Search Results

It is a motor driven, indoor and outdoor transportation vehicle with the intended use to provide mobility to a disabled or elderly person limited to a seated position.

This product consists of frame, wheels, seat, armrest, lithium battery, motor and controller with a lightweight and compact design.The entire wheelchair can be disassembled, making it easy to carry. The wheelchair comes equipped with an electronically controlled rising seat that allows the user 280mm/11" of seat height adjustment using the joystick. The armrest can be flipped upside down, which is convenient for the elderly to move. Users can drive the wheelchair by themselves through the control device. The wheelchair uses lead-acid Battery as its power source. The controls the drive leftright motor to realize the wheelchair forward, backward and turn functions. The frame of the device is carbon steel. The front wheels suitable for rotation, acceleration, retrograde and other actions of the wheels movement will be achieved by thrust generated from the rear wheels are driving wheels to control the speed and direction. The wheels are Solid PU tires. When in use, the operator drives the motor of the rear wheel by operating the controller joystick to achieve the rear wheels movement. The DC brushless motor and brake system are fixed on the rear wheels. The max loading of the device is 136KG. Only for one person sit.

The provided text is a 510(k) summary for an electrically powered wheelchair. It focuses on demonstrating substantial equivalence to a predicate device rather than presenting a study with specific acceptance criteria and detailed device performance metrics in the format requested.

Therefore, many of the requested elements of information, such as sample sizes, data provenance, expert qualifications, and MRMC study details, are not applicable or not available in this type of document, as it describes the testing and comparison of a physical medical device (a wheelchair) rather than a diagnostic AI algorithm.

However, I can extract the acceptance criteria as indicated by compliance with various ISO standards and the 'reported device performance' based on the comparison table and the general statements about meeting requirements.

Here's the closest representation of the requested information based on the provided document:

1. Table of Acceptance Criteria and Reported Device Performance

2. Sample size used for the test set and the data provenance:

• Sample Size: The document does not specify a distinct "test set" in terms of patient data or case numbers. Instead, it refers to performance and safety testing of the physical device according to various international standards (ISO). For tests involving a dummy, it mentions validation "with the maximum rated weight dummy." The number of devices tested is not explicitly stated.
• Data Provenance: The testing appears to have been conducted as part of the device development and verification process by the manufacturer in China. The specific location or date of individual tests are not detailed beyond the document preparation date (2025/2/26). The studies are inherently "prospective" in the sense that they are conducted to verify the new device's compliance.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

• This section is not applicable. The device is an electrically powered wheelchair, not an AI diagnostic algorithm or system requiring expert-established ground truth from medical images or patient data. The "ground truth" for the device's performance is established by its compliance with engineering and safety standards, which are defined by standard bodies (like ISO) and regulatory bodies (like FDA), and verified through physical testing.

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

• This section is not applicable. Adjudication methods like 2+1 or 3+1 are typically used in clinical studies to resolve discrepancies among multiple expert readers for diagnostic tasks. For a physical device like a wheelchair, performance is measured against objective engineering and safety standards, not subjective expert assessment of a diagnostic outcome.

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:

• This section is not applicable. An MRMC study is relevant for evaluating the impact of AI on human reader performance in diagnostic tasks. This document is for a physical medical device (a wheelchair) and does not involve AI for diagnostic purposes or human readers.

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

• This section is not applicable. There is no AI algorithm being evaluated for standalone performance in this document. The device is a physical wheelchair.

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

• The "ground truth" in this context is defined by the technical specifications, requirements, and acceptable tolerances outlined in the referenced international standards (e.g., ISO 7176 series, ISO 10993 series, IEC 60601-1-2) which are based on engineering principles, safety considerations, and recognized testing methodologies. It's essentially engineering and safety standard compliance verified through objective physical testing.

8. The sample size for the training set:

• This information is not applicable as this document does not describe the development or testing of an AI model with a training set.

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

• This information is not applicable as this document does not describe the development or testing of an AI model with a training set.

