Bayesian Deep Learning for Motor Imagery Classification

Status

Active, not recruiting

Phases

Phase 1

Study type

Interventional

Source

TCTR

Registry ID

TCTR20211027006

Enrollment

Registered

2021-10-27

Start date

2021-11-08

Completion date

Unknown

Last updated

2026-08-03

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

The objective of this project is to develop Motor Imagery (MI) classification algorithm that yields high accuracy and also results in Subject-Independent and Session-Independent characteristic which are desired properties in EEG-based BCI. These characteristics make EEG-based BCI practical to use due to it takes less time to collect EEG data. The proposed classifier is Bayesian Deep Learning which uses Bayesian principle for Deep Learning. Therefore, Bayesian Deep Learning is the classifier that

Interventions

Bayesian deep learning during motor imagery with brain-computer-interface,Common spatial pattern (CSP) during motor imagery with brain-computer-interface

Active Comparator Device,Active Comparator Device

Motor imagery Bayersian,Motor imagery CSP

Eligibility

Sex/Gender

All

Age

60 Years to 80 Years

Inclusion criteria

Inclusion criteria: 1. Active ageing 2. No weakness or numbness of upper limb. 3. Sit upright with the back of a chair for 60 minutes or more. 4. Agree to participate in the trial

Exclusion criteria

Exclusion criteria: 1. Have a history of diseases that weaken the arm, such as a stroke . 2. Have shoulder and elbow articular attachment. The arm can be raised less than 90 degrees forward, and the elbow is less than -20 degrees, and the elbow is flexed less than 130 degrees. 3. Have communication or intellectual problems that they are unable to follow or cooperate in the experiment . 4. Have shoulder or elbow pain with a visual analog scale of 4 or greater in the tested arm. 5. Vision after editing could not be clearly seen the screen in the experiment at a distance of approximately 1 meter.

Design outcomes

Primary

Measure	Time frame
classification accuracy after 4th session classification accuracy	—

Secondary

Measure	Time frame
classification accuracy after 4th session classification accuracy	—

Countries

Thailand

Contacts

Public ContactArpa Suwannarat

National Electronics and Computer Technology Center, 112 Thailand Science Park, Phahonyothin Road, Khlong Nueng, Khlong Luang, Pathumthani 12120, Thailand

arpa.suw@nectec.or.th0895151552

Outcome results

None listed

Source: TCTR (via WHO ICTRP) · Data processed: Aug 9, 2026