Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

Distributed learning: a reliable privacy-preserving strategy to change multicenter collaborations using AI.

Margarita Kirienko^1,2, Martina Sollini^3,4, Gaia Ninatti²

¹Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy.

European Journal of Nuclear Medicine and Molecular Imaging

|April 13, 2021

Summary

Related Concept Videos

Distribution Reliability and Automation

Distribution Reliability and Automation

Distribution reliability in electrical power systems is critical for ensuring an uninterrupted power supply to consumers at minimal cost. According to IEEE Standard Terms, reliability is the probability that a device will function without failure over a specified time period or amount of usage. For electric power distribution, this translates to maintaining continuous power supply and addressing customer concerns over power outages. Several indices, as defined by IEEE Standard 1366-2012, are...

Issues And Trends In Healthcare Delivery System

Issues And Trends In Healthcare Delivery System

The issues and trends in healthcare delivery are constantly changing. The COVID-19 pandemic is one recent issue that wreaked havoc on healthcare systems, causing a shortage of healthcare workers, high demand for medicines and supplies, and increased medical expenditure due to a lack of insurance. Other issues include rising healthcare costs and care fragmentation.
Cost Containment
Payment for healthcare services has historically promoted adoption of costly and often unnecessary or inefficient...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

A Phase 1 and Biodistribution study of Ifabotuzumab, a humanized agonistic EphA3-targeted antibody, in patients with recurrent glioblastoma.

Clinical cancer research : an official journal of the American Association for Cancer Research·2026

Same author

Quantifying the Contribution of Bone Morphology to Implant Selection in Shoulder Arthroplasty Using CT-Based Deep Learning.

Bioengineering (Basel, Switzerland)·2026

Same author

[18F]RCCB6 PET/CT in Diagnosing Clear Cell Renal Cell Carcinomas: A Prospective Comparison Study with [18F]FDG PET/CT and Conventional Imaging.

Cancer discovery·2026

Same author

Hybrid [<sup>18</sup>F]FDG PET/MR Imaging Parameters for the Prediction of Tissue Biomarkers in Invasive Ductal Breast Cancer.

Bioengineering (Basel, Switzerland)·2026

Same author

[<sup>18</sup>F]FDG PET/MRI in Endometrial Cancer: Prospective Evaluation of Preoperative Staging, Molecular Characterization and Prognostic Assessment.

Cancers·2026

Same author

Human researchers are superior to large language models in writing a medical systematic review in a comparative multitask assessment.

Scientific reports·2025

Same journal

Insights from an unfortunate FDG contamination.

European journal of nuclear medicine and molecular imaging·2026

Same journal

Clinical performance of [<sup>68</sup>Ga]Ga-DOTA-FAPI-04 PET/CT in detecting recurrent disease in medullary thyroid carcinoma: a comparative analysis with [<sup>18</sup>F]F-FDG and [<sup>68</sup>Ga]Ga-DOTATATE PET/CT, alongside identification of suitable candidates for targeted radionuclide therapy.

European journal of nuclear medicine and molecular imaging·2026

Same journal

Whole-brain metabolic activity on pretreatment [18 F]FDG PET/CT provides prognostic information beyond metabolic tumour burden in patients with breast cancer.

European journal of nuclear medicine and molecular imaging·2026

Same journal

68Ga-Trop2 PET/CT localizes a rare erector spinae metastasis in a patient with papillary thyroid cancer.

European journal of nuclear medicine and molecular imaging·2026

Same journal

Redefining robotic image-guidance - tomographic visualization of lesions during prostate cancer surgery via gantry-free robotic SPECT.

European journal of nuclear medicine and molecular imaging·2026

Same journal

Molecular phenotyping with <sup>18</sup>F-FDG and <sup>18</sup>F-FAPI PET/CT: advancing risk stratification in in-stent restenosis.

European journal of nuclear medicine and molecular imaging·2026

See all related articles

This summary is machine-generated.

Distributed learning models show comparable performance to centralized models in medical AI applications. This approach preserves sensitive data and offers a promising solution for developing robust machine learning models.

Area of Science:

Medical Artificial Intelligence
Machine Learning in Healthcare
Distributed Learning Applications

Background:

Centralized machine learning (ML) models require data aggregation, posing privacy risks.
Local training methods may lack generalizability due to limited data.
Distributed learning offers a privacy-preserving alternative for medical ML.

Purpose of the Study:

To assess the non-inferiority of distributed learning compared to centralized and local training for ML models in medicine.
To evaluate the performance of distributed learning across various medical prediction tasks.

Main Methods:

A comprehensive literature search was conducted in PubMed/MEDLINE and EMBASE using "distributed learning" OR "federated learning" up to July 21, 2020.
Studies were included if they used medical data and were excluded if they were reviews, expert opinions, or non-English.

Keywords:

Clinical trial Distributed learning Ethics Federated learning Machine learning Privacy

Related Experiment Videos

Selected studies were analyzed based on their prediction aims (risk, diagnosis, prognosis, side effects).

Main Results:

Distributed learning demonstrated comparable performance to centralized training across most medical applications.
In 21 out of 26 papers, distributed learning was compared to centralized models, showing similar efficacy.
Distributed learning outperformed locally trained models in all but two cases, highlighting its superiority over localized approaches.

Conclusions:

Distributed learning is a reliable strategy for developing medical ML models, performing on par with centralized training.
This method enhances data privacy by keeping sensitive information localized during model development.
Distributed learning is a promising solution for advancing ML in medical research and practice, especially when large, diverse datasets are needed.