Search Results

The Ekso™ (version 1.1) and Ekso GT™ (version 1.2) are intended to perform ambulatory functions in rehabilitation institutions under the supervision of a trained physical therapist for the following population:
• Individuals with hemiplegia due to stroke (upper extremity motor function of at least 4/5 in at least one arm)
• Individuals with spinal cord injuries at levels T4 to L5 (upper extremity motor function of at least 4/5 in both arms)
• Individuals with spinal cord injuries at levels of C7 to T3 (ASIA D with upper extremity motor function of at least 4/5 in both arms).
The therapist must complete a training program prior to use of the device. The devices are not intended for sports or stair climbing.

Device Description

The Ekso is a powered motorized orthosis. It consists of a fitted metal brace that supports the legs, feet, and torso. It is worn via straps on the body, legs, and feet. Battery powered motors drive knee and hip joints. It has an integrated solid torso containing the computer and power supply. It has a hand-held user interface to specify settings and initiate steps. The Ekso is used with a cane, crutch, or walker.

AI/ML Overview

Here's an analysis of the acceptance criteria and supporting study for the Ekso™ (version 1.1) and Ekso GT™ (version 1.2) based on the provided document.

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly present a table of numerical acceptance criteria with corresponding device performance metrics in the way one might see for diagnostic accuracy. Instead, the acceptance criteria are implicitly defined by the safety and effectiveness demonstrated in the clinical studies, particularly for the expanded Indications for Use. The core acceptance criterion appears to be: safe and effective ambulation for the specified patient populations, demonstrated by a lack of adverse events.

Acceptance Criteria (Implicit)	Reported Device Performance
Safety: Device operation without falls, adverse events, or significant harm during ambulation.	Study 1: 50 out of 54 patients (93%) safely performed ambulatory functions with only one arm to maintain stability. No falls or other adverse events reported.
	Study 2: 7 out of 8 patients (88%) safely performed ambulatory functions with only one arm to maintain stability. No falls or other adverse events reported.
	Overall: "The lack of adverse events over 216 walking sessions in 57 patients demonstrates the safety and effectiveness of the Ekso device..."
Effectiveness (Ambulatory Function): Patients with conditions covered by the expanded indication can ambulate.	Study 1: Focused on patients with hemiplegia due to stroke, specifically those with "upper extremity motor function of less than 4/5 in one arm" (34 patients) and others not recorded (19 patients). The study demonstrated that these patients could "safely performed ambulatory functions with only one arm to maintain stability."
	Study 2: Focused on patients with hemiplegia due to stroke, "All limited household ambulators at recruitment." Demonstrated that patients could "safely performed ambulatory functions with only one arm to maintain stability."
	Indirect Evidence: "The indirect evidence documents that regardless of upper extremity motor function, 23 patients were able to safely perform ambulatory functions in the Ekso device using only one arm to control the assistive device. The ability of such patients to support themselves and recover from a stumble is equivalent to that of a patient who has upper extremity strength of at least 4/5 in only one arm."
Training Effectiveness: Certified therapists can safely and independently operate the device.	The document describes a two-phase training and certification program for therapists, ensuring competence in screening, setup, fitting, operation, safety, and emergency management. The implication is that this training enables safe operation.

2. Sample Sizes and Data Provenance

Test Set (Clinical Studies):
- Study 1: 54 subjects total.
  - 34 subjects with affected upper extremity (UE) motor function of less than 4/5 in one arm.
  - 1 patient with affected UE motor function of 4/5.
  - 19 patients with UE motor function not recorded.
  - 216 walking sessions across both studies.
- Study 2: 8 subjects total.
  - All 8 subjects were limited household ambulators at recruitment.
- Data Provenance: Both studies are described as "Single center, prospective studies." The country of origin is not explicitly stated, but given the FDA submission, it is likely the United States.
Training Set: Not applicable for this device as it is a physical exoskeleton, not an AI/software device that requires a "training set" in the machine learning sense. The "training" described in the document refers to human therapist training.

