ABSTRACT
OBJECTIVES: To develop an [18F]FDG PET/3D-UTE model based on clinical factors, three-dimensional ultrashort echo time (3D-UTE), and PET radiomics features via machine learning for the assessment of lymph node (LN) status in non-small cell lung cancer (NSCLC). METHODS: A total of 145 NSCLC patients (training, 101 cases; test, 44 cases) underwent whole-body [18F]FDG PET/CT and chest [18F]FDG PET/MRI were enrolled. Preoperative clinical factors and 3D-UTE, CT, and PET radiomics features were analyzed. The Mann-Whitney U test, LASSO regression, and SelectKBest were used for feature extraction. Five machine learning algorithms were used to establish prediction models, which were evaluated by the area under receiver-operator characteristic (ROC), DeLong test, calibration curves, and decision curve analysis (DCA). RESULTS: A prediction model based on random forest, consisting of four clinical factors, six 3D-UTE, and six PET radiomics features, was used as the final model for PET/3D-UTE. The AUCs of this model were 0.912 and 0.791 in the training and test sets, respectively, which not only showed different degrees of improvement over individual models such as clinical, 3D-UTE, and PET (AUC-training = 0.838, 0.834, and 0.828, AUC-test = 0.756, 0.745, and 0.768, respectively) but also achieved the similar diagnostic efficacy as the optimal PET/CT model (AUC-training = 0.890, AUC-test = 0.793). The calibration curves and DCA indicated good consistency (C-index, 0.912) and clinical utility of this model, respectively. CONCLUSION: The [18F]FDG PET/3D-UTE model based on clinical factors, 3D-UTE, and PET radiomics features using machine learning methods could noninvasively assess the LN status of NSCLC. CLINICAL RELEVANCE STATEMENT: A machine learning model of 18F-fluorodeoxyglucose positron emission tomography/ three-dimensional ultrashort echo time could noninvasively assess the lymph node status of non-small cell lung cancer, which provides a novel method with less radiation burden for clinical practice. KEY POINTS: ⢠The 3D-UTE radiomics model using the PLS-DA classifier was significantly associated with LN status in NSCLC and has similar diagnostic performance as the clinical, CT, and PET models. ⢠The [18F]FDG PET/3D-UTE model based on clinical factors, 3D-UTE, and PET radiomics features using the RF classifier could noninvasively assess the LN status of NSCLC and showed improved diagnostic performance compared to the clinical, 3D-UTE, and PET models. ⢠In the assessment of LN status in NSCLC, the [18F]FDG PET/3D-UTE model has similar diagnostic efficacy as the [18F]FDG PET/CT model that incorporates clinical factors and CT and PET radiomics features.
Subject(s)
Carcinoma, Non-Small-Cell Lung , Lung Neoplasms , Humans , Fluorodeoxyglucose F18 , Carcinoma, Non-Small-Cell Lung/diagnostic imaging , Carcinoma, Non-Small-Cell Lung/surgery , Positron Emission Tomography Computed Tomography , Radiomics , Lung Neoplasms/diagnostic imaging , Positron-Emission Tomography , Machine Learning , Magnetic Resonance Imaging , Lymph Nodes/diagnostic imaging , Retrospective StudiesABSTRACT
Background: Lung cancer has become one of the deadliest tumors in the world. Non-small cell lung cancer (NSCLC) is the most common type of lung cancer, accounting for approximately 80%-85% of all lung cancer cases. This study aimed to investigate the value of diffusion kurtosis imaging (DKI), diffusion-weighted imaging (DWI) and 2-[18F]-fluoro-2-deoxy-D-glucose positron emission tomography (18F-FDG PET) in differentiating squamous cell carcinoma (SCC) and adenocarcinoma (AC) and to evaluate the correlation of each parameter with stage and proliferative status Ki-67. Methods: Seventy-seven patients with lung lesions were prospectively scanned by hybrid 3.0-T chest 18F-FDG PET/MR. Mean kurtosis (MK), mean diffusivity (MD), apparent diffusion coefficient (ADC), maximum standard uptake value (SUVmax), metabolic tumor volume (MTV) and total lesion glycolysis (TLG) were measured. The independent samples t test or Mann-Whitney U test was used