Ask a specific question about this device

It is a motor driven, indoor and outdoor transportation vehicle with the intended use to provide mobility to a disabled or elderly person limited to a seated position.

This product consists of frame, wheels, seat, armrest, lithium battery, motor and controller with a lightweight and compact design. The whole wheelchair can be folded and it can be easily carried or rolled after folding. The seat cushion is detachable. The armrest can be flipped upside down, which is convenient for the elderly to move. Users can drive the wheelchair by themselves through the control device. The wheelchair uses lithium batteries as its power source. The controls the drive left/right motor to realize the wheelchair forward, backward and turn functions. The frame of the device is Magnesium alloy. The front wheels are driven wheels suitable for rotation, acceleration, retrograde and other actions of the wheelchair. The front wheels movement will be achieved by thrust generated from the rear wheels are driving wheels to control the speed and direction. The wheels are Solid PU tires.When in use, the operator drives the motor of the rear wheel by operating the controller joystick to achieve the rear wheels movement. The DC brushless motor and brake system are fixed on the rear wheels. The max loading of the device is 120KG. Only for one person sit.

The provided text is a 510(k) summary for an Electrically Powered Wheelchair (device name: HP558) and does not contain information about acceptance criteria or a study proving device performance in the context of AI/ML or diagnostic applications.

The document discusses the substantial equivalence of the proposed device to a predicate device (HP458E) by comparing their design, functionality, materials, and compliance with various ISO and IEC standards relevant to wheelchairs. It also mentions performance data to verify that the subject device meets design specifications, but this is in the context of standard engineering and safety testing for a medical device (electrically powered wheelchair), not a study assessing AI/ML model performance.

Therefore, I cannot provide the requested information, such as:

1. A table of acceptance criteria and reported device performance: This document lists various technical specifications and safety standards for the wheelchair itself, not performance metrics of an AI algorithm.
2. Sample size, data provenance, number of experts, adjudication method for the test set: These are relevant to studies evaluating AI/ML models on a dataset, which is not what this document describes.
3. MRMC comparative effectiveness study: No such study is mentioned.
4. Standalone performance (algorithm only): The device is a physical wheelchair, not an algorithm.
5. Type of ground truth used: This concept is not applicable to the evaluation described.
6. Sample size for the training set: Not applicable.
7. How ground truth for the training set was established: Not applicable.

The document is a regulatory submission for a physical medical device (an electrically powered wheelchair) and focuses on ensuring its physical safety, effectiveness, and substantial equivalence to an existing device through engineering and biocompatibility testing against established standards. It does not involve AI or diagnostic imaging.

Ask a specific question about this device

It is a motor driven, indoor and outdoor transportation vehicle with the intended use to provide mobility to a disabled or elderly person limited to a seated position.

This product consists of frame, wheels, seat, armrest, lithium battery, motor and controller with a lightweight and compact design. The whole wheelchair can be folded and it can be easily carried or rolled after folding. The seat cushion is detachable. The armrest can be flipped upside down, which is convenient for the elderly to move. Users can drive the wheelchair by themselves through the control device. The wheelchair uses lithium batteries as its power source. The controls the drive leftright motor to realize the wheelchair forward, backward and turn functions. The frame of the device is aluminum alloy. The front wheels suitable for rotation, acceleration, retrograde and other actions of the wheelchair. The front wheels movement will be achieved by thrust generated from the rear wheels are driving wheels to control the speed and direction. The wheels are Solid PU tires. When in use, the operator drives the motor of the rear wheel by operating the controller joystick to achieve the rear wheels movement. The DC brushless motor and brake system are fixed on the rear wheels. The max loading of the device is 120KG. Only for one person sit.

The provided document is a 510(k) Premarket Notification for an Electrically Powered Wheelchair (HP206). It primarily details the device's characteristics and compares it to a predicate device (HP358EA) to demonstrate substantial equivalence.

