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

Evaluating distributed-learning on real-world obstetrics data: comparing distributed, centralized and local models

This study focused on comparing distributed learning models with centralized and local models, assessing their efficacy in predicting specific delivery and patient-related outcomes in obstetrics using real-world data. The predictions focus on key moments in the obstetric care process, including discharge and various stages of hospitalization. Our analysis: using 6 different machine learning methods like Decision Trees, Bayesian methods, Stochastic Gradient Descent, K-nearest neighbors, AdaBoost, and Multi-layer Perceptron and 19 different variables with various distributions and types, revealed that distributed models were at least equal, and often superior, to centralized versions and local versions. We also describe thoroughly the preprocessing stage in order to help others implement this method in real-world scenarios. The preprocessing steps included cleaning and harmonizing missing values, handling missing data and encoding categorical variables with multisite logic. Even though the type of machine learning model and the distribution of the outcome variable can impact the result, we reached results of 66% being superior to the centralized and local counterpart and 77% being better than the centralized with AdaBoost. Our experiments also shed light in the preprocessing steps required to implement distributed models in a real-world scenario. Our results advocate for distributed learning as a promising tool for applying machine learning in clinical settings, particularly when privacy and data security are paramount, thus offering a robust solution for privacy-concerned clinical applications.

Image-based artificial intelligence for the prediction of pathological complete response to neoadjuvant chemoradiotherapy in patients with rectal cancer: a systematic review and meta-analysis

OBJECTIVE

Artificial intelligence (AI) holds enormous potential for noninvasively identifying patients with rectal cancer who could achieve pathological complete response (pCR) following neoadjuvant chemoradiotherapy (nCRT). We aimed to conduct a meta-analysis to summarize the diagnostic performance of image-based AI models for predicting pCR to nCRT in patients with rectal cancer.

METHODS

This study followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. A literature search of PubMed, Embase, Cochrane Library, and Web of Science was performed from inception to July 29, 2023. Studies that developed or utilized AI models for predicting pCR to nCRT in rectal cancer from medical images were included. The Quality Assessment of Diagnostic Accuracy Studies-AI was used to appraise the methodological quality of the studies. The bivariate random-effects model was used to summarize the individual sensitivities, specificities, and areas-under-the-curve (AUCs). Subgroup and meta-regression analyses were conducted to identify potential sources of heterogeneity. Protocol for this study was registered with PROSPERO (CRD42022382374).

RESULTS

Thirty-four studies (9933 patients) were identified. Pooled estimates of sensitivity, specificity, and AUC of AI models for pCR prediction were 82% (95% CI: 76-87%), 84% (95% CI: 79-88%), and 90% (95% CI: 87-92%), respectively. Higher specificity was seen for the Asian population, low risk of bias, and deep-learning, compared with the non-Asian population, high risk of bias, and radiomics (all P < 0.05). Single-center had a higher sensitivity than multi-center (P = 0.001). The retrospective design had lower sensitivity (P = 0.012) but higher specificity (P < 0.001) than the prospective design. MRI showed higher sensitivity (P = 0.001) but lower specificity (P = 0.044) than non-MRI. The sensitivity and specificity of internal validation were higher than those of external validation (both P = 0.005).

CONCLUSIONS

Image-based AI models exhibited favorable performance for predicting pCR to nCRT in rectal cancer. However, further clinical trials are warranted to verify the findings.

SYNDSURV: A simple framework for survival analysis with data distributed across multiple institutions

Data sharing among different institutions represents one of the major challenges in developing distributed machine learning approaches, especially when data is sensitive, such as in medical applications. Federated learning is a possible solution, but requires fast communications and flawless security. Here, we propose SYNDSURV (SYNthetic Distributed SURVival), an alternative approach that simplifies the current state-of-the-art paradigm by allowing different centres to generate local simulated instances from real data and then gather them into a centralised hub, where an Artificial Intelligence (AI) model can learn in a standard way. The main advantage of this procedure is that it is model-agnostic, therefore prediction models can be directly applied in distributed applications without requiring particular adaptations as the current federated approaches do. To show the validity of our approach for medical applications, we tested it on a survival analysis task, offering a viable alternative to train AI models on distributed data. While federated learning has been mainly optimised for gradient-based approaches so far, our framework works with any predictive method, proving to be a comparable way of performing distributed learning without being too demanding towards each participating institute in terms of infrastructural requirements.

