Dataset Information

Federated learning enables big data for rare cancer boundary detection.

ABSTRACT: Although machine learning (ML) has shown promise across disciplines, out-of-sample generalizability is concerning. This is currently addressed by sharing multi-site data, but such centralization is challenging/infeasible to scale due to various limitations. Federated ML (FL) provides an alternative paradigm for accurate and generalizable ML, by only sharing numerical model updates. Here we present the largest FL study to-date, involving data from 71 sites across 6 continents, to generate an automatic tumor boundary detector for the rare disease of glioblastoma, reporting the largest such dataset in the literature (n = 6, 314). We demonstrate a 33% delineation improvement for the surgically targetable tumor, and 23% for the complete tumor extent, over a publicly trained model. We anticipate our study to: 1) enable more healthcare studies informed by large diverse data, ensuring meaningful results for rare diseases and underrepresented populations, 2) facilitate further analyses for glioblastoma by releasing our consensus model, and 3) demonstrate the FL effectiveness at such scale and task-complexity as a paradigm shift for multi-site collaborations, alleviating the need for data-sharing.

SUBMITTER: Pati S

PROVIDER: S-EPMC9722782 | biostudies-literature | 2022 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Federated learning enables big data for rare cancer boundary detection.

Pati Sarthak S Baid Ujjwal U Edwards Brandon B Sheller Micah M Wang Shih-Han SH Reina G Anthony GA Foley Patrick P Gruzdev Alexey A Karkada Deepthi D Davatzikos Christos C Sako Chiharu C Ghodasara Satyam S Bilello Michel M Mohan Suyash S Vollmuth Philipp P Brugnara Gianluca G Preetha Chandrakanth J CJ Sahm Felix F Maier-Hein Klaus K Zenk Maximilian M Bendszus Martin M Wick Wolfgang W Calabrese Evan E Rudie Jeffrey J Villanueva-Meyer Javier J Cha Soonmee S Ingalhalikar Madhura M Jadhav Manali M Pandey Umang U Saini Jitender J Garrett John J Larson Matthew M Jeraj Robert R Currie Stuart S Frood Russell R Fatania Kavi K Huang Raymond Y RY Chang Ken K Balaña Carmen C Capellades Jaume J Puig Josep J Trenkler Johannes J Pichler Josef J Necker Georg G Haunschmidt Andreas A Meckel Stephan S Shukla Gaurav G Liem Spencer S Alexander Gregory S GS Lombardo Joseph J Palmer Joshua D JD Flanders Adam E AE Dicker Adam P AP Sair Haris I HI Jones Craig K CK Venkataraman Archana A Jiang Meirui M So Tiffany Y TY Chen Cheng C Heng Pheng Ann PA Dou Qi Q Kozubek Michal M Lux Filip F Michálek Jan J Matula Petr P Keřkovský Miloš M Kopřivová Tereza T Dostál Marek M Vybíhal Václav V Vogelbaum Michael A MA Mitchell J Ross JR Farinhas Joaquim J Maldjian Joseph A JA Yogananda Chandan Ganesh Bangalore CGB Pinho Marco C MC Reddy Divya D Holcomb James J Wagner Benjamin C BC Ellingson Benjamin M BM Cloughesy Timothy F TF Raymond Catalina C Oughourlian Talia T Hagiwara Akifumi A Wang Chencai C To Minh-Son MS Bhardwaj Sargam S Chong Chee C Agzarian Marc M Falcão Alexandre Xavier AX Martins Samuel B SB Teixeira Bernardo C A BCA Sprenger Flávia F Menotti David D Lucio Diego R DR LaMontagne Pamela P Marcus Daniel D Wiestler Benedikt B Kofler Florian F Ezhov Ivan I Metz Marie M Jain Rajan R Lee Matthew M Lui Yvonne W YW McKinley Richard R Slotboom Johannes J Radojewski Piotr P Meier Raphael R Wiest Roland R Murcia Derrick D Fu Eric E Haas Rourke R Thompson