However, the document does not contain information about a study proving device performance against specific acceptance criteria, especially not in the context of an AI/human reader study as typically associated with the comprehensive questions asked. The device in question is an Electrically Powered Wheelchair, not a medical imaging or diagnostic AI device that would involve expert readers, ground truth establishment for a test set, multi-reader multi-case studies, or AI algorithm performance metrics like sensitivity, specificity, etc.

The "acceptance criteria" and "study that proves the device meets the acceptance criteria" in this context refer to the device's compliance with established performance standards and safety requirements for electrically powered wheelchairs.

Therefore, I cannot populate the requested table or answer most of the specific questions about AI performance, expert adjudication, or MRMC studies for this device, as these concepts are not applicable to the information provided in the 510(k) submission for a powered wheelchair.

I can, however, extract the performance data and compliance standards mentioned:

1. Table of acceptance criteria and the reported device performance

The document frames "acceptance criteria" as compliance with various ISO standards and FDA guidance. "Reported device performance" is demonstrated through testing against these standards. There isn't a specific quantitative "acceptance criteria" table with corresponding numerical "reported performance" easily extractable for each parameter the way one would for an AI diagnostic device. Instead, the document states that the device "met all design specifications" and provided "support of the substantial equivalence determination" by undergoing the listed tests.

Here's an attempt to structure it based on the available information, understanding that "acceptance criteria" is implied by compliance with the standards, and "reported performance" is that it passed the tests:

2. Sample size used for the test set and the data provenance:

• Sample Size: Not explicitly stated as a "sample size" in the conventional sense for a clinical trial or algorithm test set with individual cases. The testing refers to the device prototype(s) themselves. For mechanical and electrical performance tests, a limited number of test units (often 1-3) are typically used per test type, depending on the standard. For biocompatibility, material samples are tested.
• Data Provenance: The tests are performed to demonstrate compliance with international standards (ISO, IEC) generally by test labs. The document does not specify geographical provenance for the testing data in terms of "country of origin of the data" being retrospective or prospective from patient cohorts. It's device performance data.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

• This is not applicable to an electrically powered wheelchair device. "Ground truth" in this context would refer to the physical and electrical safety and performance parameters measured against established engineering and medical device standards. No "experts" are establishing "ground truth" in the diagnostic sense. Compliance is verified by testing according to published, objective standards.

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

• Not applicable. This refers to consensus reading for diagnostic purposes, which is not relevant here.

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:

• Not applicable. This is not an AI diagnostic device.

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

• Not applicable. This is not an AI algorithm.

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

• The "ground truth" for this device is based on established international performance and safety standards (ISO and IEC, as listed in the document) and FDA guidance for electrically powered wheelchairs. Compliance is demonstrated through physical, electrical, and materials testing.

8. The sample size for the training set:

• Not applicable. This is not an AI device trained on data.

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

• Not applicable.

Ask a specific question about this device

It is a motor driven, indoor and outdoor transportation vehicle with the intended use to provide mobility to a disabled or elderly person limited to a seated position.

The Electrically powered wheelchair is a motor driven, indoor transportation vehicle, which a device for assisting action handicapped people and disabled people to move. It is suitable for disabled people with mobility difficulties and elderly people.

It is mainly composed of front wheel, drive wheel, frame, controller (joystick), motor and drive devices, armrest, backrest, seat belt, footrest, battery box and charger.

The device is powered by Li-ion Battery pack (25V 10Ah) with 11.8 Km range, which can be recharged by an off-board battery charger that can be plugged into an AC socket outlet(100-240V, 50/60Hz) when the device is not in use.

The user can activate the controller handle (joystick) to control the speed and direction of the wheelchair movement. When the user releases the joystick will automatically go back to the central position and the wheelchair will be automatically stopped soon due to automatic electromagnetic brake system. Once the joystick is activated again move to other position, the wheelchair will be re-energized.