to compare and analyze the differences in each parameter of SCC and AC. The diagnostic efficacy was evaluated by receiver operating characteristic (ROC) curve analysis and compared with the DeLong test. A logistic regression analysis was used for the evaluation of independent predictors. Bootstrapping (1000 samples) was performed to establish a control model, and calibration curves and ROC curves were used to validate its performance. Pearson's correlation coefficient and Spearman's correlation coefficient were calculated for correlation analysis. Results: The MK and ADC values of the AC group were significantly higher than those of the SCC group (all P< 0.05), and the SUVmax, MTV, and TLG values of the SCC group were significantly higher than those of the AC group (all P<0.05). There was no significant difference in the MD value between the two groups. Moreover, MK, SUVmax, TLG and MTV were independent predictors of the NSCLC subtype, and the combination of these parameters had an optimal diagnostic efficacy (AUC, 0.876; sensitivity, 86.27%; specificity, 80.77%), which was significantly better than that of MK (AUC = 0.758, z = 2.554, P = 0.011), ADC (AUC = 0.679, z = 2.322, P = 0.020), SUVmax (AUC = 0.740, z = 2.584, P = 0.010), MTV (AUC = 0.715, z = 2.530, P = 0.011) or TLG (AUC = 0.716, z = 2.799, P = 0.005). The ROC curve showed that the validation model had high accuracy in identifying AC and SCC (AUC, 0.844; 95% CI, 0.785-0.885);. The SUVmax value was weakly positively correlated with the Ki-67 index (r = 0.340, P< 0.05), the ADC and MD values were weakly negatively correlated with the Ki-67 index (r = -0.256, -0.282, P< 0.05), and the MTV and TLG values were weakly positively correlated with NSCLC stage (r = 0.342, 0.337, P< 0.05). Conclusion: DKI, DWI and 18F-FDG PET are all effective methods for assessing the NSCLC subtype, and some parameters are correlated with stage and proliferation status.
ABSTRACT
Background: Noninvasive assessment of programmed death-ligand 1 (PD-L1) expression status in non-small cell lung cancer (NSCLC) is necessary. This study arm to investigate the value of 2-[18F]-fluoro-2-deoxy-D-glucose positron emission tomography (18F-FDG PET), diffusion-weighted imaging (DWI), intravoxel incoherent motion (IVIM), and amide proton transfer-weighted imaging (APTWI) in the assessment of PD-L1 status in NSCLC. Methods: This is a prospective diagnostic study. A total of 76 patients with NSCLC underwent chest 18F-FDG PET/magnetic resonance imaging (MRI). Parameters maximum standardized uptake value (SUVmax), quantitate the metabolic tumor volume (MTV), total lesion glycolysis (TLG), apparent diffusion coefficient (ADC), diffusion coefficient (D), pseudo diffusion coefficient (D*), and perfusion fraction (f), and magnetization transfer ratio asymmetry at 3.5 ppm [MTRasym (3.5 ppm)] from 18F-FDG PET, DWI, IVIM, and APTWI, respectively, were compared. The optimal combination of parameters was investigated using logistic regression models and evaluated by area under the receiver operating characteristic (ROC) curve (AUC). The bootstrap with 1,000 samples was used for model validation. Results: SUVmax, MTV, TLG, and MTRasym (3.5 ppm) were higher and D and f were lower in PD-L1 positive NSCLC than in PD-L1 negative NSCLC (all P<0.05). Logistic analysis showed that the combination of MTRasym (3.5 ppm), D, and SUVmax had the strongest predictive value for the differentiation of PD-L1 positive and PD-L1 negative NSCLC [AUC, 0.946; 95% confidence interval (CI): 0.869-0.985; sensitivity, 85.29%; specificity, 91.67%; P all <0.001]. The verification model showed the combination of MTRasym (3.5 ppm), D, and SUVmax had the strongest predictive value, and its ROC curve and calibration curve showed good accuracy (AUC, 0.919, 95% CI: 0.891-0.937) and consistency. Conclusions: Multi-parametric 18F-FDG PET/MRI is beneficial for the non-invasive assessment of PD-L1 status in NSCLC patients, and the combination of SUVmax, D, and MTRasym (3.5 ppm) may serve as a prognostic biomarker to guide immunotherapy.
