The Melasma Area and Severity Index (MASI) was introduced by Kimbrough-Green et al. (1994) for a vehicle-controlled trial of topical tretinoin, adapting the structure of the PASI to the face. The face is divided into four regions weighted by their share of the facial surface: the forehead, the right malar region and the left malar region at 0.3 each, and the chin at 0.1. In each region the area of involvement is graded from 0 to 6 (0 none, 1 under 10%, 2 10 to 29%, 3 30 to 49%, 4 50 to 69%, 5 70 to 89%, 6 90 to 100%), the darkness of the pigmentation compared with the surrounding normal skin from 0 to 4, and the homogeneity of the pigmentation from 0 to 4. The MASI is the sum over the regions of the weight multiplied by the area grade and by the sum of darkness and homogeneity, from 0 to 48.

Pandya et al. (2011) examined the reliability and validity of the MASI with six trained raters scoring 21 patients over two days. Area and darkness were reliable within and between raters and agreed with computer-based area measurement and with mexameter readings of the pigment difference, whereas homogeneity showed the least agreement and could be removed without any loss of reliability. The modified MASI (mMASI) therefore drops homogeneity: each region contributes its weight multiplied by the area grade and the darkness grade, and the index runs from 0 to 24. The mMASI is now the usual primary outcome in melasma trials, including the oral tranexamic acid trial of Del Rosario et al. (2018) and the thiamidol trials of Arrowitz et al. (2019). This calculator implements the mMASI by default and the original MASI as a second mode; the darkness and homogeneity grades are labelled absent, slight, mild, marked and severe (the 1994 publication labels the highest grade maximum).

No severity cut-offs have been validated for either index. Rodrigues et al. (2016) correlated the scores with a four-grade melasma severity score in 79 patients and reported mean mMASI values of 3.8 for mild, 6.5 for moderate and 8.9 for severe melasma (mean MASI 6.9, 12.4 and 20.2), with confidence intervals that overlap between neighbouring grades; the bands of 0 to 8, 8 to 16 and 16 to 24 that appear in some trial registrations are arbitrary and inconsistent with those data. Because most of the range is occupied by unusual disease, the result card shows the score without a severity label, and response is expressed as the change and the percentage change from baseline, which is how melasma trials report it. No minimal clinically important difference has been published for the MASI or the mMASI.

Both indices weight extent more heavily than pigment intensity, so a large faint patch can outscore a small dark one, and neither captures dermal versus epidermal pigment, relapse tendency or the impact on quality of life. Grading darkness is affected by lighting and by the surrounding skin tone, so serial assessments are best made by the same rater under standard lighting, ideally with standardised photographs. Split-face studies use hemi-MASI and hemi-mMASI variants with the regional weights halved (forehead 0.15, malar 0.3, chin 0.05), giving maxima of 24 and 12 for half of the face.