Decentralized Distributed Multi-institutional PET Image Segmentation Using a Federated Deep Learning Framework

Abstract

PurposeThe generalizability and trustworthiness of deep learning (DL)–based algorithms depend on the size and heterogeneity of training datasets. However, because of patient privacy concerns and ethical and legal issues, sharing medical images between different centers is restricted. Our objective is to build a federated DL-based framework for PET image segmentation utilizing a multicentric dataset and to compare its performance with the centralized DL approach.MethodsPET images from 405 head and neck cancer patients from 9 different centers formed the basis of this study. All tumors were segmented manually. PET images converted to SUV maps were resampled to isotropic voxels (3 × 3 × 3 mm3) and then normalized. PET image subvolumes (12 × 12 × 12 cm3) consisting of whole tumors and background were analyzed. Data from each center were divided into train/validation (80% of patients) and test sets (20% of patients). The modified R2U-Net was used as core DL model. A parallel federated DL model was developed and compared with the centralized approach where the data sets are pooled to one server. Segmentation metrics, including Dice similarity and Jaccard coefficients, percent relative errors (RE%) of SUVpeak, SUVmean, SUVmedian, SUVmax, metabolic tumor volume, and total lesion glycolysis were computed and compared with manual delineations.ResultsThe performance of the centralized versus federated DL methods was nearly identical for segmentation metrics: Dice (0.84 ± 0.06 vs 0.84 ± 0.05) and Jaccard (0.73 ± 0.08 vs 0.73 ± 0.07). For quantitative PET parameters, we obtained comparable RE% for SUVmean (6.43% ± 4.72% vs 6.61% ± 5.42%), metabolic tumor volume (12.2% ± 16.2% vs 12.1% ± 15.89%), and total lesion glycolysis (6.93% ± 9.6% vs 7.07% ± 9.85%) and negligible RE% for SUVmax and SUVpeak. No significant differences in performance (P > 0.05) between the 2 frameworks (centralized vs federated) were observed.ConclusionThe developed federated DL model achieved comparable quantitative performance with respect to the centralized DL model. Federated DL models could provide robust and generalizable segmentation, while addressing patient privacy and legal and ethical issues in clinical data sharing.

Decentralized collaborative multi-institutional PET attenuation and scatter correction using federated deep learning

“…Building a generalizable and robust model requires a large dataset, while privacy concerns could be addressed by FL approaches without sacrificing models’ performance. In a more recent study, Shiri et al [ 76 ] proposed a FL based multi-institutional PET image segmentation framework on head and neck studies. They enrolled 404 patients from eight different centers and reported that FL-based algorithms outperformed CB and achieved similar performance as the CZ approach.…”

“…This study inherently bears a number of limitations. The implementation of all models was performed on a server using different GPUs where the different nodes were considered as centers similar to previous FL studies [ 35 , 36 , 54 – 59 , 74 , 76 ]. The challenges of FL, such as local computer capacity, and communication bottleneck between centers and local server should be considered in the real clinical scenario.…”

Myocardial Perfusion SPECT Imaging Radiomic Features and Machine Learning Algorithms for Cardiac Contractile Pattern Recognition

“…In addition, our dataset was obtained from one institute which undermines the robustness of the findings. Further studies might gather larger and diverse dataset obtained from multiple institutes with different image acquisition parameters and patients’ ethnicity to improve the reproducibility of the models [ 51 ]. However, for proof-of-concept, this study contained enough patients.…”

Deep Learning-based Non-rigid Image Registration for High-dose Rate Brachytherapy in Inter-fraction Cervical Cancer

“…Future studies will aim to further improve the proposed method by adding more Conv layers and testing different loss functions. Although a dataset of 57 cervical patients was used in this study, increasing the training dataset would result in a model with improved accuracy and generalizability by using data augmentation [ 57 ] and decentralize federated learning [ 58 ] approaches. Another suggestion would be the use of more structure contours to take into account more anatomical considerations.…”