CODA: an open-source platform for federated analysis and machine learning on distributed healthcare data

OBJECTIVES

Distributed computations facilitate multi-institutional data analysis while avoiding the costs and complexity of data pooling. Existing approaches lack crucial features, such as built-in medical standards and terminologies, no-code data visualizations, explicit disclosure control mechanisms, and support for basic statistical computations, in addition to gradient-based optimization capabilities.

MATERIALS AND METHODS

We describe the development of the Collaborative Data Analysis (CODA) platform, and the design choices undertaken to address the key needs identified during our survey of stakeholders. We use a public dataset (MIMIC-IV) to demonstrate end-to-end multi-modal FL using CODA. We assessed the technical feasibility of deploying the CODA platform at 9 hospitals in Canada, describe implementation challenges, and evaluate its scalability on large patient populations.

RESULTS

The CODA platform was designed, developed, and deployed between January 2020 and January 2023. Software code, documentation, and technical documents were released under an open-source license. Multi-modal federated averaging is illustrated using the MIMIC-IV and MIMIC-CXR datasets. To date, 8 out of the 9 participating sites have successfully deployed the platform, with a total enrolment of >1M patients. Mapping data from legacy systems to FHIR was the biggest barrier to implementation.

DISCUSSION AND CONCLUSION

The CODA platform was developed and successfully deployed in a public healthcare setting in Canada, with heterogeneous information technology systems and capabilities. Ongoing efforts will use the platform to develop and prospectively validate models for risk assessment, proactive monitoring, and resource usage. Further work will also make tools available to facilitate migration from legacy formats to FHIR and DICOM.

Data pyramid structure for optimizing EUS-based GISTs diagnosis in multi-center analysis with missing label

This study introduces the Data Pyramid Structure (DPS) to address data sparsity and missing labels in medical image analysis. The DPS optimizes multi-task learning and enables sustainable expansion of multi-center data analysis. Specifically, It facilitates attribute prediction and malignant tumor diagnosis tasks by implementing a segmentation and aggregation strategy on data with absent attribute labels. To leverage multi-center data, we propose the Unified Ensemble Learning Framework (UELF) and the Unified Federated Learning Framework (UFLF), which incorporate strategies for data transfer and incremental learning in scenarios with missing labels. The proposed method was evaluated on a challenging EUS patient dataset from five centers, achieving promising diagnostic performance. The average accuracy was 0.984 with an AUC of 0.927 for multi-center analysis, surpassing state-of-the-art approaches. The interpretability of the predictions further highlights the potential clinical relevance of our method.

Differential privacy preserved federated transfer learning for multi-institutional 68Ga-PET image artefact detection and disentanglement

PURPOSE

Image artefacts continue to pose challenges in clinical molecular imaging, resulting in misdiagnoses, additional radiation doses to patients and financial costs. Mismatch and halo artefacts occur frequently in gallium-68 (68Ga)-labelled compounds whole-body PET/CT imaging. Correcting for these artefacts is not straightforward and requires algorithmic developments, given that conventional techniques have failed to address them adequately. In the current study, we employed differential privacy-preserving federated transfer learning (FTL) to manage clinical data sharing and tackle privacy issues for building centre-specific models that detect and correct artefacts present in PET images.

