Noninvasive monitoring of allograft rejection in a rat lung transplant model: Application of machine learning-based F-fluorodeoxyglucose positron emission tomography radiomics.

BACKGROUND

Standardized uptake values (SUVs) derived from F-fluorodeoxyglucose (F-FDG) positron emission tomography (PET) are valuable but insufficient for detecting lung allograft rejection (AR). Using a rat lung transplantation (LTx) model, we investigated correlations of AR with the SUV and PET-derived radiomics and further evaluated the performance of machine learning (ML)-based radiomics for monitoring AR.

METHODS

LTx was performed on 4 groups of rats: isograft, allograft-cyclosporine (CsA), allograft-CsA, and allograft-CsA. Each rat underwent F-FDG PET at week 3 or 6. The SUV and radiomic features were extracted from the PET images. Least absolute shrinkage and selection operator regression was used to construct a radiomics score (Rad-score). Ten modeling algorithms with 7 feature selection methods were performed to develop 70 radiomics models (49 ML models and 21 logistic regression models) for monitoring AR, validated using the bootstrap method.

RESULTS

In total, 837 radiomic features were extracted from each PET image. The SUV and Rad-score showed significant positive correlations with histopathology (p < .05). The area under the curve (AUC) of SUV for detecting AR was 0.783. The median AUC of ML models was 0.921, which was superior to that of logistic regression models (median AUC, 0.721). The optimal ML model using a random forest modeling algorithm with random forest feature selection method exhibited the highest AUC of 0.982 (95% confidence interval, 0.875-1.000) in all models.

CONCLUSIONS

SUV provided a good correlation with AR, but ML-based PET radiomics further strengthened the power of F-FDG PET functional imaging for monitoring AR in LTx.