Using apparent diffusion coefficient maps and radiomics to predict pathological grade in upper urinary tract...

Area of Science:

• Radiology
• Oncology
• Medical Imaging

Background:

• Apparent diffusion coefficient (ADC) is a quantitative biomarker for upper urinary tract urothelial carcinoma (UTUC) aggressiveness.
• Current methods primarily use mean ADC values, limiting detailed assessment.
• There is a need for advanced imaging techniques to better differentiate UTUC grades.

Purpose of the Study:

• To develop and evaluate an ADC-based radiomics model for differentiating UTUC grades.
• To incorporate texture features from ADC maps into the radiomics model.
• To compare the performance of the radiomics model against mean ADC values.

Main Methods:

• Retrospective analysis of 215 UTUC patients, divided into training and test sets.
• Extraction and screening of radiomics features from ADC maps.
• Construction and validation of an ADC-based radiomics model using ROC curve analysis.

Main Results:

• ADC values were significantly lower in high-grade UTUC compared to low-grade UTUC.
• The mean ADC values showed moderate diagnostic performance (AUC: 0.698 training, 0.628 test).
• The ADC-based radiomics model demonstrated superior performance, achieving an AUC of 1.000 in the training set and 0.786 in the test set.

Conclusions:

• The ADC-based radiomics model shows significant promise for predicting UTUC pathological grade.
• This radiomics approach offers improved classification accuracy over mean ADC values.
• Further validation with larger cohorts is recommended to confirm clinical utility.