ABSTRACT
Objective: To explore the value of PET/MRI, including diffusion kurtosis imaging (DKI), diffusion weighted imaging (DWI) and positron emission tomography (PET), for distinguishing between benign and malignant solitary pulmonary lesions (SPLs) and predicting the histopathological grading of malignant SPLs. Material and methods: Chest PET, DKI and DWI scans of 73 patients with SPL were performed by PET/MRI. The apparent diffusion coefficient (ADC), mean diffusivity (MD), mean kurtosis (MK), maximum standard uptake value (SUVmax), metabolic total volume (MTV) and total lesion glycolysis (TLG) were calculated. Student's t test or the Mann-Whitney U test was used to analyze the differences in parameters between groups. Receiver operating characteristic (ROC) curves were used to evaluate the diagnostic efficacy. Logistic regression analysis was used to evaluate independent predictors. Results: The MK and SUVmax were significantly higher, and the MD and ADC were significantly lower in the malignant group (0.59 ± 0.13, 10.25 ± 4.20, 2.27 ± 0.51[×10-3 mm2/s] and 1.35 ± 0.33 [×10-3 mm2/s]) compared to the benign group (0.47 ± 0.08, 5.49 ± 4.05, 2.85 ± 0.60 [×10-3 mm2/s] and 1.67 ± 0.33 [×10-3 mm2/s]). The MD and ADC were significantly lower, and the MTV and TLG were significantly higher in the high-grade malignant SPLs group (2.11 ± 0.51 [×10-3 mm2/s], 1.35 ± 0.33 [×10-3 mm2/s], 35.87 ± 42.24 and 119.58 ± 163.65) than in the non-high-grade malignant SPLs group (2.46 ± 0.46 [×10-3 mm2/s], 1.67 ± 0.33[×10-3 mm2/s], 20.17 ± 32.34 and 114.20 ± 178.68). In the identification of benign and malignant SPLs, the SUVmax and MK were independent predictors, the AUCs of the combination of SUVmax and MK, SUVmax, MK, MD, and ADC were 0.875, 0.787, 0.848, 0.769, and 0.822, respectively. In the identification of high-grade and non-high-grade malignant SPLs, the AUCs of MD, ADC, MTV, and TLG were 0.729, 0.680, 0.693, and 0.711, respectively. Conclusion: DWI, DKI, and PET in PET/MRI are all effective methods to distinguish benign from malignant SPLs, and are also helpful in evaluating the pathological grading of malignant SPLs.