This document is a 510(k) summary for an electrically powered wheelchair (HP202). It focuses on demonstrating "substantial equivalence" to a legally marketed predicate device (K230964) rather than proving the device meets specific acceptance criteria through a clinical study of an AI/software component.

Therefore, the request for details related to AI/software performance evaluation (like acceptance criteria for an AI model, sample sizes for test/training sets, expert consensus, MRMC studies, standalone performance, etc.) cannot be fully answered from the provided text.

The document states: "No clinical study is included in this submission." and primarily relies on non-clinical testing against established ISO standards and direct comparison of specifications to the predicate device to argue for substantial equivalence.

Here's an attempt to answer the questions based only on the provided text, highlighting where information is absent for AI-specific criteria:

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

The acceptance criteria are generally implied by conformity to specific ISO standards and the performance metrics of the predicate device. The document presents a comparison table (Table 1: General Comparison, and implicitly Table 3: Performance Comparison) rather than a formal acceptance criteria table with a "Pass/Fail" or "Meets" column. The "Remark" or "Results" columns in these tables serve a similar purpose by indicating "S.E." (Substantially Equivalent) or explaining minor differences that do not raise new safety or effectiveness concerns.

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

This information is not provided as no clinical study was conducted. The testing described is primarily physical device testing against engineering standards. The manufacturer is based in China.

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 as no clinical study or expert-based ground truth establishment (relevant for AI/image analysis) was performed.

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

This information is not provided as no clinical study or expert-based ground truth establishment was performed.

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 study was done. The device is an electrically powered wheelchair, not an AI-assisted diagnostic tool for human readers.

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

A standalone performance test for an AI algorithm was not performed because the device itself is an electrically powered wheelchair and the submission does not describe an AI algorithm component requiring such testing. The "software validation" mentioned relates to the wheelchair's control system, not an AI for interpretation or diagnosis.

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

The "ground truth" for this device's performance is established by conformance to internationally recognized engineering standards (ISO 7176 series, ISO 10993 series) and direct comparison of technical specifications and safety features to a legally marketed predicate device. This is a non-clinical "ground truth."

8. The sample size for the training set

This information is not provided as no AI training set was described or used in this submission.

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

This information is not provided as no AI training set was described or used in this submission.

In summary: The provided FDA 510(k) summary for the "Electrically powered wheelchair (HP202)" demonstrates substantial equivalence through non-clinical testing against ISO standards and direct comparison to a predicate device. It does not involve a clinical study or the evaluation of an artificial intelligence/software component where the requested details about acceptance criteria, sample sizes, expert ground truth, or human-AI interaction would be relevant.

Ask a specific question about this device

It is a motor driven, indoor and outdoor transportation vehicle with the intended use to provide mobility to a disabled or elderly person limited to a seated position.

This product consists of frame, wheels, seat, armrest, lithium battery, motor and controller with a lightweight and compact design. The whole wheelchair can be folded by one button and it can be easily carried or rolled after folding, The seat cushion is detachable. The armrest can be flipped backward, which is convenient for the wheelchair by the wheelchair by themselves through the control device.

The wheelchair uses lithium batteries as its power source. The drive left/right motor to realize the wheelchair forward, backward and turn functions.

The frame of the device is carbon fiber. The front wheels suitable for rotation, acceleration, retrograde and other actions of the wheelchair. The front wheels mover will be achieved by thrust generated from the rear wheels are driving wheels to control the speed and direction. The wheels are Solid PU tires.

When in use, the operator drives the motor of the rear wheel by operating the controller handle (joystick) to achieve the rear wheels movement.

The DC Brushless motor and Brake system are fixed on the rear wheels. The max loading of the device is 125KG. Only for one person sit.

This is an FDA 510(k) premarket notification for an Electrically Powered Wheelchair (K241632).

For devices like this (powered wheelchairs), the acceptance criteria and study that proves the device meets them typically involve demonstrating substantial equivalence to a legally marketed predicate device, primarily through non-clinical performance testing to relevant international standards. Unlike AI/ML-based medical devices, there isn't a "test set" in the sense of a dataset for algorithm evaluation, nor are there "expert readers" establishing ground truth in the same way. The "performance" refers to the physical and functional aspects of the wheelchair.