John J Ormond David Ryan DR Badve Chaitra C Sloan Andrew E AE Vadmal Vachan V Waite Kristin K Colen Rivka R RR Pei Linmin L Ak Murat M Srinivasan Ashok A Bapuraj J Rajiv JR Rao Arvind A Wang Nicholas N Yoshiaki Ota O Moritani Toshio T Turk Sevcan S Lee Joonsang J Prabhudesai Snehal S Morón Fanny F Mandel Jacob J Kamnitsas Konstantinos K Glocker Ben B Dixon Luke V M LVM Williams Matthew M Zampakis Peter P Panagiotopoulos Vasileios V Tsiganos Panagiotis P Alexiou Sotiris S Haliassos Ilias I Zacharaki Evangelia I EI Moustakas Konstantinos K Kalogeropoulou Christina C Kardamakis Dimitrios M DM Choi Yoon Seong YS Lee Seung-Koo SK Chang Jong Hee JH Ahn Sung Soo SS Luo Bing B Poisson Laila L Wen Ning N Tiwari Pallavi P Verma Ruchika R Bareja Rohan R Yadav Ipsa I Chen Jonathan J Kumar Neeraj N Smits Marion M van der Voort Sebastian R SR Alafandi Ahmed A Incekara Fatih F Wijnenga Maarten M J MMJ Kapsas Georgios G Gahrmann Renske R Schouten Joost W JW Dubbink Hendrikus J HJ Vincent Arnaud J P E AJPE van den Bent Martin J MJ French Pim J PJ Klein Stefan S Yuan Yading Y Sharma Sonam S Tseng Tzu-Chi TC Adabi Saba S Niclou Simone P SP Keunen Olivier O Hau Ann-Christin AC Vallières Martin M Fortin David D Lepage Martin M Landman Bennett B Ramadass Karthik K Xu Kaiwen K Chotai Silky S Chambless Lola B LB Mistry Akshitkumar A Thompson Reid C RC Gusev Yuriy Y Bhuvaneshwar Krithika K Sayah Anousheh A Bencheqroun Camelia C Belouali Anas A Madhavan Subha S Booth Thomas C TC Chelliah Alysha A Modat Marc M Shuaib Haris H Dragos Carmen C Abayazeed Aly A Kolodziej Kenneth K Hill Michael M Abbassy Ahmed A Gamal Shady S Mekhaimar Mahmoud M Qayati Mohamed M Reyes Mauricio M Park Ji Eun JE Yun Jihye J Kim Ho Sung HS Mahajan Abhishek A Muzi Mark M Benson Sean S Beets-Tan Regina G H RGH Teuwen Jonas J Herrera-Trujillo Alejandro A Trujillo Maria M Escobar William W Abello Ana A Bernal Jose J Gómez Jhon J Choi Joseph J Baek Stephen S Kim Yusung Y Ismael Heba H Allen Bryan B Buatti John M JM Kotrotsou Aikaterini A Li Hongwei H Weiss Tobias T Weller Michael M Bink Andrea A Pouymayou Bertrand B Shaykh Hassan F HF Saltz Joel J Prasanna Prateek P Shrestha Sampurna S Mani Kartik M KM Payne David D Kurc Tahsin T Pelaez Enrique E Franco-Maldonado Heydy H Loayza Francis F Quevedo Sebastian S Guevara Pamela P Torche Esteban E Mendoza Cristobal C Vera Franco F Ríos Elvis E López Eduardo E Velastin Sergio A SA Ogbole Godwin G Soneye Mayowa M Oyekunle Dotun D Odafe-Oyibotha Olubunmi O Osobu Babatunde B Shu'aibu Mustapha M Dorcas Adeleye A Dako Farouk F Simpson Amber L AL Hamghalam Mohammad M Peoples Jacob J JJ Hu Ricky R Tran Anh A Cutler Danielle D Moraes Fabio Y FY Boss Michael A MA Gimpel James J Veettil Deepak Kattil DK Schmidt Kendall K Bialecki Brian B Marella Sailaja S Price Cynthia C Cimino Lisa L Apgar Charles C Shah Prashant P Menze Bjoern B Barnholtz-Sloan Jill S JS Martin Jason J Bakas Spyridon S

Nature communications 20221205 1

Although machine learning (ML) has shown promise across disciplines, out-of-sample generalizability is concerning. This is currently addressed by sharing multi-site data, but such centralization is challenging/infeasible to scale due to various limitations. Federated ML (FL) provides an alternative paradigm for accurate and generalizable ML, by only sharing numerical model updates. Here we present the largest FL study to-date, involving data from 71 sites across 6 continents, to generate an auto ...[more]

