Department of Biomedical Engineering, University of Michigan, Ann Arbor, MI, USA.
John A. Paulson School of Engineering and Applied Sciences, Harvard University, Boston, MA, USA.
J Neuroeng Rehabil. 2024 Feb 8;21(1):21. doi: 10.1186/s12984-024-01307-y.
Lack of standardized assessments that explicitly quantify performance during prosthetic grip selection poses difficulty determining whether efforts to improve the design of multi-grip hands and their control approaches are successful. In this study, we developed and validated a novel assessment of multi-grip prosthetic performance: The Coffee Task.
Individuals without limb loss completed the Box and Block Test and two versions of the Coffee Task - Continuous and Segmented - with a myoelectric prosthetic emulator. On different days, participants selected prosthetic grips using pattern recognition and trigger control. Outcomes of the Continuous and Segmented Coffee Task were completion time and number of errors, respectively. Two independent raters assessed outcomes of the Coffee Task using video recordings to determine inter-rater reliability. Known-group validity was assessed by comparing outcomes with the emulator to those with an intact limb. Convergent validity was assessed through the correlation of the Coffee Task outcomes and those of the Box and Blocks Test. Responsiveness to changes with practice and control approach were assessed using the standardized response mean (SRM).
Inter-rater reliability was high for both versions of the Coffee Task (Intra-class coefficient > 0.981). Coffee Task outcomes were moderately correlated with the Box and Blocks outcomes (|r| ≥ 0.412, p ≤ 0.007). Participants completed the Coffee Task faster with their intact limb than with the emulator (p < 0.001). Both versions of the Coffee Task were responsive to changes with training (SRM ≥ 0.81) but not control approach (SRM ≤ 0.12).
The Coffee Task is reliable, has good known-group and convergent validity, and is responsive to changes due to practice. Future work should assess whether the Coffee Task is feasible and reliable for people with upper limb loss who use multi-grip prostheses.
缺乏明确量化假体抓握选择过程中表现的标准化评估方法，使得难以确定改进多指手设计及其控制方法的努力是否成功。在这项研究中，我们开发并验证了一种新的多抓握假肢性能评估方法：咖啡任务。
无肢体缺失的个体使用肌电假肢模拟器完成了 Box 和 Block 测试以及咖啡任务的两种变体（连续型和分段型）。在不同的日子里，参与者使用模式识别和触发控制选择假肢夹具。连续型和分段型咖啡任务的结果分别是完成时间和错误数。两名独立的评估者使用视频记录评估咖啡任务的结果，以确定评分者间的可靠性。通过将模拟器与完整肢体的结果进行比较来评估已知组的有效性。通过咖啡任务结果与 Box 和 Blocks 测试结果的相关性评估收敛有效性。使用标准化反应均值（SRM）评估随练习和控制方法变化的反应性。
两种咖啡任务版本的评分者间可靠性都很高（组内相关系数>0.981）。咖啡任务的结果与 Box 和 Blocks 测试的结果中度相关（|r|≥0.412，p≤0.007）。参与者使用完整肢体完成咖啡任务的速度比使用模拟器快（p<0.001）。两种咖啡任务版本都对训练变化有反应（SRM≥0.81），但对控制方法变化无反应（SRM≤0.12）。
咖啡任务可靠，具有良好的已知组和收敛有效性，并且对因练习而产生的变化敏感。未来的研究应该评估咖啡任务是否对使用多指假肢的上肢缺失人群可行和可靠。
