Related Experiment Video
Updated: Apr 18, 2026
Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging
Published on: December 15, 2014
Application of Full Tumor ADC Histogram Parameters Combined with Fractal Dimension in Evaluating Perineural and
Haisheng Wang1, Rong Gao1, Long Yuan1
1Radiology of Department, Lanzhou University Second Hospital, Cuiyingmen No.82, Chengguan District, Lanzhou 730030, People's Republic of China (H.W., R.G., L.Y., G.L., Y.L., K.Z., P.L., H.L.); Second Clinical School, Lanzhou University, Lanzhou, People's Republic of China (H.W., R.G., L.Y., G.L., Y.L., K.Z., P.L., H.L.); Key Laboratory of Medical Imaging of Gansu Province, Lanzhou, People's Republic of China (H.W., R.G., L.Y., G.L., Y.L., K.Z., P.L., H.L.); Gansu International Scientific and Technological Cooperation Base of Medical Imaging Artificial Intelligence, Lanzhou, People's Republic of China (H.W., R.G., L.Y., G.L., Y.L., K.Z., P.L., H.L.).
Rationale And Objectives:
To explore the value of combining full tumor apparent diffusion coefficient (ADC) histogram parameters with fractal dimension (FD) in predicting perineural and lymphovascular invasion (PNI/LVI) and postoperative recurrence risk in rectal adenocarcinoma.
Materials And Methods:
A multicenter retrospective analysis was conducted on the clinical and magnetic resonance imaging (MRI) data of 144 pathologically confirmed rectal adenocarcinoma patients (112 for training, 32 for external validation). Full tumor ADC histogram parameters (e.g., mean, standard deviation, percentiles) and FD were extracted from 3.0 T MRI images. Univariate and multivariate analyses identified independent predictors for PNI/LVI and recurrence. A binary logistic regression model (combined model 1) and a Cox regression model (combined model 2) were constructed. Model performance was evaluated using the area under the receiver operating characteristic curve area under the curve (AUC), concordance index (C-index), and validated internally (5-fold cross‑validation) and externally.
Results:
The FD value was significantly higher in the PNI/LVI‑positive group than in the negative group (1.12 vs. 1.05, p < 0.001). ADC histogram parameters (e.g., ADC_90per, standard deviations (StDev) and mrEMVI also showed significant differences (p < 0.001). The combined model for predicting PNI/LVI (ADC_90per + FD + mrEMVI) achieved an AUC of 0.927 (95% CI: 0.877-0.977) in the training set, superior to individual parameters, the ADC histogram model, or the clinical model (all p < 0.05). External validation yielded an AUC of 0.922. For recurrence prediction, multivariate Cox analysis identified FD (HR = 1.60, p = 0.003) and lymph node metastasis (HR = 2.54, p = 0.013) as independent predictors. The Cox model had a C-index of 0.727 (0.644-0.811) internally and 0.808 externally.
Conclusion:
The combination of full tumor ADC histogram parameters and FD can non-invasively predict the PNI/LVI status and recurrence risk in rectal adenocarcinoma, providing imaging-based evidence for personalized treatment decisions.
More Related Videos
06:05Author Spotlight: Multiplex Immunofluorescence Combined with Spatial Image Analysis for the Clinical and Biological Assessment of the Tumor Microenvironment
Published on: June 2, 2023
05:24Multifractal Spectrum Analysis for Assessing Pulmonary Nodule Malignancy
Published on: January 10, 2025