Construction of A Multimodal Digital Assessment Model for Myasthenia Gravis
试验速览
- 阶段
- 不适用
- 状态
- 招募中
- 入组人数
- 180
- 试验地点
- 1
- 主要终点
- Descriptive Analysis and Comparison of Digital Phenotypic Data across Subgroups
研究概览
简要总结
This research is a single-center, exploratory, observational study to be carried out in the outpatient or inpatient ward of the Neurology Department at Huashan Hospital, affiliated to Fudan University. The aim is to develop a digital assessment model for Myasthenia Gravis by gathering multimodal digital phenotypic data from MG patients. This includes physiological signals, facial videos, eye movements, speech, limb movements, various scales, and quality of life metrics.
详细描述
The goal is to define the multimodal digital phenotypes of myasthenia gravis patients, determine the specificities of their symptoms, and develop a digital evaluation model and remote assessment system that is objective, precise, and user-friendly. This will provide a scientific foundation and technical support for diagnosing, treating, and rehabilitating individuals with MG.
Key issues to be addressed include:
- Whether digital phenotyping can comprehensively represent the disease characteristics and severity gradations in MG.
- The feasibility of using multimodal digital phenotypic modeling for the objective evaluation of MG.
- The challenges and obstacles faced by the AI-enhanced medical model in clinical demonstration applications.
The study is focused on three main objectives:
- Perform a descriptive analysis and comparison of phenotypic data across various subgroups to pinpoint key characteristics associated with MG disease grade and scale score.
- Construct a digital evaluation model for MG patients using chosen features, validate the model with prospectively gathered data, and conduct a correlative analysis with the clinical functional scales to assess its effectiveness on predicting MG symptom grading and disease progression.
- Develop a patient-centric remote evaluation system utilizing the refined MG digital evaluation model to facilitate its application in real-world clinical settings.
研究设计
- 研究类型
- Observational
- 观察模型
- Cohort
- 时间视角
- Prospective
入排标准
- 年龄范围
- 18 Years 至 —（Adult, Older Adult）
- 性别
- All
- 接受健康志愿者
- 是
入选标准
- •For Patients with MG
- •Patients have a confirmed diagnosis of myasthenia gravis and be over 18 years old.
- •Patients have the clinical classification of Myasthenia Gravis Foundation of America (MGFA) within I-IV, or be asymptomatic after treatment.
- •Patients must sign the informed consent form and the privacy confidentiality agreement.
- •For Healthy Participants
- •Age group matched with MG participants
排除标准
- •For Patients with MG
- •Patients are in the crisis stage of myasthenia gravis (MGFA class V) and unable to cooperate with scoring.
- •Patients are with severe cardiopulmonary diseases and unable to cooperate with the scoring.
- •Patients with any psychiatric disorder or cognitive dysfunction that, in the investigator's judgment, may interfere with their participation in the study.
- •Patients with any other unspecified unstable medical condition.
结局指标
主要结局
Descriptive Analysis and Comparison of Digital Phenotypic Data across Subgroups
时间窗: At baseline (single study visit)
Quantitative descriptive analysis of multimodal digital phenotypic data (motor performance, ocular metrics, and speech-derived features) to compare subgroup-specific patterns among patients with myasthenia gravis. Metrics will include summary statistics (mean, standard deviation, distribution profiles) for each modality, with subgroup comparisons performed to assess variability.
Correlation Between Digital Evaluation Model and Quantitative Myasthenia Gravis (QMG) Scale
时间窗: At baseline (single study visit)
This outcome measure will assess the convergent validity of the digital evaluation model by calculating the correlation coefficient (Pearson's or Spearman's) between the model-derived composite score and the Quantitative Myasthenia Gravis (QMG) clinical scale score in patients with myasthenia gravis.
次要结局
- Interclass Correlation Coefficient (ICC) of the Digital Outcome Assessment Model(Baseline and Week 2)
- Prospectively validate the model effectiveness(1 year)