ABSTRACT
Objective: To evaluate the application value of monoexponential, fractional order calculus (FROC) diffusion models and PET imaging to distinguish between benign and malignant solitary pulmonary lesions (SPLs) and malignant SPLs with different pathological types and explore the correlation between each parameter and Ki67 expression. Methods: A total of 112 patients were enrolled in this study. Prior to treatment, all patients underwent a dedicated thoracic 18F-FDG PET/MR examination. Five parameters [including apparent diffusion coefficient (ADC) derived from the monoexponential model; diffusion coefficient (D), a microstructural quantity (µ), and fractional order parameter (ß) derived from the FROC model and maximum standardized uptake value (SUVmax) derived from PET] were compared between benign and malignant SPLs and different pathological types of malignant SPLs. Independent sample t test, Mann-Whitney U test, DeLong test and receiver operating characteristic (ROC) curve analysis were used for statistical evaluation. Pearson correlation analysis was used to calculate the correlations between Ki-67 and ADC, D, µ, ß, and SUVmax. Results: The ADC and D values were significantly higher and the µ and SUVmax values were significantly lower in the benign group [1.57 (1.37, 2.05) µm2/ms, 1.59 (1.52, 1.72) µm2/ms, 5.06 (3.76, 5.66) µm, 5.15 ± 2.60] than in the malignant group [1.32 (1.03, 1.51) µm2/ms, 1.43 (1.29, 1.52) µm2/ms, 7.06 (5.87, 9.45) µm, 9.85 ± 4.95]. The ADC, D and ß values were significantly lower and the µ and SUVmax values were significantly higher in the squamous cell carcinoma (SCC) group [1.29 (0.66, 1.42) µm2/ms, 1.32 (1.02, 1.42) µm2/ms, 0.63 ± 0.10, 9.40 (7.76, 15.38) µm, 11.70 ± 5.98] than in the adenocarcinoma (AC) group [1.40 (1.28, 1.67) µm2/ms, 1.52 (1.44, 1.64) µm2/ms, 0.70 ± 0.10, 5.99 (4.54, 6.87) µm, 8.76 ± 4.18]. ROC curve analysis showed that for a single parameter, µ exhibited the best AUC value in discriminating between benign and malignant SPLs groups and AC and SCC groups (AUC = 0.824 and 0.911, respectively). Importantly, the combination of monoexponential, FROC models and PET imaging can further improve diagnostic performance (AUC = 0.872 and 0.922, respectively). The Pearson correlation analysis showed that Ki67 was positively correlated with µ value and negatively correlated with ADC and D values (r = 0.402, -0.346, -0.450, respectively). Conclusion: The parameters D and µ derived from the FROC model were superior to ADC and SUVmax in distinguishing benign from malignant SPLs and adenocarcinoma from squamous cell carcinoma, in addition, the combination of multiple parameters can further improve diagnostic performance. The non-Gaussian FROC diffusion model is expected to become a noninvasive quantitative imaging technique for identifying SPLs.
ABSTRACT
BACKGROUND: Lung cancer is one of the most devastating diseases worldwide, and it is meaningful to assess the subtype and epidermal growth factor receptor (EGFR) status in lung cancer noninvasively by imaging methods. PURPOSE: To differentiate noninvasively small cell lung cancer (SCLC) from nonsmall cell lung cancer (NSCLC), and EGFR mutation-type from wild-type NSCLC by comparing amide proton transfer-weighted imaging (APTWI), diffusion-weighted imaging (DWI), and 2-[18 F]-fluoro-2-deoxy-d-glucose positron emission tomography (18 F-FDG PET). STUDY TYPE: Prospective. POPULATION: A total of 99 patients with lung cancer. FIELD STRENGTH/SEQUENCE: APTWI and DWI at 18 F-FDG PET/MRI 3.0 T. ASSESSMENT: The apparent diffusion coefficient (ADC), magnetization transfer ratio asymmetry (MTRasym [3.5 ppm]), maximum standardized uptake value (SUVmax ), metabolic tumor volume (MTV), and total lesion glycolysis (TLG) were calculated and compared. STATISTICAL TESTS: Individual sample t-test, Mann-Whitney U test, Logistic regression, and P < 0.05 were considered statistically significant. RESULTS: In NSCLC, MTRasym (3.5 ppm), MTV, and TLG were significantly lower and ADC was significantly higher than in SCLC; MTRasym (3.5 ppm) was significantly higher and SUVmax , MTV, and TLG were significantly lower in EGFR mutation-type NSCLC than in EGFR wild-type NSCLC. In the identification of SCLC and NSCLC, MTRasym (3.5 ppm), ADC, and MTV were independent predictors, the AUCs of the combination of independent predictors, MTV, TLG, MTRasym (3.5 ppm), and ADC were 0.942, 0.875, 0.843, 0.814, and 0.687, respectively. In the identification of EGFR mutation-type and wild-type NSCLC, MTRasym (3.5 ppm) and MTV were independent predictors, the AUCs of the combination of independent predictors, TLG, MTV, MTRasym (3.5 ppm), and SUVmax were 0.919, 0.834, 0.813, 0.795, and 0.771, respectively. DATA CONCLUSION: In the noninvasive assessment of lung cancer subtype and EGFR status, APTWI has similar utility to diffusion and metabolic parameters. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY STAGE: 2.
