA 510(k) interface; the public device search in EUDAMED, the EU's medical device database; a small model built for extraction, such as Meta's Muse Ember (Notes: "The jobs agents repeat"), to pull claims out of listing text, with every row checked by hand.
First step
Save the 64 Amazon US results for that search and record the claims in the first 20 listings.
Done looks like
At least 200 listings from at least three marketplaces in at least two regions, each with its claimed wavelength, pulse length and intended user, and a clearance or registration number where one exists, with 50 rows checked by a second person in full agreement.