METHODS

Altogether, 1413 patients with 68Ga prostate-specific membrane antigen (PSMA)/DOTA-TATE (TOC) PET/CT scans from 3 countries, including 8 different centres, were enrolled in this study. CT-based attenuation and scatter correction (CT-ASC) was used in all centres for quantitative PET reconstruction. Prior to model training, an experienced nuclear medicine physician reviewed all images to ensure the use of high-quality, artefact-free PET images (421 patients' images). A deep neural network (modified U2Net) was trained on 80% of the artefact-free PET images to utilize centre-based (CeBa), centralized (CeZe) and the proposed differential privacy FTL frameworks. Quantitative analysis was performed in 20% of the clean data (with no artefacts) in each centre. A panel of two nuclear medicine physicians conducted qualitative assessment of image quality, diagnostic confidence and image artefacts in 128 patients with artefacts (256 images for CT-ASC and FTL-ASC).

RESULTS

The three approaches investigated in this study for 68Ga-PET imaging (CeBa, CeZe and FTL) resulted in a mean absolute error (MAE) of 0.42 ± 0.21 (CI 95%: 0.38 to 0.47), 0.32 ± 0.23 (CI 95%: 0.27 to 0.37) and 0.28 ± 0.15 (CI 95%: 0.25 to 0.31), respectively. Statistical analysis using the Wilcoxon test revealed significant differences between the three approaches, with FTL outperforming CeBa and CeZe (p-value < 0.05) in the clean test set. The qualitative assessment demonstrated that FTL-ASC significantly improved image quality and diagnostic confidence and decreased image artefacts, compared to CT-ASC in 68Ga-PET imaging. In addition, mismatch and halo artefacts were successfully detected and disentangled in the chest, abdomen and pelvic regions in 68Ga-PET imaging.

CONCLUSION

The proposed approach benefits from using large datasets from multiple centres while preserving patient privacy. Qualitative assessment by nuclear medicine physicians showed that the proposed model correctly addressed two main challenging artefacts in 68Ga-PET imaging. This technique could be integrated in the clinic for 68Ga-PET imaging artefact detection and disentanglement using multicentric heterogeneous datasets.

Federated and distributed learning applications for electronic health records and structured medical data: a scoping review

OBJECTIVES

Federated learning (FL) has gained popularity in clinical research in recent years to facilitate privacy-preserving collaboration. Structured data, one of the most prevalent forms of clinical data, has experienced significant growth in volume concurrently, notably with the widespread adoption of electronic health records in clinical practice. This review examines FL applications on structured medical data, identifies contemporary limitations, and discusses potential innovations.

MATERIALS AND METHODS

We searched 5 databases, SCOPUS, MEDLINE, Web of Science, Embase, and CINAHL, to identify articles that applied FL to structured medical data and reported results following the PRISMA guidelines. Each selected publication was evaluated from 3 primary perspectives, including data quality, modeling strategies, and FL frameworks.

RESULTS

Out of the 1193 papers screened, 34 met the inclusion criteria, with each article consisting of one or more studies that used FL to handle structured clinical/medical data. Of these, 24 utilized data acquired from electronic health records, with clinical predictions and association studies being the most common clinical research tasks that FL was applied to. Only one article exclusively explored the vertical FL setting, while the remaining 33 explored the horizontal FL setting, with only 14 discussing comparisons between single-site (local) and FL (global) analysis.

CONCLUSIONS

The existing FL applications on structured medical data lack sufficient evaluations of clinically meaningful benefits, particularly when compared to single-site analyses. Therefore, it is crucial for future FL applications to prioritize clinical motivations and develop designs and methodologies that can effectively support and aid clinical practice and research.

Decentralized collaborative multi-institutional PET attenuation and scatter correction using federated deep learning

PURPOSE

Attenuation correction and scatter compensation (AC/SC) are two main steps toward quantitative PET imaging, which remain challenging in PET-only and PET/MRI systems. These can be effectively tackled via deep learning (DL) methods. However, trustworthy, and generalizable DL models commonly require well-curated, heterogeneous, and large datasets from multiple clinical centers. At the same time, owing to legal/ethical issues and privacy concerns, forming a large collective, centralized dataset poses significant challenges. In this work, we aimed to develop a DL-based model in a multicenter setting without direct sharing of data using federated learning (FL) for AC/SC of PET images.