Here's a breakdown based on the provided document:

Acceptance Criteria and Reported Device Performance

The document states that the device's acceptance criteria are met by demonstrating compliance with various international ISO and IEC standards relevant to electrically powered wheelchairs. The reported "performance" is inherently tied to passing these standards.

Table 1: Acceptance Criteria and Reported Device Performance

The document explicitly states: "The conclusions drawn from the nonclinical tests demonstrate that the subject device is as safe, as effective, and performs as well as the legally marketed predicate device K231508."

Study Details

Given the nature of the device (electrically powered wheelchair) and the information provided in the 510(k) summary, the "study" is a series of non-clinical, bench-top, and possibly simulated-use tests to demonstrate compliance with the referenced international standards.

1. Sample size used for the test set and the data provenance:

   • Sample Size: The document does not specify a numerical sample size (e.g., number of wheelchairs tested). For non-clinical performance testing of physical devices, testing typically involves a representative sample or a single unit (depending on the test standard and design).
   • Data Provenance: The document does not explicitly state the country of origin for the testing data itself. The applicant is Kunshan Hi-Fortune Health Products Co., Ltd in China, and their consultant is in Shanghai, China. It is highly probable that the testing was conducted in laboratories in China, often by accredited testing houses. The testing would be considered prospective in the sense that it was conducted specifically to demonstrate compliance for this 510(k) submission.

2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

   • This question is not applicable in the context of this device and 510(k) submission. "Ground truth" established by experts (like radiologists for AI algorithms) is not a component of demonstrating substantial equivalence for an electrically powered wheelchair. The "ground truth" here is the pass/fail criteria defined by the requirements of the international standards themselves. Compliance is verified by engineers and technicians performing the tests according to standard protocols.

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

   • This is not applicable. Adjudication methods like 2+1 or 3+1 are used for expert consensus on image interpretation or clinical outcomes, typically in studies involving human readers or clinical trials. For non-clinical performance testing against engineering standards, compliance is objectively measured.

4. 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:

   • This is not applicable. MRMC studies are used to evaluate AI-assisted diagnostic devices in a clinical reading setting. This 510(k) is for a powered wheelchair, not a diagnostic device.

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

   • This refers to AI algorithm performance. This is not applicable as this is not an AI/ML device. The "standalone performance" refers to the physical functionality of the wheelchair according to the specified standards.

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

   • The "ground truth" for this device's performance is objective compliance with the pass/fail criteria and specifications outlined in the referenced international standards (e.g., maximum speed, turning radius, battery performance, mechanical strength, electromagnetic compatibility limits, flammability). It is a standards-based performance evaluation, not clinical outcomes or expert interpretation.

7. The sample size for the training set:

   • This is not applicable. There is no "training set" in the context of a physical device like a powered wheelchair unless it uses an AI/ML component (which is not described or implied here). The design and engineering of the wheelchair are based on established principles, not a data training set.

8. How the ground truth for the training set was established:

   • This is not applicable for the same reasons as above.

Ask a specific question about this device

The device is a motor-driven, and indoor transportation vehicle with the intended use to provide mobility to a disabled or an elderly person limited to a seated position.

The proposed device, Electrically Powered Wheelchair, mainly powered by battery, motivated by DC motor, driven by user controlling joystick and adjusting speed.

The Electrically Powered Wheelchair consists of two foldable armrests, a backrest, a cushion, a foldable frame, two rear driving wheels with hub seat motor/electromagnetic brake assemblies, two pivoting casters, a Li-ion batteries, an off-board battery charger, a control panel, and an electric motor controller.

The electric wheelchair is intended to provide mobility to a person with a disability or an older adult limited to a sitting position.

