A multidisciplinary research team has developed a method to monitor the progression of movement disorders using motion capture technology and artificial intelligence.
In two groundbreaking studies published in the journal Nature Medicine, an interdisciplinary team of artificial intelligence and clinical researchers has shown that by combining human movement data collected by wearable technology with powerful new medical artificial intelligence techniques, Combined, they were able to identify clear patterns of movement, predict future disease progression and dramatically increase the efficiency of clinical trials of two very different rare diseases, Duchenne muscular dystrophy (DMD) and Friedreich’s ataxia ( FA).
DMD and FA are rare degenerative genetic diseases that affect movement and eventually lead to paralysis. Neither disease is currently curable, but the researchers hope the results will significantly speed up the search for new treatments.
Tracking the progression of FA and DMD is usually done through intensive testing in a clinical setting. These papers provide a more precise assessment while also improving the accuracy and objectivity of the data collected.
The researchers estimate that using these disease markers would mean that far fewer patients would be needed to develop new drugs than current approaches. This is especially important for rare diseases where identifying suitable patients is difficult.
The scientists hope that, in addition to using the technology to monitor patients in clinical trials, it could one day be used to monitor or diagnose a range of common conditions that affect motor behaviour, such as dementia, stroke and orthopedic disorders.
The senior author and corresponding author of both papers is Professor Aldo Faisal from the Department of Bioengineering and Computing at Imperial College London, who is also Director of the UKRI Center for Doctoral Training in Medical Artificial Intelligence, and the University of Bayreuth Chair in Digital Health (Germany) and A UKRI Turing AI fellowship winner said: “Our method collects a huge amount of data from a person’s full-body movement – more than any neuroscientist has observed in patients with precision or time. Our “AI technology constructs a digital twin of a patient, allowing us to make unprecedentedly accurate predictions of an individual patient’s disease progression. We believe that the same AI technology works in two very different diseases, showing just how powerful it can be.” Hopefully it will apply to many diseases and help us develop treatments for many more diseases faster, cheaper and more precisely.”
The two papers highlight the extensive collaboration of researchers and experts in AI technology, engineering, genetics, and clinical specialties. These include the Department of Bioengineering and Computing at Imperial College, MRC London Institute of Medical Sciences (MRC LMS), UKRI Center for AI in Healthcare, UCL Great Ormond Street Institute of Child Health (UCL GOS ICH), NIHR Great Researchers Ormond Street Hospital Biomedical Research Center (NIHR GOSH BRC), Imperial College London, UCL Queen Square Institute of Neurology Ataxia Centre, Great Ormond Street Hospital, National Hospital for Neurology and Neurosurgery , National Hospital of Neurology and Neurosurgery (UCLH and UCL/UCL BRC), University of Bayreuth, Germany, and Gemelli Hospital, Rome, Italy.
Motion Fingerprint – Detailed Test
In the DMD-focused study, researchers and clinicians at Imperial College London, Great Ormond Street Hospital and University College London tested the drug in 21 children with DMD and 17 age-matched healthy controls Wearable sensor kit. Children wear the sensors while taking standard clinical assessments, such as the 6-minute walk test, and performing everyday activities, such as eating lunch or playing.
In the FA study, teams from the Ataxia Center at Imperial College London and UCL Queen Square Institute of Neurology worked with patients to identify key movement patterns and predict genetic markers for the disease. FA, the most common inherited ataxia, is caused by an abnormally large triple repeat of DNA that turns off the FA gene. Using this new AI technique, the team was able to use motion data to accurately predict the “off” of the FA gene, measuring how active it was, without taking any biological samples from the patient.
The team was able to administer a rating scale to determine the degree of disability in ataxia SARA and functional assessments such as ambulation, hand/arm movement (SCAFI) in nine FA patients and matched controls. The results of these validated clinical assessments are then compared with those obtained by using the new technique on the same patients and controls. The latter showed higher sensitivity in predicting disease progression.
In both studies, all data from sensors was collected and fed into artificial intelligence technology to create personal avatars and analyze movements. This large dataset and powerful computational tools allowed the researchers to define key motor fingerprints seen in children with DMD as well as adults with FA that were different in controls. Many of these AI-based motor patterns have not previously been described clinically in DMD or FA.
