Quantitative differentiation of malignant and benign thyroid nodules with multi-parameter diffusion-weighted imaging

Figure 2 Data from a 35-year-old male with a malignant nodule of follicular thyroid carcinoma in the right lateral of thyroid. A: T2WI shows the lesion in the right lateral of thyroid (orange arrow). Multiparametric diffusion parameter maps fitted by mono-exponential [B = ADC_{(0 and 1000)}; C = ADC_{(0 to 1000)}; D = ADC_{(0 and 2000)}; E = ADC_{(0 to 2000)}], bi-exponential (F = D; G = D*; H = f), stretched exponential (I = DDC; J = α), and kurtosis diffusion-weighted imaging (K = Dapp; L = Kapp) models show the lesion in the right lateral of thyroid (orange arrow); M: Histopathological H&E image (original magnification, 40×) demonstrate high cell density (stars) and capsular invasion (write arrow); N: Plot of the decay of diffusion weighted signal intensity fitted by a mono-exponential (green), biexponential (blue), stretched-exponential (dark), and kurtosis (red) models for the nodule; O: Plot of the decay of diffusion weighted signal intensity fitted by a mono-exponential with b = 0 and 1000 (dark), b = 0 to 1000 (red), b = 0 and 2000 (blue), and b = 0 to 2000 (green) for the nodule. ADC: Apparent diffusion coefficient; D: Slow diffusion coefficient; D*: Fast diffusion coefficient; f: Fraction of fast diffusion; DDC: Distributed diffusion coefficient; α: Anomalous exponent term; Dapp: Mean diffusivity; Kapp: Mean kurtosis; ROC: Receiver operating characteristic.