Implementing Machine Learning Into Optical Pedography: Development and Validation of a Novel Instrument: a Validation Study
Trial Snapshot
- Phase
- Not Applicable
- Status
- Recruiting
- Sponsor
- Palacky University
- Enrollment
- 150
- Locations
- 2
- Primary Endpoint
- Criterion validity of foot area measurement
Study Overview
Brief Summary
This study aims to evaluate the validity and reliability of a proposed plantar pressure assessment intrument based on an implementation of machine learning into optical pedography.
Detailed Description
This study aims to evaluate the validity and reliability of a proposed plantar pressure assessment intrument based on an implementation of machine learning into optical pedography. The whole process will be divided into distinct steps required for the targeted outcome, which includes:
- collecting visual data (podoscope foot pictures) and training a segmentation machine learning-based algorithm designed for recognizing only feet area, a total of atleast 30 participants performing 9 different standing positions - over 270 usable pictures for training and functionality validation
- collecting personal, visual and pressure data (participant weight, podoscope foot pictures, pedobarographic platform measurements) and training a machine/deep learning-based model designed for feet pressure distribution areas identification and quantification, a total of estimated 60 participants undergoing 5 alternating, as similar as possible, measurements on podoscope and pedobarographic platform
- evaluating the validity and reliability of a new plantar pressure measuring instrument following the same imaging procedure as described in step 2, a total of estimated 60 participants
Study Design
- Study Type
- Observational
- Observational Model
- Other
- Time Perspective
- Prospective
Eligibility Criteria
- Ages
- 18 Years to — (Adult, Older Adult)
- Sex
- All
- Accepts Healthy Volunteers
- Yes
Inclusion Criteria
- •A healthy individual with no absence of lower extremity
- •Aged ≥18 years
- •Willingness to participate and ability to follow the assessors instructions.
Exclusion Criteria
- •Presence of foot diseases alternating foot contact area
- •Cognitive or psychiatric disorders limiting cooperation
- •Lack of informed consent or non-compliance during imaging
Outcomes
Primary Outcomes
Criterion validity of foot area measurement
Time Frame: Day 1
The total contact area of the foot during a static stance will be measured by the machine learning based optical pedography Instrument and compared to the gold standard pedobarographic platform. The validity will be determined by the Pearson Correlation Coefficient (r) between the two devices. Unit of measure: Pearson Correlation Coefficient (r) ranging from -1 to 1.
Criterion validity of peak plantar pressure distribution
Time Frame: Day 1
The distribution of pressure across the plantar surface (specifically mean peak pressure) will be measured. Validity will be assessed by calculating the Intraclass Correlation Coefficient (ICC) between the machine learning based optical pedography instrument and the gold standard pedobarographic platform. Unit of measure: Intraclass Correlation Coefficient (ICC) ranging from 0 to 1.
Intrasession reliability of foot area measurement
Time Frame: Day 1
The consistency of the total contact area (cm 2) measured across 5 repeated trials within a single session. Reliability will be assessed using the Intraclass Correlation Coefficient (ICC 3,5) to determine the degree of agreement between the five captures of the same participant's feet. Units of measure: Intraclass Correlation Coefficient (ICC) ranging from 0 to 1
Intrasession reliability of mean peak pressure
Time Frame: Day 1
The consistency of mean peak pressure measurements across 5 repeated trials within a single session. Reliability will be assessed by calculating the Intraclass Correlation Coefficient (ICC 3,5) for the five measurements taken on the machine learning based optical pedography instrument. Unit of measure: Intraclass Correlation Coefficient (ICC) ranging from 0 to 1
Secondary Outcomes
No secondary outcomes reported
Investigators
Jonatan Dvoracek
PhD student
Palacky University