Deep learning for the detection of microsatellite instability from histology images in colorectal cancer: A systematic literature review

Echle, Amelie; Laleh, Narmin Ghaffari; Schrammen, Peter L.; West, Nicholas P.; Brinker, Titus J.; Gruber, Stephen B.; Buelow, Roman D.; Grabsch, Heike; Quirke, Philip; Kather, Jakob Nikolas

doi:10.1016/j.immuno.2021.100008

Cited by 30 publications

(15 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…AUROC was selected as the primary metric to evaluate algorithm performance and potential clinical utility. AUROC is the most widely used evaluation criterion for binary classification tasks in computational pathology and was chosen to enable a comparison with the findings of previous studies 54 . The AUROCs of five training runs (technical replicates with different random seeds) of a given model were compared.…”

Section: Methodsmentioning

confidence: 99%

Swarm learning for decentralized artificial intelligence in cancer histopathology

Saldanha

Quirke

West

et al. 2022

Nat Med

Self Cite

110

View full text Add to dashboard Cite

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.

show abstract

Section: Methodsmentioning

confidence: 99%

Swarm learning for decentralized artificial intelligence in cancer histopathology

Saldanha

Quirke

West

et al. 2022

Nat Med

Self Cite

110

View full text Add to dashboard Cite

show abstract

“…Even before the DL era, pathological predictors of MSI in colorectal cancer were known [12]. Consecutively, it was demonstrated that DL can predict MSI status from histology as well [28,36,57,60,61,64,66,67,69,83,112]. When a DL system is trained on thousands of patients, the predictive power is superb and can reach AUROCs of above 0.95 [61,62], and a recent head-to-head comparison showed that DL outperforms pathologists [66].…”

Section: Defective Dna Repair Mechanisms: Msi and Hrdmentioning

confidence: 99%

Artificial intelligence to identify genetic alterations in conventional histopathology

Cifci

Foersch

Kather

2022

The Journal of Pathology

View full text Add to dashboard Cite

Precision oncology relies on the identification of targetable molecular alterations in tumor tissues. In many tumor types, a limited set of molecular tests is currently part of standard diagnostic workflows. However, universal testing for all targetable alterations, especially rare ones, is limited by the cost and availability of molecular assays. From 2017 to 2021, multiple studies have shown that artificial intelligence (AI) methods can predict the probability of specific genetic alterations directly from conventional hematoxylin and eosin (H&E) tissue slides. Although these methods are currently less accurate than gold standard testing (e.g. immunohistochemistry, polymerase chain reaction or next-generation sequencing), they could be used as pre-screening tools to reduce the workload of genetic analyses. In this systematic literature review, we summarize the state of the art in predicting molecular alterations from H&E using AI. We found that AI methods perform reasonably well across multiple tumor types, although few algorithms have been broadly validated. In addition, we found that genetic alterations in FGFR, IDH, PIK3CA, BRAF, TP53, and DNA repair pathways are predictable from H&E in multiple tumor types, while many other genetic alterations have rarely been investigated or were only poorly predictable. Finally, we discuss the next steps for the implementation of AI-based surrogate tests in diagnostic workflows.

show abstract

“… 14 , 15 , 16 , 17 There are growing evidences supporting the possible use of deep learning (DL) for H&E stained image-based MMR status detection in CRC, with an area-under-ROC curves (AUROC) between 0·77 and 0·96. 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 Thus, DL is a promising technology that could be further improve the increase detection accuracy.…”

Section: Introductionmentioning

confidence: 99%

Clinical actionability of triaging DNA mismatch repair deficient colorectal cancer from biopsy samples using deep learning

Mei

et al. 2022

eBioMedicine

View full text Add to dashboard Cite

Deep learning for the detection of microsatellite instability from histology images in colorectal cancer: A systematic literature review

Cited by 30 publications

References 50 publications

Swarm learning for decentralized artificial intelligence in cancer histopathology

Swarm learning for decentralized artificial intelligence in cancer histopathology

Artificial intelligence to identify genetic alterations in conventional histopathology

Clinical actionability of triaging DNA mismatch repair deficient colorectal cancer from biopsy samples using deep learning

Contact Info

Product

Resources

About