AUTHOR=Huang Yecai , Zhu Yuxin , Yang Qiang , Luo Yangkun , Zhang Peng , Yang Xuegang , Ren Jing , Ren Yazhou , Lang Jinyi , Xu Guohui 

TITLE=Automatic tumor segmentation and metachronous single-organ metastasis prediction of nasopharyngeal carcinoma patients based on multi-sequence magnetic resonance imaging

JOURNAL=Frontiers in Oncology

VOLUME=Volume 13 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2023.953893

DOI=10.3389/fonc.2023.953893

ISSN=2234-943X

ABSTRACT=Background: Distant metastases is the main failure mode of nasopharyngeal carcinoma. However, early prediction of distant metastases in NPC is extremely challenging. Deep learning has made great progress in recent years. This study intends to explore and construct the metachronous single organ metastases (MSOM) based on multimodal magnetic resonance imaging. Patients and methods: The magnetic resonance imaging data of 186 patients with nasopharyngeal carcinoma before treatment was collected, and the Gross tumor volume (GTV) and metastatic lymph nodes (GTVln) prior to treatment was defined on T1WI, T2WI and CE-T1WI. After image normalization, the deep learning platform Python was used in Ubuntu 20.04.1 LTS to construct automatic tumor detection and MSOO prediction model. Results: 85 of 186 patients had MSOM (including 32 liver metastases, 25 lung metastases, and 28 bone metastases). The median time to MSOM was 13 months after treatment (7-36 months). Patients were randomly assigned to training set (N=140) , validation set (N=46).  By comparison, we found that the overall performance of automatic tumor detection model based on CE-T1WI was the best (mAP@0.5=57.6). The performance of automatic detection for primary tumor(GTV) and lymph node gross tumor volume(GTVln) based on CE-T1WI model are better than that of models based on T1WI and T2WI(AP@0.5 is 59.6 and 55.6). The prediction model based on CE-T1WI for MSOM prediction has the best overall performance, and it obtained the largest AUC value (AUC=0.733) in the validation set. The precision, recall, precision and AUC of the prediction model based on CE-T1WI are 0.727, 0.533, 0.730 and 0.733(95%CI 0.557-0.909), respectively. When clinical data were added to the deep learning prediction model, a better performance of the model could be obtained, the AUC of the integrated model based on T2WI,T1WI and CE-T1WI were 0.719,0.727 and 0.775 respectively. By comparing the 3-year survival of high-risk and low-risk patients based on the fusion model, we found that the 3-year DMFS of low and high Conclusion: The intelligent prediction model based on magnetic resonance imaging alone or combined with clinical data has excellent performance in automatic tumor detection and MSOM prediction for NPC patients, and is worthy of clinical application.