METHODS

Non-attenuation/scatter corrected and CT-based attenuation/scatter corrected (CT-ASC) ¹⁸F-FDG PET images of 300 patients were enrolled in this study. The dataset consisted of 6 different centers, each with 50 patients, with scanner, image acquisition, and reconstruction protocols varying across the centers. CT-based ASC PET images served as the standard reference. All images were reviewed to include high-quality and artifact-free PET images. Both corrected and uncorrected PET images were converted to standardized uptake values (SUVs). We used a modified nested U-Net utilizing residual U-block in a U-shape architecture. We evaluated two FL models, namely sequential (FL-SQ) and parallel (FL-PL) and compared their performance with the baseline centralized (CZ) learning model wherein the data were pooled to one server, as well as center-based (CB) models where for each center the model was built and evaluated separately. Data from each center were divided to contribute to training (30 patients), validation (10 patients), and test sets (10 patients). Final evaluations and reports were performed on 60 patients (10 patients from each center).

RESULTS

In terms of percent SUV absolute relative error (ARE%), both FL-SQ (CI:12.21-14.81%) and FL-PL (CI:11.82-13.84%) models demonstrated excellent agreement with the centralized framework (CI:10.32-12.00%), while FL-based algorithms improved model performance by over 11% compared to CB training strategy (CI: 22.34-26.10%). Furthermore, the Mann-Whitney test between different strategies revealed no significant differences between CZ and FL-based algorithms (p-value > 0.05) in center-categorized mode. At the same time, a significant difference was observed between the different training approaches on the overall dataset (p-value < 0.05). In addition, voxel-wise comparison, with respect to reference CT-ASC, exhibited similar performance for images predicted by CZ (R² = 0.94), FL-SQ (R² = 0.93), and FL-PL (R² = 0.92), while CB model achieved a far lower coefficient of determination (R² = 0.74). Despite the strong correlations between CZ and FL-based methods compared to reference CT-ASC, a slight underestimation of predicted voxel values was observed.

CONCLUSION

Deep learning-based models provide promising results toward quantitative PET image reconstruction. Specifically, we developed two FL models and compared their performance with center-based and centralized models. The proposed FL-based models achieved higher performance compared to center-based models, comparable with centralized models. Our work provided strong empirical evidence that the FL framework can fully benefit from the generalizability and robustness of DL models used for AC/SC in PET, while obviating the need for the direct sharing of datasets between clinical imaging centers.

Federated Learning in Health care Using Structured Medical Data

The success of machine learning-based studies is largely subjected to accessing a large amount of data. However, accessing such data is typically not feasible within a single health system/hospital. Although multicenter studies are the most effective way to access a vast amount of data, sharing data outside the institutes involves legal, business, and technical challenges. Federated learning (FL) is a newly proposed machine learning framework for multicenter studies, tackling data-sharing issues across participant institutes. The promise of FL is simple. FL facilitates multicenter studies without losing data access control and allows the construction of a global model by aggregating local models trained from participant institutes. This article reviewed recently published studies that utilized FL in clinical studies with structured medical data. In addition, challenges and open questions in FL in clinical studies with structured medical data were discussed.

