A Multi-Modal Deep Learning Approach for Predicting Eligibility for Adaptive Radiation Therapy in Nasopharyngeal Carcinoma Patients

TL;DR

This paper introduces a deep learning model that predicts which nasopharyngeal cancer patients will need adaptive radiation therapy before treatment starts, aiming to improve efficiency and personalized care.

Contribution

A novel multi-modal deep learning model that fuses imaging and clinical data to predict ART eligibility before treatment initiation.

Findings

The model achieved an AUC of 0.9070, outperforming other deep learning networks in predicting ART eligibility.

The approach combines CT, MRI, and clinical data using ResNet-50 and cross-attention mechanisms for accurate classification.

Abstract

Nasopharyngeal carcinoma (NPC) is a type of cancer that often requires radiation therapy as a primary treatment. During therapy, some patients may experience anatomical changes that make adaptive radiation therapy (ART) necessary to improve treatment outcomes. However, identifying which patients will need ART is usually carried out only after treatment has started, which can be time consuming and resource intensive. In this study, we developed a deep learning model that combines medical imaging and clinical data to predict ART eligibility before treatment begins. This early prediction has the potential to help clinicians to better plan therapy in advance, reduce unnecessary delays, and make more efficient use of medical resources. By identifying suitable ART candidates ahead of time, our approach has the potential to improve personalized cancer care and support faster clinical decision-making.

Background: Adaptive radiation therapy (ART) can improve prognosis for nasopharyngeal carcinoma (NPC) patients. However, the inter-individual variability in anatomical changes, along with the resulting extension of treatment duration and increased workload for the radiologists, makes the selection of eligible patients a persistent challenge in clinical practice. The purpose of this study was to predict eligible ART candidates prior to radiation therapy (RT) for NPC patients using a classification neural network. By leveraging the fusion of medical imaging and clinical data, this method aimed to save time and resources in clinical workflows and improve treatment efficiency. Methods: We collected retrospective data from 305 NPC patients who received RT at Hong Kong Queen Elizabeth Hospital. Each patient sample included pre-treatment computed tomographic (CT) images, T1-weighted magnetic resonance imaging (MRI) data, and T2-weighted MRI images, along with clinical data. We developed and trained a novel multi-modal classification neural network that combines ResNet-50, cross-attention, multi-scale features, and clinical data for multi-modal fusion. The patients were categorized into two labels based on their re-plan status: patients who received ART during RT treatment, as determined by the radiation oncologist, and those who did not. Results: The experimental results demonstrated that the proposed multi-modal deep prediction model outperformed other commonly used deep learning networks, achieving an area under the curve (AUC) of 0.9070. These results indicated the ability of the model to accurately classify and predict ART eligibility for NPC patients. Conclusions: The proposed method showed good performance in predicting ART eligibility among NPC patients, highlighting its potential to enhance clinical decision-making, optimize treatment efficiency, and support more personalized cancer care.

Figures