Numerous imaging modalities have been used to non-invasively assess fat concentration in living organs. Although each has its advantages and disadvantages, fat quantification using MRI has steadily increased in popularity and accuracy. MRI can also be used at the initial diagnosis without exposure to radiation. The method used, often called the Dixon method after its originator, analyses the signal from the protons to quantify the fat fraction (proton density fat fraction : PDFF). The basic principle of the Dixon method is to separate the signals from water and fat protons using the difference in resonance frequency between water and fat protons. This is done by taking images when the phases are aligned (in-phase images) and when the phases are 180 degrees out of phase (opposed-phase images). The Dixon method has always shown promise for clinical application, but there have been issues with image degradation due to differences in the relaxation time of fat and water protons and various artifacts. However, recent technological developments have improved its accuracy considerably. As a result of these advances, MRI fat quantification is now widely used in clinical practice and has achieved many results in the quantitative measurement of fat and iron deposition in the liver and other tissues, as well as in the differential diagnosis of tumors. In this review, the first half of the article will describe the fundamental principles and the development of the Dixon method, and the second half will show the method with clinical examples.