3. Number of Experts and Qualifications for Ground Truth

Ground Truth for Test Set: The "ground truth" here is the clinical assessment of safety and ability to ambulate. This would have been established by the trained physical therapists supervising the sessions and likely by the principal investigator(s) of the studies. The document states the device is used "under the supervision of a trained physical therapist," and the therapists themselves undergo certification. While no specific number of "experts" are listed for establishing ground truth post-hoc, the safety and effectiveness were directly observed and recorded during the prospective studies by qualified clinical personnel. Their qualifications are implied by their role as physical therapists and study investigators.
Training Set Ground Truth: Not applicable.

4. Adjudication Method for the Test Set

The document does not describe a formal adjudication method (like 2+1 or 3+1 consensus) for the outcome of the clinical studies. The results are presented as direct observations and reports from the prospective studies ("No falls or other adverse events reported"). This implies the safety and functional performance were observed and recorded directly by the study investigators/therapists.

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

No, an MRMC comparative effectiveness study was not done. This type of study (comparing human readers with and without AI assistance) is relevant for diagnostic imaging AI. The Ekso device is a physical exoskeleton for rehabilitation, not a diagnostic AI.

6. Standalone (Algorithm Only) Performance

Not applicable. The Ekso is an exoskeleton requiring human interaction (the patient wearing it, the therapist supervising it). It does not have a "standalone" algorithm performance in the way an AI diagnostic tool would. Its performance is intrinsically linked to the human operator and patient.

7. Type of Ground Truth Used

Clinical Observation and Outcomes Data:
- The primary ground truth for the clinical studies was the direct clinical observation of patient safety (absence of falls, adverse events) and their ability to perform ambulatory functions while using the device, as assessed by trained physical therapists and study investigators.
- The "results" section directly references "safely performed ambulatory functions" and "no falls or other adverse events reported," which are observed clinical outcomes.

8. Sample Size for the Training Set

Not applicable. As noted above, this is a physical medical device, not an AI/machine learning algorithm requiring a "training set" of data. The "training" refers to physical therapist training.

9. How the Ground Truth for the Training Set Was Established

Not applicable. See point 8. The training for therapists involves demonstrating competence in operating the device and supervising patients safely, as evaluated by the "Ekso Bionics Clinical Training Team."

Ask a Question

Ask a specific question about this device

K Number

K143690

Device Name

Ekso (version 1.1) and Ekso GT (version 1.2)

Manufacturer

EKSO BIONICS, INC.

Date Cleared

2016-04-01

(464 days)

Product Code

Regulation Number

Type

Panel

Reference & Predicate Devices

K131798

Intended Use

Individuals with hemiplegia due to stroke
Individuals with spinal cord injuries at levels T4 to L5
Individuals with spinal cord injuries at levels of C7 to T3 (ASIA D).
The therapist must complete a training program prior to use of the device. The devices are not intended for sports or stair climbing.

Device Description

AI/ML Overview

The provided text describes the Ekso™ (version 1.1) and Ekso GT™ (version 1.2) powered lower extremity exoskeletons and their substantial equivalence to a predicate device (ReWalk™). The text primarily focuses on comparing the characteristics of the Ekso device with its predicate and detailing clinical studies to support its safety and performance for its stated indications for use.

Here's an analysis of the requested information based on the provided text:

1. A table of acceptance criteria and the reported device performance

The document does not explicitly state acceptance criteria in a quantitative, pass/fail format typical of certain medical device pre-market notifications. Instead, it presents a comparison with a predicate device and then details clinical study results to demonstrate safety and performance. The "acceptance criteria" can be inferred from the "Significant Differences" column in Table 1 and the "Results" section of the clinical studies, demonstrating that the device performs comparably or acceptably for its intended use without raising new safety or efficacy concerns.

