Parameter Estimation for the Intravoxel Incoherent Motion-diffusional Kurtosis Imaging Model Using Synthetic Q-space Learning Toward Breast Tumor Diagnosis
Abstract
PURPOSE: Synthetic Q-space learning (synQSL), which uses only synthetic data in the training of regressors, has demonstrated promising results in parameter estimation for brain diffusion MRI (dMRI). This study aimed to evaluate the performance of synQSL in breast dMRI for intravoxel incoherent motion-diffusional kurtosis imaging (IVIM-DKI) parameter estimation, with comparing several types of regressors and conventional fitting by nonlinear least squares fitting (NL-LSF). METHODS: samples, and trained multi-layer perceptron (MLP), Kolmogorov-Arnold networks (KAN), and random forest (RF) regressors for synQSL. The performance of the parameter estimation methods including NL-LSF was evaluated using a digital phantom including various value combinations of IVIM-DKI parameters and clinical data of 67 cases (13 benign and 54 malignant lesions) through quantitative analysis and visual assessment. RESULTS: , and D than NL-LSF (P < 0.05) with less variation among the parameter sets. In clinical datasets, synQSL not only improved the visual quality of parameter maps but also showed significant differences between benign and malignant lesions in parameters while the NL-LSF failed. Furthermore, the estimation times for all synQSL regressors were substantially shorter than that of NL-LSF. Among the synQSL regressors, MLP showed superior properties including computational cost. CONCLUSION: synQSL demonstrated superior parameter estimation performance compared to NL-LSF in breast IVIM-DKI analysis. In this study, MLP was considered the most suitable regressor for synQSL among those we examined, based on the balance between estimation accuracy and computational costs.
Related
- Clinical utility and reproducibility of intravoxel incoherent motion MRI in rectal cancer: a structured narrative review
- Influence of self-reported vascular health and physical activity level on intravoxel incoherent motion imaging response to exercise in chronic low back pain
- Varying diffusion curvature, intravoxel incoherent motion, and conventional monoexponential diffusion models for evaluating Ki-67 expression in rectal carcinoma
- Bayesian Parameter Estimation of Cannibalism Dynamics in Larval Atractosteus tropicus Under Different Feeding Conditions
- In Vivo Cardiac Biomechanical Model Parameter Estimation from Ultrafast Shear Wave Elastography
BibTeX
@article{konya2026parameter,
title = {{Parameter Estimation for the Intravoxel Incoherent Motion-diffusional Kurtosis Imaging Model Using Synthetic Q-space Learning Toward Breast Tumor Diagnosis}},
author = {Kousei Konya and Yuki Ichinoseki and Erina Kato and Naoko Mori and Hideki Ota and Shunji Mugikura and Kei Takase and Yoshitaka Masutani},
journal = {Magnetic Resonance in Medical Sciences},
year = {2026},
month = aug,
doi = {10.2463/mrms.mp.2026-0045},
url = {https://www.jstage.jst.go.jp/article/mrms/25/4/25_mp.2026-0045/_article},
note = {CC BY},
}
RIS
TY - JOUR
TI - Parameter Estimation for the Intravoxel Incoherent Motion-diffusional Kurtosis Imaging Model Using Synthetic Q-space Learning Toward Breast Tumor Diagnosis
AU - Kousei Konya
AU - Yuki Ichinoseki
AU - Erina Kato
AU - Naoko Mori
AU - Hideki Ota
AU - Shunji Mugikura
AU - Kei Takase
AU - Yoshitaka Masutani
PY - 2026
DA - 2026/08/19
JO - Magnetic Resonance in Medical Sciences
DO - 10.2463/mrms.mp.2026-0045
UR - https://www.jstage.jst.go.jp/article/mrms/25/4/25_mp.2026-0045/_article
AB - PURPOSE: Synthetic Q-space learning (synQSL), which uses only synthetic data in the training of regressors, has demonstrated promising results in parameter estimation for brain diffusion MRI (dMRI). This study aimed to evaluate the performance of synQSL in breast dMRI for intravoxel incoherent motion-diffusional kurtosis imaging (IVIM-DKI) parameter estimation, with comparing several types of regressors and conventional fitting by nonlinear least squares fitting (NL-LSF). METHODS: samples, and trained multi-layer perceptron (MLP), Kolmogorov-Arnold networks (KAN), and random forest (RF) regressors for synQSL. The performance of the parameter estimation methods including NL-LSF was evaluated using a digital phantom including various value combinations of IVIM-DKI parameters and clinical data of 67 cases (13 benign and 54 malignant lesions) through quantitative analysis and visual assessment. RESULTS: , and D than NL-LSF (P < 0.05) with less variation among the parameter sets. In clinical datasets, synQSL not only improved the visual quality of parameter maps but also showed significant differences between benign and malignant lesions in parameters while the NL-LSF failed. Furthermore, the estimation times for all synQSL regressors were substantially shorter than that of NL-LSF. Among the synQSL regressors, MLP showed superior properties including computational cost. CONCLUSION: synQSL demonstrated superior parameter estimation performance compared to NL-LSF in breast IVIM-DKI analysis. In this study, MLP was considered the most suitable regressor for synQSL among those we examined, based on the balance between estimation accuracy and computational costs.
SN - 1880-2206
SN - 1347-3182
ER -