Folding design: This foldable electric wheelchair has main frame, rear frame, backrest frame, seat frame, front wheel frame and battery frame. Release the locking device of the backrest frame and the rear frame, push the backrest frame downward, drive the seat frame to rotate toward the backrest frame with the rotating mechanism, the two armrest frames rotate towards the direction of the seat frame. At the same time two front wheel frames move to the direction of the seat frame. When the back cushion is in contact with the seat cushion, the folding is completed.

Joystick: This controls the speed and direction of the wheelchair. Push the joystick in the direction you wish to go. The further you push it, the faster the speed. Releasing the joystick stops the wheelchair and automatically applies the brakes.

Controller: The controller includes a power switch, a speed adjustment button, a horn button and a direction joystick. The operation interface controller receives the operation signal and transmits it to the main board of the controller body, and sends a control signal to the electric wheelchair to adjust and operate the electric wheelchair.

Wheel and frame connection method: First, the left and right frames are connected to the motor respectively, and then the motor shaft is connected to the wheel. The motor rotates to drive the wheels to rotate to drive the wheelchair.

Frame design: 1. The frame tube of this product is mostly is aluminum tube.

2. The operation mode of the folding mechanism is to remove the locking device by someone else and manually push the backrest frame to complete the folding operation under non-riding state.

The subject device has 6 inch front wheel and 10 inch rear tire.

The motor of electric wheelchair is DC24V 120W; the battery is 25.2V 10.4AH, Li-ion battery; the charger is 24V/2A.

Max. loading can not be over than 120Kgs.

The following surfaces are re commended NOT to operate on:

Sand surface

Wet or icy surface

Road maintenance hole metal cover

Do not use on stairs Do not use escalators. Use the elevator. Too steep incline over 6 degrees. Obstacle climbing ability: 1.97" Do not use outdoors

This document is a 510(k) Premarket Notification for an Electrically Powered Wheelchair (Model: HP458E). It focuses on demonstrating substantial equivalence to a predicate device (Powered Wheelchair DYW30A(D09), K170787) rather than providing detailed acceptance criteria and a study proving device performance in the context of an AI/ML-driven medical device.

Therefore, the requested information regarding acceptance criteria, sample sizes for test and training sets, number and qualifications of experts for ground truth, adjudication methods, MRMC studies, standalone performance, and ground truth establishment cannot be found in this document, as it pertains to a different type of medical device submission (electrically powered wheelchair) and does not involve AI/ML.

The document primarily relies on non-clinical testing against established ISO standards for wheelchairs and a comparison of technological characteristics to assert substantial equivalence.

However, I can extract information related to the device's technical specifications and the standards it meets, which serve as its "acceptance criteria" in the context of mechanical and electrical performance.

Here's an interpretation based on the provided document, assuming "acceptance criteria" generally refers to the performance specifications and safety standards the device is required to meet:

1. Table of Acceptance Criteria and Reported Device Performance

For an electrically powered wheelchair, "acceptance criteria" are defined by compliance with relevant international standards and performance specifications that demonstrate safety and effectiveness. The reported device performance is demonstrated through testing against these standards and by comparing its technical specifications to a predicate device.

2. Sample size used for the test set and the data provenance

• Test Set: Not applicable in the biological/imaging data sense, as this is a mechanical/electrical device. The "test set" here refers to the actual physical device model (HP458E) that underwent non-clinical performance and safety testing.
• Data Provenance: The tests were conducted according to international standards (ISO, IEC). The document does not specify the country of origin for the testing data or whether it was retrospective or prospective, beyond stating "non clinical tests were conducted to verify that the proposed device met all design specifications."

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

• Not applicable. This is not an AI/ML device relying on expert interpretation for ground truth. Ground truth for a wheelchair's performance is established by objective measurements and standardized testing procedures (e.g., measuring speed, stability angles, braking distance according to ISO standards).

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

• Not applicable. No human interpretation requiring adjudication.

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

• Not applicable. This is not an AI-assisted diagnostic device.