Federated machine learning for a facilitated implementation of Artificial Intelligence in healthcare - a proof of concept study for the prediction of coronary artery calcification scores

The implementation of Artificial Intelligence (AI) still faces significant hurdles and one key factor is the access to data. One approach that could support that is federated machine learning (FL) since it allows for privacy preserving data access. For this proof of concept, a prediction model for coronary artery calcification scores (CACS) has been applied. The FL was trained based on the data in the different institutions, while the centralized machine learning model was trained on one allocation of data. Both algorithms predict patients with risk scores ≥5 based on age, biological sex, waist circumference, dyslipidemia and HbA1c. The centralized model yields a sensitivity of c. 66% and a specificity of c. 70%. The FL slightly outperforms that with a sensitivity of 67% while slightly underperforming it with a specificity of 69%. It could be demonstrated that CACS prediction is feasible via both, a centralized and an FL approach, and that both show very comparable accuracy. In order to increase accuracy, additional and a higher volume of patient data is required and for that FL is utterly necessary. The developed "CACulator" serves as proof of concept, is available as research tool and shall support future research to facilitate AI implementation.

Artificial intelligence and machine learning in cancer imaging

An increasing array of tools is being developed using artificial intelligence (AI) and machine learning (ML) for cancer imaging. The development of an optimal tool requires multidisciplinary engagement to ensure that the appropriate use case is met, as well as to undertake robust development and testing prior to its adoption into healthcare systems. This multidisciplinary review highlights key developments in the field. We discuss the challenges and opportunities of AI and ML in cancer imaging; considerations for the development of algorithms into tools that can be widely used and disseminated; and the development of the ecosystem needed to promote growth of AI and ML in cancer imaging.

Image Embeddings Extracted from CNNs Outperform Other Transfer Learning Approaches in Classification of Chest Radiographs

To identify the best transfer learning approach for the identification of the most frequent abnormalities on chest radiographs (CXRs), we used embeddings extracted from pretrained convolutional neural networks (CNNs). An explainable AI (XAI) model was applied to interpret black-box model predictions and assess its performance. Seven CNNs were trained on CheXpert. Three transfer learning approaches were thereafter applied to a local dataset. The classification results were ensembled using simple and entropy-weighted averaging. We applied Grad-CAM (an XAI model) to produce a saliency map. Grad-CAM maps were compared to manually extracted regions of interest, and the training time was recorded. The best transfer learning model was that which used image embeddings and random forest with simple averaging, with an average AUC of 0.856. Grad-CAM maps showed that the models focused on specific features of each CXR. CNNs pretrained on a large public dataset of medical images can be exploited as feature extractors for tasks of interest. The extracted image embeddings contain relevant information that can be used to train an additional classifier with satisfactory performance on an independent dataset, demonstrating it to be the optimal transfer learning strategy and overcoming the need for large private datasets, extensive computational resources, and long training times.

Diagnostic accuracy and potential covariates of artificial intelligence for diagnosing orthopedic fractures: a systematic literature review and meta-analysis

OBJECTIVES

To systematically quantify the diagnostic accuracy and identify potential covariates affecting the performance of artificial intelligence (AI) in diagnosing orthopedic fractures.

METHODS

PubMed, Embase, Web of Science, and Cochrane Library were systematically searched for studies on AI applications in diagnosing orthopedic fractures from inception to September 29, 2021. Pooled sensitivity and specificity and the area under the receiver operating characteristic curves (AUC) were obtained. This study was registered in the PROSPERO database prior to initiation (CRD 42021254618).

RESULTS

Thirty-nine were eligible for quantitative analysis. The overall pooled AUC, sensitivity, and specificity were 0.96 (95% CI 0.94-0.98), 90% (95% CI 87-92%), and 92% (95% CI 90-94%), respectively. In subgroup analyses, multicenter designed studies yielded higher sensitivity (92% vs. 88%) and specificity (94% vs. 91%) than single-center studies. AI demonstrated higher sensitivity with transfer learning (with vs. without: 92% vs. 87%) or data augmentation (with vs. without: 92% vs. 87%), compared to those without. Utilizing plain X-rays as input images for AI achieved results comparable to CT (AUC 0.96 vs. 0.96). Moreover, AI achieved comparable results to humans (AUC 0.97 vs. 0.97) and better results than non-expert human readers (AUC 0.98 vs. 0.96; sensitivity 95% vs. 88%).