Subject(s)
Carcinoma, Non-Small-Cell Lung , Lung Neoplasms , Amides , Carcinoma, Non-Small-Cell Lung/diagnostic imaging , Carcinoma, Non-Small-Cell Lung/genetics , ErbB Receptors/genetics , Fluorodeoxyglucose F18 , Humans , Lung Neoplasms/diagnostic imaging , Lung Neoplasms/genetics , Neoplasm Staging , Positron-Emission Tomography , Prognosis , Prospective Studies , Protons , Radiopharmaceuticals , Retrospective Studies , Tumor BurdenABSTRACT
BACKGROUND: Noninvasive identification of the histological features of endometrioid adenocarcinoma is necessary. This study aimed to investigate whether amide proton transfer-weighted imaging (APTWI) and multimodel (monoexponential, biexponential, and stretched exponential) diffusion-weighted imaging (DWI) could predict the histological grade of endometrial adenocarcinoma (EA). In addition, we analyzed the correlation between each parameter and the Ki-67 index. METHODS: A total of 90 EA patients who received pelvic magnetic resonance imaging (MRI) were enrolled. The magnetization transfer ratio asymmetry [MTRasym (3.5 ppm)], apparent diffusion coefficient (ADC), diffusion coefficient (D), pseudo-diffusion coefficient (D*), perfusion fraction (f), distributed diffusion coefficient (DDC), and water molecular diffusion heterogeneity index (α) were measured and compared. Correlation coefficients between each parameter and histological grade and the Ki-67 index were calculated. Statistical methods included the independent samples t test, Spearman's correlation, and logistic regression. RESULTS: MTRasym (3.5 ppm) [(3.72%±0.31%) vs. (3.27%±0.48%)], f [(3.15%±0.36%) vs. (2.69%±0.83%)], and α [(0.89±0.05) vs. (0.81±0.09)] were higher and ADC [(0.82±0.08) vs. (0.89±0.10) ×10-3 mm2/s], D [(0.67±0.09) vs. (0.81±0.11) ×10-3 mm2/s], and DDC [(1.04±0.09) vs. (1.13±0.13) ×10-3 mm2/s] were lower in high-grade EA than in low-grade EA (P<0.05). MTRasym (3.5 ppm) and D were independent predictors for the histological grade of EA. The combination of MTRasym (3.5 ppm) and D were better able to identify high- and low-grade EA than was each parameter. MTRasym (3.5 ppm) and α were moderately and weakly positively correlated, respectively, with histological grade and the Ki-67 index (r=0.528, r=0.514, r=0.395, and r=0.367; P<0.05). D was moderately negatively correlated with histological grade and the Ki-67 index (r=-0.540 and r=-0.529; P<0.05). DDC was weakly and moderately negatively correlated with histological grade and the Ki-67 index, respectively (r=-0.473 and r=-0.515; P<0.05). ADC was weakly negatively correlated with histological grade and the Ki-67 index (r=-0.417 and r=-0.427; P<0.05). f was weakly positively correlated with histological grade and the Ki-67 index (r=0.294 and r=0.355; P<0.05). CONCLUSIONS: Our study found that both multimodel DWI and APTWI could be used to estimate the histological grade and Ki-67 index of EA, and the combination of high MTRasym (3.5 ppm) and low D may be an effective imaging marker for predicting the grade of EA.
