A wearable motion capture suit and machine learning predict disease progression in Friedreich’s ataxia

Kadirvelu, Balasundaram; Gavriel, Constantinos; Nageshwaran, Sathiji; Chan, Jackson Ping Kei; Nethisinghe, Suran; Athanasopoulos, Stavros; Ricotti, Valeria; Voït, Thomas; Giunti, Paola; Festenstein, Richard; Faisal, A. Aldo

doi:10.1038/s41591-022-02159-6

Cited by 36 publications

(18 citation statements)

References 62 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recent studies have yielded encouraging outcomes by using wearable sensors to quantify clinical evaluations such as SARA and mFARS, aiming to mitigate the inaccuracies associated with subjective clinical assessments [23][24][25][26][27][28]. However, these approaches still require supervised assessments, which is only feasible within clinical settings.…”

Section: Discussionmentioning

confidence: 99%

At-home Wearable-based Monitoring and Machine Learning Predicts Clinical Measures and Biological Biomarkers of Disease Severity in Friedreich's Ataxia

Vaziri,

Mishra,

Nunes

et al. 2024

Preprint

View full text Add to dashboard Cite

Friedreich ataxia (FRDA) results in progressive impairment in gait, upper extremity coordination, and speech. Currently, these symptoms are assessed through expert examination at clinical visits. Such in-clinic assessments are time-consuming, subjective, of limited sensitivity, and provide only a limited perspective of the daily disability of patients. In this study, we recruited 39 FRDA patients and remotely monitored their physical activity and upper extremity function using a set of wearable sensors for 7 consecutive days. We compared the sensor-derived metrics of lower and upper extremity function as measured during activities of daily living with FDRA clinical measures (e.g., mFARS and FA-ADL) and biological biomarkers of disease severity (guanine-adenine-adenine (GAA) and frataxin (FXN) levels). The results showed significant correlations with moderate to high effect size between multiple sensor-derived metrics and the FRDA clinical and biological outcomes. Finally, we developed multiple machine learning-based models to predict disease severity in FRDA using demographic, biological, and sensor-derived metrics. When sensor-derived metrics were included, the model performance enhanced 1.5-fold and 2-fold in terms of coefficient of determination for predicting FRDA clinical measures and biological biomarkers of disease severity, respectively. Our results signify the potential of at-home remote monitoring in assessing disease severity and monitoring motor dysfunction in FRDA.

show abstract

Section: Discussionmentioning

confidence: 99%

At-home Wearable-based Monitoring and Machine Learning Predicts Clinical Measures and Biological Biomarkers of Disease Severity in Friedreich's Ataxia

Vaziri,

Mishra,

Nunes

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…The solution resides in a clear association with a hard disease milestone or a clinically significant event, but this requires large amounts of data of high quality. Recent papers in DMD [ 14 ] and Friedreich’s ataxia [ 133 ] showed the feasibility and the potential of machine learning, but these proof of mechanism study were conducted using different sensors in a controlled setting [ 14 ].…”

Section: Discussionmentioning

confidence: 99%

The use of digital outcome measures in clinical trials in rare neurological diseases: a systematic literature review

Poleur

Markati

Servais

2023

Orphanet J Rare Dis

View full text Add to dashboard Cite

Developing drugs for rare diseases is challenging, and the precision and objectivity of outcome measures is critical to this process. In recent years, a number of technologies have increasingly been used for remote monitoring of patient health. We report a systematic literature review that aims to summarize the current state of progress with regard to the use of digital outcome measures for real-life motor function assessment of patients with rare neurological diseases. Our search of published literature identified 3826 records, of which 139 were included across 27 different diseases. This review shows that use of digital outcome measures for motor function outside a clinical setting is feasible and employed in a broad range of diseases, although we found few outcome measures that have been robustly validated and adopted as endpoints in clinical trials. Future research should focus on validation of devices, variables, and algorithms to allow for regulatory qualification and widespread adoption.

show abstract

“…Machine learning is superior to traditional statistical models, and can develop predictive models for associations between high-throughput sequencing results and features [8][9][10]. It is becoming an integral part of modern data mining and clinical diagnosis [11][12][13].…”

Section: Introductionmentioning

confidence: 99%

IL18 in acute pancreatitis: Machine learning and two-sample Mendelian randomization study

Zhou,

Liu,

Bao

et al. 2024

Preprint

View full text Add to dashboard Cite

Objective Acute pancreatitis is a common disease whose treatment is limited to symptomatic support, thus finding effective biomarkers is of great significance for early diagnosis and therapy. Methods Bioinformatics and machine learning were applied to evaluate the expression, clinical features, biological function and immunological effects of the characteristic genes in AP. Meanwhile, AP mice models were constructed to verify the results in vivo. Finally, Mendelian randomization studies were performed to determine the causal relationship between IL-18 and AP through genome-wide association studies. Results A total of 100 core genes were obtained via differential analysis and PPI interaction network. IL18 was identified as the characteristic gene for AP by machine learning through three algorithmic. The expression of IL18 was increased significantly in AP (P < 0.001). The AUC value of IL18 in the diagnosis of AP was 0.917, exhibiting high clinical value. Moreover, IL18 was associated with various immune cells involved in the progression of AP. Through inverse variance weighting (IVW), we found that the OR for IL18 and AP was 0.908 (95%CI = 0.843–0.978, p = 0.011). Conclusions IL18 is a pivotal biomarker predicting the clinical prognosis and immune response in AP, which is proved to serve as a protective factor.

show abstract

A wearable motion capture suit and machine learning predict disease progression in Friedreich’s ataxia

Cited by 36 publications

References 62 publications

At-home Wearable-based Monitoring and Machine Learning Predicts Clinical Measures and Biological Biomarkers of Disease Severity in Friedreich's Ataxia

At-home Wearable-based Monitoring and Machine Learning Predicts Clinical Measures and Biological Biomarkers of Disease Severity in Friedreich's Ataxia

The use of digital outcome measures in clinical trials in rare neurological diseases: a systematic literature review

IL18 in acute pancreatitis: Machine learning and two-sample Mendelian randomization study

Contact Info

Product

Resources

About