Restriction Spectrum Imaging and Diffusion Kurtosis Imaging for Assessing Proliferation Status in Rectal Carcinoma
Yingying Cui,Xinhui Wang,Ying Wang,Nan Meng,Yaping Wu,Yu Shen,Neil Roberts,Yan Bai,Xiaosheng Song,Guofeng Shen,Yongjun Guo,Jinxia Guo,Meiyun Wang
DOI: https://doi.org/10.1016/j.acra.2024.08.021
2024-08-26
Abstract:Objectives: To investigate the application of the three-compartment restriction spectrum imaging (RSI) model, diffusion kurtosis imaging (DKI), and diffusion-weighted imaging (DWI) in predicting Ki-67 status in rectal carcinoma. Methods: A total of 80 rectal carcinoma patients, including 47 high-proliferation (Ki-67 > 50%) cases and 33 low-proliferation (Ki-67 ≤ 50%) cases, underwent pelvic MRI were enrolled. Parameters derived from RSI (f1, f2, and f3), DKI (MD and MK), and DWI (ADC) were calculated and compared between the two groups. Logistic regression (LR) analysis was conducted to identify independent predictors and assess combined diagnosis. Area under the receiver operating characteristic curve (AUC), DeLong analysis, and calibration curve analyses were performed to evaluate diagnostic performance. Results: The patients with high-proliferation rectal carcinoma exhibited significantly higher f1 and MK values and significantly lower ADC, MD, f2, and f3 values than those with low-proliferation rectal carcinoma (P < 0.05). LR analysis showed that MD, MK, and f2 were independent predictors for Ki-67 status in rectal carcinoma. Moreover, the combination of these three parameters achieved an optimal diagnostic efficacy (AUC = 0.877, sensitivity = 80.85%, specificity = 84.85%) that was significantly better than that obtained using ADC (AUC = 0.783, Z = 2.347, P = 0.019), f2 (AUC = 0.732, Z = 2.762, P = 0.006), and f3 (AUC = 0.700, Z = 3.071, P = 0.002). The combined diagnosis also showed good performance (AUC = 0.859) in the internal validation analysis based on 1000 bootstrap samples, while the calibration curve demonstrated that the combined diagnosis provided good stability. Conclusion: RSI, DKI, and DWI can effectively differentiate between patients with high- and low-proliferation rectal carcinoma. Furthermore, the MD, MK, and f2 imaging parameters may be a novel and promising combination biomarker for examining Ki-67 status in rectal carcinoma.