Locally produced AI-designed transtibial prosthetic sockets in rural Sierra Leone – a prospective cohort study

Abstract

OBJECTIVE: To evaluate AI-based transtibial prosthetic socket shapes in a rural, low-income setting. DESIGN: This prospective cohort study evaluates AI-designed prosthetic sockets. Baseline and ten-week follow-up assessments were conducted to evaluate satisfaction, functionality, and quality of life. SETTING: The study was conducted at the 3D laboratory of Masanga Hospital in rural Sierra Leone. PARTICIPANTS: 34 participants with a transtibial amputation without a functional prosthesis were enrolled. At follow-up, 27 participants completed the assessments. INTERVENTIONS: Local coworkers with limited prosthetic experience autonomously produced transtibial sockets using user-friendly software powered by an AI algorithm trained on expert data. MAIN OUTCOME(S) AND MEASURE(S): Baseline and ten-week follow-up assessments included: (1) satisfaction with the device and service using the adapted Orthotics and Prosthetics User's Survey (OPUS); (2) functioning by quantifying prosthetic wear time and step count (before and after receiving the prosthesis) with activity sensors and the Lower Extremity Functional Status (LEFS) using the adapted OPUS survey (range: 0-60); (3) disease-specific Health-Related Quality of Life (D-HRQoL) using the adapted OPUS survey (range: 0-36) and Generic Health-Related Quality of Life (G-HRQoL) using the EQ-5D-3L and the VAS score; (4) participants' personal goals were evaluated. RESULTS: The activity sensors confirmed that all participants used the prosthesis. Median prosthetic wear time was 9.3 (1.6-11.4) hours per day, with a median of 3978 (1746-6545) steps per day, of which 39% (16-76%) were with the prosthesis. Median scores indicated a significant improvement in OPUS LEFS from 45 (39-49) to 52 (47-54) and OPUS D-HRQoL from 25 (20-30) to 31 (25-35). EQ-5D-3L and the VAS score did not significantly change; 56% achieved their desired activity goal. CONCLUSION AND RELEVANCE: This study shows that easy-to-use software and AI can support socket design in areas with limited access to skilled prosthetists.

Ähnliche Arbeiten