CONCLUSIONS

AI demonstrated high accuracy in diagnosing orthopedic fractures from medical images. Larger-scale studies with higher design quality are needed to validate our findings.

KEY POINTS

• Multicenter study design, application of transfer learning, and data augmentation are closely related to improving the performance of artificial intelligence models in diagnosing orthopedic fractures. • Utilizing plain X-rays as input images for AI to diagnose fractures achieved results comparable to CT (AUC 0.96 vs. 0.96). • AI achieved comparable results to humans (AUC 0.97 vs. 0.97) but was superior to non-expert human readers (AUC 0.98 vs. 0.96, sensitivity 95% vs. 88%) in diagnosing fractures.

Novel technical and privacy-preserving technology for artificial intelligence in ophthalmology

PURPOSE OF REVIEW

The application of artificial intelligence (AI) in medicine and ophthalmology has experienced exponential breakthroughs in recent years in diagnosis, prognosis, and aiding clinical decision-making. The use of digital data has also heralded the need for privacy-preserving technology to protect patient confidentiality and to guard against threats such as adversarial attacks. Hence, this review aims to outline novel AI-based systems for ophthalmology use, privacy-preserving measures, potential challenges, and future directions of each.

RECENT FINDINGS

Several key AI algorithms used to improve disease detection and outcomes include: Data-driven, imagedriven, natural language processing (NLP)-driven, genomics-driven, and multimodality algorithms. However, deep learning systems are susceptible to adversarial attacks, and use of data for training models is associated with privacy concerns. Several data protection methods address these concerns in the form of blockchain technology, federated learning, and generative adversarial networks.

SUMMARY

AI-applications have vast potential to meet many eyecare needs, consequently reducing burden on scarce healthcare resources. A pertinent challenge would be to maintain data privacy and confidentiality while supporting AI endeavors, where data protection methods would need to rapidly evolve with AI technology needs. Ultimately, for AI to succeed in medicine and ophthalmology, a balance would need to be found between innovation and privacy.

Decentralized Distributed Multi-institutional PET Image Segmentation Using a Federated Deep Learning Framework

PURPOSE

The generalizability and trustworthiness of deep learning (DL)-based algorithms depend on the size and heterogeneity of training datasets. However, because of patient privacy concerns and ethical and legal issues, sharing medical images between different centers is restricted. Our objective is to build a federated DL-based framework for PET image segmentation utilizing a multicentric dataset and to compare its performance with the centralized DL approach.

METHODS

PET images from 405 head and neck cancer patients from 9 different centers formed the basis of this study. All tumors were segmented manually. PET images converted to SUV maps were resampled to isotropic voxels (3 × 3 × 3 mm3) and then normalized. PET image subvolumes (12 × 12 × 12 cm3) consisting of whole tumors and background were analyzed. Data from each center were divided into train/validation (80% of patients) and test sets (20% of patients). The modified R2U-Net was used as core DL model. A parallel federated DL model was developed and compared with the centralized approach where the data sets are pooled to one server. Segmentation metrics, including Dice similarity and Jaccard coefficients, percent relative errors (RE%) of SUVpeak, SUVmean, SUVmedian, SUVmax, metabolic tumor volume, and total lesion glycolysis were computed and compared with manual delineations.

RESULTS

The performance of the centralized versus federated DL methods was nearly identical for segmentation metrics: Dice (0.84 ± 0.06 vs 0.84 ± 0.05) and Jaccard (0.73 ± 0.08 vs 0.73 ± 0.07). For quantitative PET parameters, we obtained comparable RE% for SUVmean (6.43% ± 4.72% vs 6.61% ± 5.42%), metabolic tumor volume (12.2% ± 16.2% vs 12.1% ± 15.89%), and total lesion glycolysis (6.93% ± 9.6% vs 7.07% ± 9.85%) and negligible RE% for SUVmax and SUVpeak. No significant differences in performance (P > 0.05) between the 2 frameworks (centralized vs federated) were observed.