Inferred Acceptance Criteria & Reported Performance (from comparison and clinical studies):

Acceptance Criteria (Inferred from Predicate Comparison & Clinical Studies)	Reported Device Performance (Ekso™)
Safety:
- No additional safety or efficacy concerns compared to predicate for expanded indications (stroke, higher SCI injury range).	- Clinical data evaluated safety for higher level SCI and stroke patients, limiting use to those with adequate upper extremity motor strength (at least 4/5 in both arms).
- Safety during ambulation and sitting/standing transitions.	- Clinical data supports safe use for ambulation and sitting/standing transitions despite larger range of motion.
- Failsafe feature allows user to recover during a fault.	- In event of power failure, knees become locked and hips free, allowing user to remain standing and recover.
- No serious adverse events (SAEs) such as falls or injuries reported in clinical studies.	- Study 1: No falls reported. 1 subject with blisters, 5 with pain (self-limiting).

Study 2: 1 subject had 2 falls without injury. No injuries reported.
Study 3: No falls or other adverse events reported.
Study 4: No falls or other adverse events reported.
Study 5: No falls or other adverse events reported. |
| Device Characteristics & Performance: | |
| - Similar indications for use but potentially expanded to include stroke patients and a wider range of SCI levels (C7-T3 ASIA D). | - Intended for hemiplegia due to stroke, SCI T4-L5, and SCI C7-T3 (ASIA D), with upper extremity motor function of at least 4/5 in both arms. Deemed "Similar" without additional safety concerns. |
| - Mobility Aid: Support for various mobility aids (walker, crutches, cane). | - Utilizes walker, crutches, cane. Deemed "Similar" with no additional safety or efficacy concerns for added options. |
| - Ability to sit, stand, walk, and turn. | - Can sit, stand, walk, and turn (same as predicate). |
| - Walking Speed: Comparable to predicate. | - ~2 km/hr (same as predicate). |
| - Grade of Inclination: Suitable for intended flat environments. | - 1.15 deg (lower than predicate's 5 deg, but consistent with indoor use, so no additional safety/efficacy concerns). |
| - Type of Surface: Suitable for smooth, cement, carpet. | - Smooth, cement, carpet. Not walking on grass presents no additional safety concerns as intended for indoor use. |
| - Patient Population: Adaptable to intended patient population. | - Adults over 18 with hemiplegia due to stroke, SCI T4-L5, and SCI C7-L5 ASIA D. Evaluated for these differences. |
| - Control Method: Efficient and safe for initiating steps. | - Handheld interface for PT; weight shift to initiate steps. Deemed "Similar" with no additional safety or efficacy concerns. |
| - Range of Motion: Allows for comfortable and functional movement. | - Hips: 135° flexion to 20° extension, Knees: 130° flexion to 0° extension, Ankles: 10° flexion to 10° extension (larger than predicate, supporting sitting/standing transitions and comfort). |
| - Weight: Lighter is preferable (no additional safety/efficacy concerns). | - 50 lbs (23 kg) - lighter than predicate, no safety/efficacy concerns. |
| - Battery: Provides necessary power and operates safely. | - Rechargeable lithium ion batteries (48.1V, 30A peak current, 1 hr continuous usage). Tested per specification, no additional safety/efficacy concerns. |
| - Battery Charge Time: Manageable for clinical use. | - 1 hour (shorter than predicate, allows for swapping, no additional safety/efficacy concerns). |
| - Expected Useable Life: Reasonable for medical device. | - 4 years (shorter than predicate, but within acceptable limits, no additional safety/efficacy concerns). |
| - Training Program and Certification: Ensures safe operation by trained professionals. | - Yes, extensive two-phase training and certification program for therapists (Level 1: basic, Level 2: advanced/independent). |
| - Clinical Effectiveness: Improvement in ambulation parameters (10MWT, 6MWT, FIM scores). | - Study 1 (SCI): Mean 10MWT improved (66s pre to 40s post).
Study 2 (SCI): Mean 10MWT improved (160s pre to 78s post), Mean 6MWT improved (82m pre to 152m post).
Study 3 (Stroke): Mean 10MWT improved (150s pre to 79s post), Mean 6MWT improved (40m pre to 88m post), Mean FIM scores improved (24 pre to 49 post).
Study 4 (Stroke): Mean FIM scores improved (25 pre to 51 post).
Study 5 (Stroke): Mean 10MWT improved (76s pre to 37s post), Mean 6MWT improved (48m pre to 83m post). |