The scientists also found that the new AI technique could also significantly improve predictions of how an individual patient’s disease will progress over a six-month period, compared with the current gold-standard assessment. Such precise predictions allow for more efficient clinical trials, giving patients faster access to new treatments, and can also aid in more precise drug delivery.
The number of future clinical trials is small
This new method of analyzing whole-body motion measurements provides clinical teams with clear disease markers and progression predictions. These are valuable tools for measuring the benefit of new treatments during clinical trials.
The new technique could help researchers conduct clinical trials of conditions that affect movement more quickly and accurately. In the DMD study, the researchers showed that the new technique could reduce the number of children needed to test whether a new treatment works to as little as a quarter of that required by current methods.
Likewise, in the FA study, the researchers showed that they could achieve the same precision with 10 patients instead of more than 160. This AI technique is especially powerful when studying rare diseases, where the number of patients is small. In addition, the technology allows the study of patients during life-altering disease events such as inability to walk, whereas current clinical trials target ambulatory or non-ambulatory patient populations.
Co-author of both studies, Professor Thomas Voit, Director of the NIHR Great Ormond Street Biomedical Research Center (NIHR GOSH BRC) and UCL GOS ICH Professor of Developmental Neuroscience, said: “These studies show how innovative technologies can significantly improve the way we live our lives every day. Research diseases. The impact of this, together with specialized clinical knowledge, will not only increase the efficiency of clinical trials, but also has the potential to translate into various conditions affecting sports. This is thanks to the cross-research collaboration between institutes, hospitals, clinical specialties and dedicated With more patients and families, we can begin to address the challenges facing rare disease research.”
Dr Balasundaram Kadirvelu, Postdoctoral Fellow in the Department of Computational and Bioengineering at Imperial College London and co-first author of both studies, said: “We were surprised to see that our AI algorithm was able to discover some new ways of analyzing human movement. We Call them “behavioral fingerprints” because, just like your finger prints allow us to identify a person, these digital fingerprints can accurately characterize disease, whether the patient is in a wheelchair or walking, being evaluated in a clinic or over a coffee restaurant for lunch”
Co-first author of the DMD study and co-author of the FA study, Dr Valeria Ricotti, Honorary Clinical Lecturer at GOS ICH at University College London, said: “The cost and logistical challenges of studying rare diseases are much greater, meaning patients miss out on Potential new treatments. Improving the efficiency of clinical trials gives us hope that we can successfully test many more treatments.”
Co-author Professor Paola Giunti, Head of the UCL Ataxia Centre, Queen Square Institute of Neurology, and Honorary Consultant at the National Hospital of Neurology and Neurosurgery, UCL Hospital, said: “We are thrilled with the results of this project, which demonstrates have shown how artificial intelligence is close to being sure to be better at capturing disease progression in rare diseases like Friedreich’s ataxia. With this new approach, we can revolutionize the design of clinical trials for new drugs and monitor them with an accuracy unknown to previous methods. effects of existing drugs.”
“As well as our significant input into the clinical programme, the large number of FA patients at the UCL Queen Square Institute of Neurology Ataxia Center are well characterized clinically and genetically, making this project possible. We also Thanks to all of our patients who participated in this project.”
Co-author of both studies, Professor Richard Festenstein, from the MRC London Institute of Medical Sciences and Imperial College London’s Department of Brain Sciences, said: “Patients and families often want to know how their disease is progressing and motion capture combined with AI can help. provide this information. We hope this study has the potential to transform clinical testing of rare movement disorders and improve the diagnosis and monitoring of patients with above-human performance levels.”
The research was funded by the UKRI Turing Fellowship in Artificial Intelligence, Professor Faisal, NIHR Imperial College Biomedical Research Center (BRC), MRC London Institute of Medical Sciences, Duchenne Research Fund, NIHR Great Ormond Street Hospital (GOSH) BRC, University College London/UCLH BRC and UK Medical Research Council.
Kadirvelu, B., and others. (2023) Wearable motion capture suit and machine learning to predict disease progression in Friedreich’s ataxia. natural medicine. doi.org/10.1038/s41591-022-02159-6.