Multimodal deep learning model on interim [18F]FDG PET/CT for predicting primary treatment failure in diffuse large B-cell lymphoma

Objectives The prediction of primary treatment failure (PTF) is necessary for patients with diffuse large B-cell lymphoma (DLBCL) since it serves as a prominent means for improving front-line outcomes. Using interim F-18-fluoro-2-deoxyglucose ([F-18]FDG) positron emission tomography/computed tomography (PET/CT) imaging data, we aimed to construct multimodal deep learning (MDL) models to predict possible PTF in low-risk DLBCL. Methods Initially, 205 DLBCL patients undergoing interim [F-18]FDG PET/CT scans and the front-line standard of care were included in the primary dataset for model development. Then, 44 other patients were included in the external dataset for generalization evaluation. Based on the powerful backbone of the Conv-LSTM network, we incorporated five different multimodal fusion strategies (pixel intermixing, separate channel, separate branch, quantitative weighting, and hybrid learning) to make full use of PET/CT features and built five corresponding MDL models. Moreover, we found the best model, that is, the hybrid learning model, and optimized it by integrating the contrastive training objective to further improve its prediction performance. Results The final model with contrastive objective optimization, named the contrastive hybrid learning model, performed best, with an accuracy of 91.22% and an area under the receiver operating characteristic curve (AUC) of 0.926, in the primary dataset. In the external dataset, its accuracy and AUC remained at 88.64% and 0.925, respectively, indicating its good generalization ability. Conclusions The proposed model achieved good performance, validated the predictive value of interim PET/CT, and holds promise for directing individualized clinical treatment.

Automated summary

In this study, a multimodal deep learning (MDL) model was constructed using interim PET/CT imaging data to predict possible primary treatment failure (PTF) in low-risk DLBCL patients. The best model, optimized with a contrastive training objective, showed good prediction performance and generalization ability in both the primary and external dataset.