ABSTRACT
BACKGROUND: Amide proton transfer weighted imaging (APTw), intravoxel incoherent motion (IVIM), and positron emission tomography (PET) imaging all have the potential to characterize solitary pulmonary lesions (SPLs). PURPOSE: To compare APTw and IVIM with PET imaging for distinguishing between benign and malignant SPLs and their subtypes. STUDY TYPE: Prospective. POPULATION: Ninety-five patients, 78 with malignant SPLs (including 48 with adenocarcinoma [AC] and 17 with squamous cell carcinoma [SCC]), and 17 with benign SPLs. FIELD STRENGTH/SEQUENCE: Fast spin-echo (FSE) T2WI, FSE APTw and echo-planar imaging IVIM, MR-base attenuation correction (MRAC), and PET imaging on a 3-T whole-body PET/MR system. ASSESSMENT: The magnetization transfer ratio asymmetry (MTRasym) at 3.5 ppm, diffusion coefficient (D), pseudo diffusion coefficient (D*), perfusion fraction (f), and the maximum standardized uptake value (SUVmax) were analyzed. STATISTICAL TESTS: Individual sample t-test, Delong test, Pearson's correlation analysis, and area under the receiver operating characteristic curve (AUC). P < 0.05 indicated statistical significance. RESULTS: The MTRasym and SUVmax were significantly higher, and D was significantly lower in the malignant group (3.3 ± 2.6 [%], 7.8 ± 5, and 1.2 ± 0.3 [×10-3 mm2 /second]) compared to the benign group (-0.3 ± 1.6 [%], 3.1 ± 3.8, and 1.6 ± 0.3 [×10-3 mm2 /second]). The MTRasym and D were significantly lower, and SUVmax was significantly higher in the SCC group (0.8 ± 1.0 [%], 1.0 ± 0.2 [×10-3 mm2 /second] than in the AC group (4.1 ± 2.6 [%], 1.3 ± 0.3 [×10-3 mm2 /second], 6.7 ± 4.6). Besides, the combination (AUC = 0.964) of these three methods showed higher diagnostic efficacy than any individual method (AUC = 0.917, 0.851, 0.82, respectively) in identifying malignant and benign SPLs. However, APTw showed better diagnostic efficacy than the combination of three methods or PET imaging alone in distinguishing SCC and AC groups (AUC = 0.934, 0.781, 0.725, respectively). DATA CONCLUSION: APTw, IVIM, and PET imaging are all effective methods to distinguish benign and malignant SPLs and their subtypes. APTw is potentially more capable than PET imaging of distinguishing lung SCC from AC. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY: Stage 2.
Subject(s)
Adenocarcinoma , Carcinoma, Squamous Cell , Amides , Carcinoma, Squamous Cell/diagnostic imaging , Diffusion Magnetic Resonance Imaging/methods , Fluorodeoxyglucose F18 , Humans , Lung , Magnetic Resonance Imaging , Motion , Positron-Emission Tomography , Prospective Studies , ProtonsABSTRACT
BACKGROUND: Ki-67 proliferation index (PI) is important for providing information on tumor behavior, treatment response, and prognosis. Integrated positron emission tomography/magnetic resonance (PET/MR) may have the potential to assess Ki-67 PI in patients with lung adenocarcinoma. PURPOSE: To explore the value of simultaneous 18 F-fluorodeoxyglucose (18 F-FDG) PET/MR-derived parameters in assessing the proliferation status of lung adenocarcinoma and to determine the best combination of parameters. STUDY TYPE: Prospective. POPULATION: Seventy-eight patients with lung adenocarcinoma and with Ki-67 PI. FIELD STRENGTH/SEQUENCE: 3.0 T, simultaneous PET/MRI including diffusion-weighted imaging (DWI) and 18 F-FDG PET. ASSESSMENT: DWI-derived parameters, namely, apparent diffusion coefficient (ADC), true diffusion coefficient (D), pseudo diffusion coefficient (D*), perfusion fraction (f), diffusion heterogeneity index (α), and distributed