PMID: 36470898

Similar Datasets

Project description:Malignant glioma is the uncontrollable growth of cells in the spinal cord and brain that look similar to the normal glial cells. The most essential part of the nervous system is glial cells, which support the brain's functioning prominently. However, with the evolution of glioma, tumours form that invade healthy tissues in the brain, leading to neurological impairment, seizures, hormonal dysregulation, and venous thromboembolism. Medical tests, including medical resonance imaging (MRI), computed tomography (CT) scans, biopsy, and electroencephalograms are used for early detection of glioma. However, these tests are expensive and may cause irritation and allergic reactions due to ionizing radiation. The deep learning models are highly optimal for disease prediction, however, the challenge associated with it is the requirement for substantial memory and storage to amalgamate the patient's information at a centralized location. Additionally, it also has patient data-privacy concerns leading to anonymous information generalization, regulatory compliance issues, and data leakage challenges. Therefore, in the proposed work, a distributed and privacy-preserved horizontal federated learning-based malignant glioma disease detection model has been developed by employing 5 and 10 different clients' architectures in independent and identically distributed (IID) and non-IID distributions. Initially, for developing this model, the collection of the MRI scans of non-tumour and glioma tumours has been done, which are further pre-processed by performing data balancing and image resizing. The configuration and development of the pre-trained MobileNetV2 base model have been performed, which is then applied to the federated learning(FL) framework. The configurations of this model have been kept as 0.001, Adam, 32, 10, 10, FedAVG, and 10 for learning rate, optimizer, batch size, local epochs, global epochs, aggregation, and rounds, respectively. The proposed model has provided the most prominent accuracy with 5 clients' architecture as 99.76% and 99.71% for IID and non-IID distributions, respectively. These outcomes demonstrate that the model is highly optimized and generalizes the improved outcomes when compared to the state-of-the-art models.

Project description:BackgroundFederated learning (FL) is a newly proposed machine-learning method that uses a decentralized dataset. Since data transfer is not necessary for the learning process in FL, there is a significant advantage in protecting personal privacy. Therefore, many studies are being actively conducted in the applications of FL for diverse areas.ObjectiveThe aim of this study was to evaluate the reliability and performance of FL using three benchmark datasets, including a clinical benchmark dataset.MethodsTo evaluate FL in a realistic setting, we implemented FL using a client-server architecture with Python. The implemented client-server version of the FL software was deployed to Amazon Web Services. Modified National Institute of Standards and Technology (MNIST), Medical Information Mart for Intensive Care-III (MIMIC-III), and electrocardiogram (ECG) datasets were used to evaluate the performance of FL. To test FL in a realistic setting, the MNIST dataset was split into 10 different clients, with one digit for each client. In addition, we conducted four different experiments according to basic, imbalanced, skewed, and a combination of imbalanced and skewed data distributions. We also compared the performance of FL to that of the state-of-the-art method with respect to in-hospital mortality using the MIMIC-III dataset. Likewise, we conducted experiments comparing basic and imbalanced data distributions using MIMIC-III and ECG data.ResultsFL on the basic MNIST dataset with 10 clients achieved an area under the receiver operating characteristic curve (AUROC) of 0.997 and an F1-score of 0.946. The experiment with the imbalanced MNIST dataset achieved an AUROC of 0.995 and an F1-score of 0.921. The experiment with the skewed MNIST dataset achieved an AUROC of 0.992 and an F1-score of 0.905. Finally, the combined imbalanced and skewed experiment achieved an AUROC of 0.990 and an F1-score of 0.891. The basic FL on in-hospital mortality using MIMIC-III data achieved an AUROC of 0.850 and an F1-score of 0.944, while the experiment with the imbalanced MIMIC-III dataset achieved an AUROC of 0.850 and an F1-score of 0.943. For ECG classification, the basic FL achieved an AUROC of 0.938 and an F1-score of 0.807, and the imbalanced ECG dataset achieved an AUROC of 0.943 and an F1-score of 0.807.ConclusionsFL demonstrated comparative performance on different benchmark datasets. In addition, FL demonstrated reliable performance in cases where the distribution was imbalanced, skewed, and extreme, reflecting the real-life scenario in which data distributions from various hospitals are different. FL can achieve high performance while maintaining privacy protection because there is no requirement to centralize the data.

Dataset Information

Federated learning enables big data for rare cancer boundary detection.

Publications

Federated learning enables big data for rare cancer boundary detection.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets