Search Results

The ARGO ReWalk™ orthotically fits to the lower limbs and part of the upper body and is intended to enable individuals with spinal cord injury at levels T7 to L5 to perform ambulatory functions with supervision of a specially trained companion in accordance with the user assessment and training certification program. The device is also intended to enable individuals with spinal cord injury at levels T4 to T6 to perform ambulatory functions in rehabilitation institutions in accordance with the user assessment and training certification program. The ReWalk™ is not intended for sports or stair climbing.

The device is a wearable exoskeleton device (refer to Figure 2 below) that allows the user to enable ambulation over the course of the day (refer to Figure 1 below). The control of the device is achieved through a wrist-worn user-operated wireless communicator, tilt sensor and specific body movements. The movement of the swing leg is controlled by a set of gears and DC motors at the knee and hip joints.

Here's a breakdown of the acceptance criteria and the studies that prove REWALKTM meets them, based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria for the ReWalk™ device are not explicitly stated as quantitative performance metrics in a single table, but rather as a combination of safety, functionality, and performance demonstrations across various studies. The reported device performance is extracted from the clinical study results.

2. Sample Sizes and Data Provenance

• Study 1 (Pilot Study):
  • Test Set Sample Size: 7 subjects enrolled. 6 subjects completed the 6MWT and 10MWT.
  • Data Provenance: Not explicitly stated, but implies a prospective clinical trial in a single center given the "pilot study" designation.
• Study 2 (Multi-Center Study):
  • Test Set Sample Size: 24 subjects enrolled. Results provided for 20 subjects for 6MWT, 22 subjects for 10MWT, and 13 subjects for Ashworth scale.
  • Data Provenance: Prospective, multi-center study conducted at two sites: Moss Rehab and Centro Villa Beretta.
• Study 3 (Exoskeletal-Assisted Walking for Persons with Motor-Complete Paraplegia):
  • Test Set Sample Size: A convenience sample of 12 potential participants screened, 7 included in the results (5 were excluded).
  • Data Provenance: Pilot study (single group, pre/post intervention), likely prospective, but not explicitly stated. A footnote indicates the source as "Force Sustainment: Rehabilitation, Regeneraton and Prosthetics for Reintegration to duty. Meeting Proceedings STO-MP-HFM-228, Paper 6." It also mentions "Additional retrospective data from this study was provided in a separate master file."

3. Number of Experts and Qualifications for Ground Truth

The concept of "ground truth" as typically used in AI/medical device contexts (e.g., expert consensus on image interpretation) is not directly applicable here. This device's performance is measured directly by patient physical performance metrics and observed adverse events. There were no "experts" establishing ground truth in the sense of reviewing or labeling a dataset for algorithmic training or testing. The outcomes (e.g., distance walked, time taken, presence of skin lesions) are direct measurements from the subjects using the device.

4. Adjudication Method for the Test Set

Not applicable. The clinical studies measured direct physical performance (e.g., 6MWT, 10MWT) and observed adverse events. There was no need for expert adjudication of case interpretations.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

No. An MRMC comparative effectiveness study was not conducted for the ReWalk™ device. This type of study is typically used for diagnostic devices where human readers interpret medical images or data, and the AI's impact on their performance is being evaluated. The ReWalk™ is a therapeutic/assistive device, and its effectiveness is measured more directly by functional outcomes. There is no mention of "human readers" or "AI assistance" in this context.

6. Standalone (Algorithm Only) Performance Study

No, a standalone (algorithm only) performance study was not explicitly mentioned in the sense of the device operating without a human user. The device inherently requires a human user-in-the-loop for its operation (e.g., wearing the device, using crutches, operating the wrist communicator, providing body movements for walking). The "device performance" relies on the human-device interaction.

However, extensive non-clinical/bench studies were performed that tested the device in a standalone manner for its mechanical integrity, electrical safety, durability, and software functionality (e.g., cyclic testing of joints, battery testing, EMC/EMI testing, software verification/validation). These tests ensure the components and underlying algorithms perform as intended and safely, independent of a human user, but the overall "performance of ambulation" requires a human.

7. Type of Ground Truth Used

The ground truth used for these studies was primarily:

• Objective Physiological Measurements: Distances walked (6MWT, 10MWT), time taken for 10MWT.
• Clinical Assessments: Ashworth scale for spasticity, Visual Analog Scale (VAS) for pain and fatigue (results not extensively detailed in the provided text).
• Observable Adverse Events: Falls, hematomas, skin lesions, bruising, blisters, edema.
• Direct Observation of Functional Skills: Standing, walking, sit-to-stand, stand-to-sit, and maneuvering on various surfaces.

8. Sample Size for the Training Set

There is no mention of a separate "training set" in the context of an AI/machine learning model whose performance on a "test set" is being evaluated. The subjects in the clinical studies (1, 2, and 3) were participants who received training on how to use the device. Their performance after training was then measured. The term "training set" in this document refers to the population of subjects who underwent the manufacturer's training program to learn how to operate the ReWalk™ device, not a data set for machine learning.

• Study 1: 7 subjects
• Study 2: 24 subjects
• Study 3: 7 subjects

9. How Ground Truth for the Training Set Was Established

Given that "training set" here refers to human subjects undergoing training rather than an algorithm's dataset, the concept of "ground truth for the training set" doesn't apply in the typical sense of labeled data. Instead, the training program itself defines what constitutes correct and proficient use. The "ground truth" for successful training was established through:

• Manufacturer-Developed Training Program: The sponsor proposed a tiered training system (Table 7) with defined basic and advanced skills (Tables 8 & 9) and companion skills (Tables 12 & 13).
• Scoring Metrics: Proficiency was assessed using scoring metrics, including a FIM (Functional Independence Measure) based scoring system (Tables 10 & 11) for individual skills, where a "Pass" score (5 or greater) indicated successful completion with supervision or independently. For companion skills, a score of 2 indicated full understanding.

These scoring criteria and the structured program set the "ground truth" for what constitutes successful human user training and proficiency with the device.

Ask a specific question about this device