Objective To investigate the application of histogram features of apparent diffusion coefficient (ADC), mean diffusion coefficient (MD), mean kurtosis (MK), and fat fraction (FF) for assessing multiple myeloma (MM) activity. Materials and methods From April 2023 to December 2023, 24 multiple myeloma patients who underwent zonally oblique multi-slice diffusion-weighted imaging (ZOOM-DWI), zonally oblique multi-slice diffusion kurtosis imaging (ZOOM-DKI), and modified Dixon quantification imaging (mDixon-Quant) were enrolled in this retrospective study. Histogram features of ADC, MD, MK, and FF were analyzed. Significant variables were selected for multivariate logistic regression analysis using a backward stepwise selection method. The diagnostic efficacies of individual histogram features and combined models were evaluated by the receiver operating characteristic curve (ROC) and the area under the curve (AUC). Results A total of 24 participants were enrolled in this study, 12 males and 12 females, with an age range of 39–73 (57.88 ± 8.55) years. The ADC and MD histogram features did not correlate with the disease status in multiple myeloma, whereas the MK and FF histogram features were significantly associated with the disease status (p < 0.05). MK_InterquartileRange and FF_Median had the highest AUC values in multiple myeloma activity assessment. The differences in diagnostic efficacies between MK_InterquartileRange and FF_Median, MK_InterquartileRange and the MK model, FF_Median and the FF model, the MK model and the FF model were not statistically different (p = 0.70, 0.54, 0.09 and 0.09, respectively). Conclusion Histogram features of MK and FF are valuable in assessing disease activity in multiple myeloma patients. Key Points <Emphasis Type="BoldItalic">Question</Emphasis> Accurate assessment of the disease status of multiple myeloma is crucial for enhancing individualized treatment, yet current non-invasive tools remain inadequate. <Emphasis Type="BoldItalic">Findings</Emphasis> Mean kurtosis (MK) and fat fraction (FF) histogram features are associated with disease activity of multiple myeloma, providing rich and accurate parameters for assessing the disease status. <Emphasis Type="BoldItalic">Clinical relevance</Emphasis> The MK and FF histogram features facilitate non-invasive assessment of multiple myeloma activity, potentially guiding the development of personalized treatment strategies. Graphical Abstract

The value of multiparametric functional MRI histogram features in assessing multiple myeloma activity

期刊论文
发表日期：

2025年3月18日

查看全文

Ying Lin,
Hongjie Huang,
Zebin Xiao,
Shufang Shi,
Qiang Weng,
Zhanhai Tu

摘要

Objective

To investigate the application of histogram features of apparent diffusion coefficient (ADC), mean diffusion coefficient (MD), mean kurtosis (MK), and fat fraction (FF) for assessing multiple myeloma (MM) activity.

Materials and methods

From April 2023 to December 2023, 24 multiple myeloma patients who underwent zonally oblique multi-slice diffusion-weighted imaging (ZOOM-DWI), zonally oblique multi-slice diffusion kurtosis imaging (ZOOM-DKI), and modified Dixon quantification imaging (mDixon-Quant) were enrolled in this retrospective study. Histogram features of ADC, MD, MK, and FF were analyzed. Significant variables were selected for multivariate logistic regression analysis using a backward stepwise selection method. The diagnostic efficacies of individual histogram features and combined models were evaluated by the receiver operating characteristic curve (ROC) and the area under the curve (AUC).

Results

A total of 24 participants were enrolled in this study, 12 males and 12 females, with an age range of 39–73 (57.88 ± 8.55) years. The ADC and MD histogram features did not correlate with the disease status in multiple myeloma, whereas the MK and FF histogram features were significantly associated with the disease status (p < 0.05). MK_InterquartileRange and FF_Median had the highest AUC values in multiple myeloma activity assessment. The differences in diagnostic efficacies between MK_InterquartileRange and FF_Median, MK_InterquartileRange and the MK model, FF_Median and the FF model, the MK model and the FF model were not statistically different (p = 0.70, 0.54, 0.09 and 0.09, respectively).

Conclusion

Histogram features of MK and FF are valuable in assessing disease activity in multiple myeloma patients.

Key Points

Question Accurate assessment of the disease status of multiple myeloma is crucial for enhancing individualized treatment, yet current non-invasive tools remain inadequate.

Findings Mean kurtosis (MK) and fat fraction (FF) histogram features are associated with disease activity of multiple myeloma, providing rich and accurate parameters for assessing the disease status.

Clinical relevance The MK and FF histogram features facilitate non-invasive assessment of multiple myeloma activity, potentially guiding the development of personalized treatment strategies.

Graphical Abstract