Wearable activity trackers and artificial intelligence in the management of rheumatic diseases

Davergne, Thomas; Kedra, Joanna; Gossec, Laure

doi:10.1007/s00393-021-01100-5

Cited by 16 publications

(9 citation statements)

References 29 publications

(38 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…As ML methods used to date are mostly "supervised", large datasets are needed to train the models adequately. Additionally, the accuracy of these data must be guaranteed, as poor quality data could lead to erroneous results and therefore, to erroneous conclusions; nevertheless, implementing data quality control can be time consuming, expensive, and laborious (88). Moreover, standardization of data (especially imaging data) acquisition is still challenging.…”

Section: -Limitations Of Machine Learning Methodsmentioning

confidence: 99%

“…Technical issues are related to the fact that huge amounts of validated data are required to properly train "supervised" ML modelswhich represent most ML methods used nowadays (88). Indeed, depending on the physician or research team, disease parameters are not assessed in the same way (e.g., disease activity can be assessed in 28 or 44 joints, or using erythrocyte sedimentation rate or C reactive protein), which results in heterogeneous data.…”

Section: Expert Opinionmentioning

confidence: 99%

“…Methodological issues are mostly related to the fact that, although providing interesting results on specific datasets, only a few ML models are applicable to other datasets (88). As a matter of fact, most ML models presented previously have good predictive performances, but this could reflect "overfitting" of the models to the dataset they were trained on.…”

Section: Expert Opinionmentioning

confidence: 99%

See 2 more Smart Citations

Machine learning approaches to improve disease management of patients with rheumatoid arthritis: review and future directions

Kedra

Davergne

Braithwaite³

et al. 2021

Expert Review of Clinical Immunology

Self Cite

View full text Add to dashboard Cite

Introduction: Although the management of rheumatoid arthritis (RA) has improved in major way over the last decades, this disease still leads to an important burden for patients and society, and there is a need to develop more personalized approaches. Machine learning (ML) methods are more and more used in health-related studies and can be applied to different sorts of data (clinical, radiological, or "omics" data). Such approaches may improve the management of patients with RA. Areas covered: In this paper, we propose a review regarding ML approaches applied to RA. A scoping literature search was performed in PubMed, in September 2021 using the following MeSH terms: "arthritis, rheumatoid" and "machine learning". Based on this search, the usefulness of ML methods for RA diagnosis, monitoring and prediction of response to treatment and RA outcomes, is discussed. Expert opinion: ML methods have the potential to revolutionize RA-related research and improve disease management and patient care. Nevertheless, these models are not yet ready to contribute fully to rheumatologists' daily practice. Indeed, these methods raise technical, methodological, and ethical issues, which should be addressed properly to allow their implementation. Collaboration between data scientists, clinical researchers and physicians is therefore required to move this field forward.

show abstract

Section: -Limitations Of Machine Learning Methodsmentioning

confidence: 99%

Section: Expert Opinionmentioning

confidence: 99%

Section: Expert Opinionmentioning

confidence: 99%

See 1 more Smart Citation

Machine learning approaches to improve disease management of patients with rheumatoid arthritis: review and future directions

Kedra

Davergne

Braithwaite³

et al. 2021

Expert Review of Clinical Immunology

Self Cite

View full text Add to dashboard Cite

show abstract

“…For participants with complete data and a body-worn smartphone, this confidence interval would be much narrower than for participants with high amounts of missing data, or who had their phone lying on a desk. If accuracy and validity is insufficient, researchers should consider using body-worn devices, such as wearable activity trackers [ 101 ] or smartwatches [ 22 ].…”

Section: Main Textmentioning

confidence: 99%

Smartphones for musculoskeletal research – hype or hope? Lessons from a decennium of mHealth studies

Beukenhorst

Druce

Cock

2022

BMC Musculoskelet Disord

View full text Add to dashboard Cite

Background Smartphones provide opportunities for musculoskeletal research: they are integrated in participants’ daily lives and can be used to collect patient-reported outcomes as well as sensor data from large groups of people. As the field of research with smartphones and smartwatches matures, it has transpired that some of the advantages of this modern technology are in fact double-edged swords. Body In this narrative review, we illustrate the advantages of using smartphones for data collection with 18 studies from various musculoskeletal domains. We critically appraised existing literature, debunking some myths around the advantages of smartphones: the myth that smartphone studies automatically enable high engagement, that they reach more representative samples, that they cost little, and that sensor data is objective. We provide a nuanced view of evidence in these areas and discuss strategies to increase engagement, to reach representative samples, to reduce costs and to avoid potential sources of subjectivity in analysing sensor data. Conclusion If smartphone studies are designed without awareness of the challenges inherent to smartphone use, they may fail or may provide biased results. Keeping participants of smartphone studies engaged longitudinally is a major challenge. Based on prior research, we provide 6 actions by researchers to increase engagement. Smartphone studies often have participants that are younger, have higher incomes and high digital literacy. We provide advice for reaching more representative participant groups, and for ensuring that study conclusions are not plagued by bias resulting from unrepresentative sampling. Costs associated with app development and testing, data storage and analysis, and tech support are substantial, even if studies use a ‘bring your own device’-policy. Exchange of information on costs, collective app development and usage of open-source tools would help the musculoskeletal community reduce costs of smartphone studies. In general, transparency and wider adoption of best practices would help bringing smartphone studies to the next level. Then, the community can focus on specific challenges of smartphones in musculoskeletal contexts, such as symptom-related barriers to using smartphones for research, validating algorithms in patient populations with reduced functional ability, digitising validated questionnaires, and methods to reliably quantify pain, quality of life and fatigue.

show abstract

“…Particular interest was also given to the use of ML solutions for osteoporosis in [35]. The applicability of AI in the management of RMDs, with data collected from wearable activity trackers, was examined in [36].…”

Section: Past Reviewsmentioning

confidence: 99%

Understanding the role and adoption of artificial intelligence techniques in rheumatology research: an in-depth review of the literature

Madrid

Merino‐Barbancho

Rodríguez-González

et al. 2022

Preprint

View full text Add to dashboard Cite

The outstanding and upward trend in the number of published research related to rheumatic and musculoskeletal diseases, in which artificial intelligence plays a key role, has exhibited the interest of rheumatology researchers in using these techniques to answer their research questions. In this review, we analyse the original research articles that combine both worlds in a five-year period (2017-2021). In contrast to other published papers on the same topic, we first studied the review and recommendation articles that were published during that period, including up to October 2022, as well as the publication trends. Secondly, we review the published research articles and classify them into one of the following categories: disease classification, disease prediction, predictors identification, patient stratification and disease subtype identification, disease progression and activity, and treatment response. Thirdly, we provide a table with illustrative studies in which artificial intelligence techniques have played a central role in more than twenty rheumatic and musculoskeletal diseases. Finally, the findings of the research articles, in terms of disease and/or data science techniques employed are highlighted in a discussion. Therefore, the present review arises with the aim of characterising how researchers are employing data science techniques in the rheumatology medical field.

show abstract

Wearable activity trackers and artificial intelligence in the management of rheumatic diseases

Cited by 16 publications

References 29 publications

Machine learning approaches to improve disease management of patients with rheumatoid arthritis: review and future directions

Machine learning approaches to improve disease management of patients with rheumatoid arthritis: review and future directions

Smartphones for musculoskeletal research – hype or hope? Lessons from a decennium of mHealth studies

Understanding the role and adoption of artificial intelligence techniques in rheumatology research: an in-depth review of the literature

Contact Info

Product

Resources

About