diffusion coefficient (DDC); and PET-derived parameters, namely, maximum standardized uptake value (SUVmax ), metabolic tumor volume (MTV), and total lesion glycolytic volume (TLG), were calculated and compared between the high (>25%) and low (≤25%) Ki-67 PI groups. The correlations between PET-derived parameters and DWI-derived parameters were analyzed. STATISTICAL TESTS: Student's t-test, Mann-Whitney U test, chi-square test, and receiver operating characteristic (ROC) curves. A P-value <0.05 was considered statistically significant. RESULTS: The SUVmax , MTV, TLG, ADC, D, and DDC values were significantly different between the high (N = 35) and low Ki-67 PI groups (N = 43). D, SUVmax , and MTV independently predicted the Ki-67 PI status. The combination of D, SUVmax , and MTV had the largest area under the ROC curve (AUC = 0.900), which was significantly larger than the AUC alone of DDC (AUC = 0.725), SUVmax (AUC = 0.815), MTV (AUC = 0.774), or TLG (AUC = 0.783). The perfusion fraction did not correlate with SUVmax , MTV, or TLG (r = -0.03, -0.11, and -0.04, respectively; P = 0.786, 0.348, and 0.733). DATA CONCLUSION: The combination of D, SUVmax , and MTV may predict Ki-67 PI status. No correlation was observed between perfusion parameters and metabolic parameters. LEVEL OF EVIDENCE: 1 TECHNICAL EFFICACY: Stage 2.
Subject(s)
Adenocarcinoma of Lung , Fluorodeoxyglucose F18 , Cell Proliferation , Diffusion Magnetic Resonance Imaging/methods , Humans , Ki-67 Antigen , Magnetic Resonance Imaging , Prospective Studies , Retrospective StudiesABSTRACT
BACKGROUND: Endometrial carcinoma (EC) risk stratification is generally based on histological assessment. It would be beneficial to perform risk stratification noninvasively by MRI. PURPOSE: To investigate the application of amide proton transfer-weighted imaging (APTWI), monoexponential, biexponential, and stretched exponential intravoxel incoherent motion (IVIM), and diffusion kurtosis imaging (DKI) for the evaluation of risk stratification in early-stage EC. STUDY TYPE: Prospective. POPULATION: Eighty patients with early-stage EC (47 classified as low risk, 20 as medium risk, and 13 as high risk by histological grade and International Federation of Gynecology and Obstetrics stage). FIELD STRENGTH/SEQUENCE: T1-weighted imaging, T2-weighted imaging, IVIM, APTWI, and DKI MRI at 3 T. ASSESSMENT: The magnetization transfer ratio asymmetry (MTRasym [3.5 ppm]), apparent diffusion coefficient (ADC), diffusion coefficient (D), pseudo diffusion coefficient (D*), perfusion fraction (f), distributed diffusion coefficient (DDC), water molecular diffusion heterogeneity index (α), mean kurtosis (MK), and mean diffusivity (MD) were calculated and compared between low-risk and non-low-risk groups. STATISTICAL TESTS: Individual sample t test, analysis of variance, and logistic regression. A P-value <0.05 was considered statistically significant. RESULTS: The α, ADC, D, DDC, and MD were significantly higher and the f, MK, and MTRasym (3.5 ppm) were significantly lower in the low-risk group than in the non-low-risk group. The difference in D* between the two groups was not significant (P = 0.289). MTRasym (3.5 ppm), D, and MK were independent predictors of risk stratification. The combination of these three parameters was better able to identify low- and non-low-risk groups than each individual parameter. DATA CONCLUSION: The IVIM, DKI, and APTWI parameters have potential as imaging markers for risk stratification in early-stage EC. LEVEL OF EVIDENCE: 2 TECHNICAL EFFICACY: Stage 3.