Swarm learning for decentralized artificial intelligence in cancer histopathology

Abstract

Artificial intelligence (AI) can predict the presence of molecular alterations directly from routine histopathology slides. However, training robust AI systems requires large datasets for which data collection faces practical, ethical and legal obstacles. These obstacles could be overcome with swarm learning (SL), in which partners jointly train AI models while avoiding data transfer and monopolistic data governance. Here, we demonstrate the successful use of SL in large, multicentric datasets of gigapixel histopathology images from over 5,000 patients. We show that AI models trained using SL can predict BRAF mutational status and microsatellite instability directly from hematoxylin and eosin (H&E)-stained pathology slides of colorectal cancer. We trained AI models on three patient cohorts from Northern Ireland, Germany and the United States, and validated the prediction performance in two independent datasets from the United Kingdom. Our data show that SL-trained AI models outperform most locally trained models, and perform on par with models that are trained on the merged datasets. In addition, we show that SL-based AI models are data efficient. In the future, SL can be used to train distributed AI models for any histopathology image analysis task, eliminating the need for data transfer.

Self-supervised deep learning for pan-cancer mutation prediction from histopathology

“…A key limitation is that many clinically relevant genes were not analyzable due to having fewer than 25 mutants in TCGA. Large-scale efforts are needed to create datasets with a sufficient size, which could be facilitated by federated 22 or swarm 23 learning. Since the early 2000s, studies have shown a link between genetic alterations and histological phenotypes 24 , which DL can exploit.…”

“…A key limitation is that many clinically relevant genes were not analyzable due to having fewer than 25 mutants in TCGA. Large-scale efforts are needed to create datasets with a sufficient size, which could be facilitated by federated 22 or swarm 23 learning.…”

When blockchain meets artificial intelligence: An application to cancer histopathology

“… 9 When training multicentric models on medical data, SL has its advantages over FL, because it centralizes neither the data nor the models ( Figure 1 ). Prior to the recent study by Saldanha et al., 10 there was no application of SL to cancer histopathology data, making the study a pioneer in the field of computational histopathology.

Figure 1 Different model training strategies Conventional merged data training involves data sharing and aggregation (left).

…”

Direct prediction of genetic aberrations from pathology images in gastric cancer with swarm learning

“…In the last five years, decentralized machine learning approaches have been proposed which could alleviate the need for physical data exchange. The most prominent examples include federated learning (FL) and swarm learning (SL) [15][16][17]. In these approaches, multiple datasets are located on physically separate computers, with the DL model trained on each computer separately [16].…”