Distributed learning: Developing a predictive model based on data from multiple hospitals without data leaving the hospital – A real life proof of concept

Jochems, Arthur; Deist, Timo M.; Soest, Johan van; Eble, Michael J.; Bulens, Paul; Coucke, Philippe; Dries, Wim; Lambin, Philippe; Dekker, André

doi:10.1016/j.radonc.2016.10.002

Cited by 161 publications

(111 citation statements)

References 38 publications

Supporting

Mentioning

111

Contrasting

Order By: Relevance

“…Methods that enable the models to learn across institutional cohorts (i.e., distributed learning), rather than requiring the data to be centrally stored can create viable alternatives for effective data learning and interpretation while being cognizant of potential privacy concerns. 19 6. RECOMMENDATIONS FOR NEXT STEPS 6.A.…”

Section: Develop a Standard Nomenclature For Data Collectionmentioning

confidence: 99%

Genomics, bio specimens, and other biological data: Current status and future directions

et al. 2018

View full text Add to dashboard Cite

Section: Develop a Standard Nomenclature For Data Collectionmentioning

confidence: 99%

Genomics, bio specimens, and other biological data: Current status and future directions

et al. 2018

View full text Add to dashboard Cite

“…For these reasons, distributed learning [4] has been suggested for multiple applications including the medical eld [5] [6].…”

Section: Introductionmentioning

confidence: 99%

Secure Architectures Implementing Trusted Coalitions for Blockchained Distributed Learning (TCLearn)

et al. 2019

View full text Add to dashboard Cite

Distributed learning across a coalition of organizations allows the members of the coalition to train and share a model without sharing the data used to optimize this model. In this paper, we propose new secure architectures that guarantee preservation of data privacy, trustworthy sequence of iterative learning and equitable sharing of the learned model among each member of the coalition by using adequate encryption and blockchain mechanisms. We exemplify its deployment in the case of the distributed optimization of a deep learning convolutional neural network trained on medical images. a a is work has been partially performed during eNTERFACE'18. It has been partly funded by Wal-e-Cities, a research project co-funded by the European Regional Development Fund (ERDF) and Wallonia, Belgium.

show abstract

“…Over the last couple of decades considerable effort has been invested into compiling radiotherapy treatment planning datasets, collected through multicenter clinical trials, for retrospectively identifying predictors of treatment outcome . More recently, infrastructure has been developed to undertake such analyses on clinical databases in situ in geographically distributed clinics . One specific process being utilized in such analyses is autosegmentation of regions of interest (“structures”) enabling the rapid retrospective segmentation of possibly complex structures, while ensuring segmentation consistency.…”

Section: Introductionmentioning

confidence: 99%

“…1,2 More recently, infrastructure has been developed to undertake such analyses on clinical databases in situ in geographically distributed clinics. 3 One specific process being utilized in such analyses is autosegmentation of regions of interest ("structures") enabling the rapid retrospective segmentation of possibly complex structures, while ensuring segmentation consistency.…”

Section: Introductionmentioning

confidence: 99%

Similarity clustering‐based atlas selection for pelvic CT image segmentation

et al. 2019

View full text Add to dashboard Cite

Purpose To demonstrate selection of a small representative subset of images from a pool of images comprising a potential atlas (PA) pelvic CT set to be used for autosegmentation of a separate target image set. The aim is to balance the need for the atlas set to represent anatomical diversity with the need to minimize resources required to create a high quality atlas set (such as multiobserver delineation), while retaining access to additional information available for the PA image set. Methods Preprocessing was performed for image standardization, followed by image registration. Clustering was used to select the subset that provided the best coverage of a target dataset as measured by postregistration image intensity similarities. Tests for clustering robustness were performed including repeated clustering runs using different starting seeds and clustering repeatedly using 90% of the target dataset chosen randomly. Comparisons of coverage of a target set (comprising 711 pelvic CT images) were made for atlas sets of five images (chosen from a PA set of 39 pelvic CT and MR images) (a) at random (averaged over 50 random atlas selections), (b) based solely on image similarities within the PA set (representing prospective atlas development), (c) based on similarities within the PA set and between the PA and target dataset (representing retrospective atlas development). Comparisons were also made to coverage provided by the entire PA set of 39 images. Results Exemplar selection was highly robust with exemplar selection results being unaffected by choice of starting seed with very occasional change to one of the exemplar choices when the target set was reduced. Coverage of the target set, as measured by best normalized cross‐correlation similarity of target images to any exemplar image, provided by five well‐selected atlas images (mean = 0.6497) was more similar to coverage provided by the entire PA set (mean = 0.6658) than randomly chosen atlas subsets (mean = 0.5977). This was true both of the mean values and the shape of the distributions. Retrospective selection of atlases (mean = 0.6497) provided a very small improvement over prospective atlas selection (mean = 0.6431). All differences were significant (P < 1.0E‐10). Conclusions Selection of a small representative image set from one dataset can be utilized to develop an atlas set for either retrospective or prospective autosegmentation of a different target dataset. The coverage provided by such a judiciously selected subset has the potential to facilitate propagation of numerous retrospectively defined structures, utilizing additional information available with multimodal imaging in the atlas set, without the need to create large atlas image sets.

show abstract

Distributed learning: Developing a predictive model based on data from multiple hospitals without data leaving the hospital – A real life proof of concept

Cited by 161 publications

References 38 publications

Genomics, bio specimens, and other biological data: Current status and future directions

Genomics, bio specimens, and other biological data: Current status and future directions

Secure Architectures Implementing Trusted Coalitions for Blockchained Distributed Learning (TCLearn)

Similarity clustering‐based atlas selection for pelvic CT image segmentation

Contact Info

Product

Resources

About