Capturing Key MG-symptoms Using Smartphone Recordings.
Trial statusRecruiting
Trial phaseNot listed
Trial typeObservational
Biological sexAll
Age18+
SponsorLeiden University Medical Center
About this trial
This study will make use of a cross-sectional design of MG patients and non-MG participants to quantitatively assess key MG symptoms, and to explore the applicability of machine learning algorithms to their measurement.
Eligibility criteria
This trial accepts healthy volunteersQualifiers
Age ≥ 18 years
Ability to understand the requirements of the study and provide written informed consent.
A clinical diagnosis of myasthenia gravis (ocular or generalized) as defined by the Dutch national guideline (category "definite" or "probable" MG).
MGFA Clinical Classification of disease severity I-IV.
Disqualifiers
Not willing to be audio-recorded for the study assessments.
Not willing to be video-recorded for the study assessments.
Subjects currently taking part in a clinical trial of an Investigational Medicinal Product.
Subjects who have used an immediate release pyridostigmine-based medication in the 12 hours prior to their participation and participants on prolonged release pyridostigmine.
Trial population
We aim to include 150 patients with Myasthenia Gravis, recruited from the national Dutch-Belgia MG registry, the patient organisation 'Spierziekten Nederland' or identified by the clinical team of Leiden University Medical Center (LUMC). Non-MG participants are preferentially sampled from friends and family of enrolled MG patients. The highest preference will be same-household family members, lowest preference will be non-household non-family members. This acts as both a natural source of control participants and a natural control for any environmental confounders. Furthermore, non-MG participants will be recruited separately by placing flyers in crowded areas in the LUMC.
Trial design
Design model
Case-control
Time perspective
Cross-sectional
Treatments tested in this trial
Not listed
Trial groups
225 Participants
are grouped into 2 trial groupsGroup A: MG patients
Group B: Non-MG participants
Trial outcomes
Primary outcomes
1
Differentiating between MG-patients and non-MG participants using digital features of dysarhtria, dysphonia, proximal arm fatigue and ptosis.
Using machine-learning algorithms.
Time frame
Assessed at a single time point during outpatient visitSecondary outcomes
1
Correlating digital features of dysarthria, dysphonia, proximal arm fatigue and ptosis in MG patients with disease severity as measured by the MGC score.
Myasthenia Gravis Composite (MGC) is scored on an ordinal scale with four possible categories and weighted and consists of 10 items with a total score of 0-50; a higher score indicating more severe disease.
Time frame
Assessed at a single time point during outpatient visit2
Correlating digital features of dysarthria, dysphonia, proximal arm fatigue and ptosis in MG patients with the impact of MG on daily activities as measured by the MG-ADL.
MG Activities of Daily Living (MG-ADL) consists of 8 items with a total score of 0-24; a higher score indicating more severe disease. .
Time frame
Assessed at a single time point during outpatient visit3
The performance of automated signal processing of speech recordings collected through smartphone microphone for detection of dysarthria and dysphonia compared to clinical assessment.
Clinical assessment by two independent trained speech and language pathologists (SLP) using the Therapy Outcome Measure scale (TOMS) for dysarthria and GRBAS scale (Grade, Roughness, Breathiness, Asthenia and Strain) for dysphonia.
Time frame
Assessed at a single time point during outpatient visit4
The performance of automated measurement of proximal arm-fatiguing exercises through computer vision techniques applied to smartphone camera recordings for detection of proximal arm muscle weakness and fatigability compared to clinical assessment.
Clinical assessment by accredited clinicans.
Time frame
Assessed at a single time point during outpatient visitSponsors and contacts
Click on the lead sponsor to view all of their trials.
Leiden University Medical Center
Lead sponsor
Huma
Collaborator
Source ClinicalTrials.gov