2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)

The document lists five clinical studies:

Study 1 - SCI:
- Sample Size: 44 subjects (35 ASIA A-B, 21 ASIA C-D, some overlap likely given the numbers provided - total patients 44 in text).
- Data Provenance: Multi-center, prospective, open-label, non-comparative, non-randomized. Conducted at 6 sites from 5 countries.
Study 2 - SCI:
- Sample Size: 12 subjects (9 ASIA A, 3 ASIA B).
- Data Provenance: Single center, open-label, non-comparative, non-randomized, prospective.
Study 3 - Stroke:
- Sample Size: 54 subjects.
- Data Provenance: Single center, exploratory retrospective analysis.
Study 4 - Stroke:
- Sample Size: 54 subjects.
- Data Provenance: Single center, open-label, non-comparative, non-randomized, prospective.
Study 5 - Stroke:
- Sample Size: 8 subjects (6 in Arm 1, 2 in Arm 2).
- Data Provenance: Single center, prospective, two-arm study.

The countries of origin for the studies are only explicitly stated for "Study 1 - SCI" (6 sites from 5 countries).

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience)

This information is not provided in the document. The clinical studies report objective measures (e.g., 10MWT, 6MWT, FIM scores, heart rate, blood pressure) and adverse events, which would typically be recorded by trained clinical staff like physical therapists and physicians. The establishment of "ground truth" in terms of expert consensus on diagnosis or a gold standard interpretation is not applicable here as the studies assess functional outcomes and safety for device use. The document does state that the device is used "under the supervision of a trained physical therapist" and mentions a "certification program" for therapists.

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set

This information is not provided in the document. The clinical studies describe objective measurements of functional performance and adverse event reporting. There is no mention of an adjudication process by multiple experts for subjective outcomes or interpretations.

5. If a multi reader multi case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance

No MRMC comparative effectiveness study was done or mentioned. This type of study (MRMC) is typically relevant for diagnostic imaging AI devices where human readers (e.g., radiologists) interpret images with and without AI assistance. The Ekso device is a physical rehabilitation exoskeleton, not an AI diagnostic tool.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done

The Ekso device is an exoskeleton designed for use by patients under the supervision of a trained physical therapist. Its core function involves mechanical assistance for ambulation with human interaction and supervision. Therefore, a "standalone algorithm only" performance study is not applicable or relevant for this type of device. The performance is inherently human-in-the-loop.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)

The "ground truth" in these studies consists of clinical outcomes data and safety event reporting. This includes:

Functional performance metrics:
- 10-Meter Walk Test (10MWT) time (seconds)
- 6-Minute Walk Test (6MWT) distance (meters)
- Functional Independence Measure (FIM) scores
Physiological measurements: Heart rate, blood pressure.
Adverse events: Blisters, pain, falls (with or without injury).

These are direct measurements of patient response and device safety, not an "expert consensus" on a diagnosis or a "pathology" result.

8. The sample size for the training set

The document does not provide a specific sample size for a "training set" in the context of machine learning or AI. The clinical studies mentioned are for evaluating the device's performance and safety, not for training an algorithm within the device. The "training" described in the document refers to the program for physical therapists to learn how to operate the Ekso device.

9. How the ground truth for the training set was established

Since there is no "training set" in the context of machine learning described for the device itself, this question is not applicable. The "ground truth" for the therapist training is established through competency demonstration as assessed by the "Ekso Bionics Clinical Training Team" against defined operational and safety criteria.

Ask a Question

Ask a specific question about this device

Page 1 of 1