A combination of Fourier transform infrared spectroscopy (FTIR) and multivariate statistics (chemometry) was applied as a screening tool for the quantitative determination of carbohydrates, such as glucose, fructose and sucrose, in 11 processed commercial apple juices and 2 genuine juices obtained from squeezed apples. For calibration, a number of 24 mixtures of glucose, fructose and sucrose solutions (synthetic samples), at different concentrations were prepared and scanned in the 885 and 1500 cm-1 spectral range, using attenuated total reflectance (ATR) FTIR spectroscopy. Based on the mixtures spectra in the mid IR (MIR) region, we realized calibration models for each carbohydrate, using partial least squares (PLS) regression. The models were then used to predict the glucose, fructose and sucrose concentration in commercial apple juices, comparatively with concentrations in fresh, genuine juices, in order to assess the juice authenticity. The glucose concentrations (%, w/w) predicted for commercial juices ranged from 1.664 to 3.133 versus 3.1 for genuine fresh juices. The fructose concentrations (%, w/w) predicted for commercial juices ranged from 3.701 to 6.941 versus 9.2 for genuine fresh juices, while the sucrose concentrations (%, w/w) predicted for commercial juices ranged from 0.746 to 5.795 versus 1.38 for genuine fresh juices. The standard deviations of most predicted values are below 10%. Most juices exhibited glucose, fructose and sucrose concentrations in the expected range. However, several samples showed discrepancies from average concentration values, thus the authenticity of these juices could not be confirmed. Also, high sucrose concentration can flag adulterated juices, or indicates sucrose addition to maintain the juice sweetness intensity. Our results indicate ATR-MIR spectroscopy to be a rapid, accurate, non-destructive and cost-effective tool for routine monitoring of multiple constituents in apple juices, as quality and safety indicators.