CONCLUSION

The developed federated DL model achieved comparable quantitative performance with respect to the centralized DL model. Federated DL models could provide robust and generalizable segmentation, while addressing patient privacy and legal and ethical issues in clinical data sharing.

Recent Applications of Artificial Intelligence in Radiotherapy: Where We Are and Beyond

In recent decades, artificial intelligence (AI) tools have been applied in many medical fields, opening the possibility of finding novel solutions for managing very complex and multifactorial problems, such as those commonly encountered in radiotherapy (RT). We conducted a PubMed and Scopus search to identify the AI application field in RT limited to the last four years. In total, 1824 original papers were identified, and 921 were analyzed by considering the phase of the RT workflow according to the applied AI approaches. AI permits the processing of large quantities of information, data, and images stored in RT oncology information systems, a process that is not manageable for individuals or groups. AI allows the iterative application of complex tasks in large datasets (e.g., delineating normal tissues or finding optimal planning solutions) and might support the entire community working in the various sectors of RT, as summarized in this overview. AI-based tools are now on the roadmap for RT and have been applied to the entire workflow, mainly for segmentation, the generation of synthetic images, and outcome prediction. Several concerns were raised, including the need for harmonization while overcoming ethical, legal, and skill barriers.

Ensemble Learning Using Individual Neonatal Data for Seizure Detection

Objective: Sharing medical data between institutions is difficult in practice due to data protection laws and official procedures within institutions. Therefore, most existing algorithms are trained on relatively small electroencephalogram (EEG) data sets which is likely to be detrimental to prediction accuracy. In this work, we simulate a case when the data can not be shared by splitting the publicly available data set into disjoint sets representing data in individual institutions. Methods and procedures: We propose to train a (local) detector in each institution and aggregate their individual predictions into one final prediction. Four aggregation schemes are compared, namely, the majority vote, the mean, the weighted mean and the Dawid-Skene method. The method was validated on an independent data set using only a subset of EEG channels. Results: The ensemble reaches accuracy comparable to a single detector trained on all the data when sufficient amount of data is available in each institution. Conclusion: The weighted mean aggregation scheme showed best performance, it was only marginally outperformed by the Dawid–Skene method when local detectors approach performance of a single detector trained on all available data. Clinical impact: Ensemble learning allows training of reliable algorithms for neonatal EEG analysis without a need to share the potentially sensitive EEG data between institutions.

Privacy preserving distributed learning classifiers - Sequential learning with small sets of data

BACKGROUND

Artificial intelligence (AI) typically requires a significant amount of high-quality data to build reliable models, where gathering enough data within a single institution can be particularly challenging. In this study we investigated the impact of using sequential learning to exploit very small, siloed sets of clinical and imaging data to train AI models. Furthermore, we evaluated the capacity of such models to achieve equivalent performance when compared to models trained with the same data over a single centralized database.

METHODS

We propose a privacy preserving distributed learning framework, learning sequentially from each dataset. The framework is applied to three machine learning algorithms: Logistic Regression, Support Vector Machines (SVM), and Perceptron. The models were evaluated using four open-source datasets (Breast cancer, Indian liver, NSCLC-Radiomics dataset, and Stage III NSCLC).

FINDINGS

The proposed framework ensured a comparable predictive performance against a centralized learning approach. Pairwise DeLong tests showed no significant difference between the compared pairs for each dataset.

INTERPRETATION

Distributed learning contributes to preserve medical data privacy. We foresee this technology will increase the number of collaborative opportunities to develop robust AI, becoming the default solution in scenarios where collecting enough data from a single reliable source is logistically impossible. Distributed sequential learning provides privacy persevering means for institutions with small but clinically valuable datasets to collaboratively train predictive AI while preserving the privacy of their patients. Such models perform similarly to models that are built on a larger central dataset.