Joint Lymphoma Lesion Segmentation and Prognosis Prediction From Baseline FDG-PET Images via Multitask Convolutional Neural Networks

Objective:
Lymphoma lesion segmentation and prognosis prediction from baseline FDG-PET images are valuable for tailoring and adapting a treatment plan for patients with Diffuse Large B-cell Lymphoma (DLBCL). However, the tasks are challenging due to the fact that DLBCL is a highly heterogeneous group of neoplasms and that the lymphoma cells are large and arranged in a diffuse pattern.
Methods:
We propose a novel multi-task 3D convolutional neural network model for simultaneous lymphoma lesion segmentation and prognosis prediction from baseline FDG-PET images. In our model, the learned image features of one task are shared and thereby mutually reinforce the learning of the other task. Since the dataset is limited, to reduce overfitting and to facilitate network convergence, we further introduce deep supervision into both the segmentation task and the prognosis prediction task.
Results:
Evaluated on a dataset of 269 patients, our method achieves an average Dice similarity coefficient of 0.868 for lesion segmentation, an average AUC (area under the curve) of 0.823 and an average accuracy of 0.821 for prognosis prediction. Its predictions can differentiate patients with different PFS (progression-free survival) and OS (overall survival) (
${p} < 0.0001$
).
Conclusion:
Our method achieves joint lymphoma lesion segmentation and prognosis prediction from baseline FDG-PET scans.
Significance:
Our model may be used to asset the physician as a second opinion while making the final decision.