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

• Not applicable. This is a physical device, not an algorithm.

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

• The "ground truth" for this device's performance and safety is defined by international consensus standards (ISO, IEC) that specify test methods and acceptable performance limits for wheelchairs. This includes objective measurements of physical properties (e.g., dimensions, weight), dynamic performance (speed, braking, climbing), stability, and material properties (biocompatibility, flame retardancy).

8. The sample size for the training set

• Not applicable. This is not an AI/ML device that requires a training set.

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

• Not applicable.

Ask a specific question about this device

The device is intended for medical purposes to provide mobility to persons restricted to a sitting position.

Not Found

This document is a FDA 510(k) clearance letter for an electrically powered wheelchair. It does not contain information about the acceptance criteria or a study proving that the device meets acceptance criteria, as one would find for a diagnostic or AI-enabled medical device.

Therefore, I cannot extract the requested information regarding acceptance criteria, study details, sample sizes, expert qualifications, or ground truth establishment. This type of information is typically found in performance testing reports, clinical study summaries, or validation reports, which are not part of this 510(k) clearance letter.

Ask a specific question about this device

The device is a motor-driven, and indoor transportation vehicle with the intended use to provide mobility to a disabled or an elderly person limited to a seated position.

The Electrically Powered Wheelchair - Model HP358E is powered by Li-ion battery, driven by DC motor. Users control direction and adjust speed by controller. The main frame of the wheelchair is made of aluminum alloy. It has a seat base with four-wheeled and two with a back cover on the frame. Wheels are connected to the stainless steel axle receivers via stainless steel axles. On the front end of the frame are assembled two caster housings. Caster forks are mounted to these housings via steel axles. There are two rear caster wheels in back and anti-tip wheels in front to prevent the power chair from tipping. On the front end of the frame, upon the caster, a foot platform is assembled for foot holding. Two non-adjustable armrests are mounted on the frame for user's arm holding. Two grips mounted on the top end of the frame with a brake, and another brake is designed under the seat base. The backrest can be adjusted by squeezing yellow catches on the back together. A storage compartment located between the seat and foot platform. The joystick can be mounted on the left or right side using the quick release mechanism. The subject device has two 2 inch Anti-tippers, two 8 inch front wheels and two 12 inch rear wheels. The motor is DC24V 250W; the battery is 25.2 V 10.4Ah Li-ion battery; the charger is DC 24V, 2A. Max. loading can not be over than 125Kg, the maximum distance of travel on the fully charged battery is 10km and its maximum speed is 6km/h. The braking time is about 2~3s, and the braking distance is ≤1.0m.

The provided text is a 510(k) Summary for an Electrically Powered Wheelchair (Model HP358E). It focuses on demonstrating substantial equivalence to a predicate device rather than presenting a study of a device's performance against specific acceptance criteria in the context of AI or advanced diagnostic algorithms. Therefore, much of the requested information regarding AI device evaluation (e.g., sample size for test set, provenance, expert adjudication, MRMC studies, standalone performance, training set details) is not applicable or present in this document.

However, I can extract the relevant "acceptance criteria" (which in this context are performance and safety standards) and the "device performance" as reported for the proposed device, along with other applicable details.

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

The "acceptance criteria" for this wheeled device are derived from various ISO and IEC standards that it claims to comply with, and the "reported device performance" are the measured specifications of the HP358E. These are primarily found in the "Non-Clinical Test Summary" and the "Performance Comparison" table.

Acceptance Criteria & Reported Device Performance for Electrically Powered Wheelchair - Model HP358E

2. Sample sized used for the test set and the data provenance:

This document describes the testing of a physical medical device (an electrically powered wheelchair) against engineering and safety standards. The "test set" here refers to the actual device prototypes or samples subjected to the described non-clinical tests. The number of individual devices tested is not specified, but it's standard practice in such non-clinical testing to test a representative sample size to ensure reliability and repeatability according to the testing standards (e.g., ISO and IEC).

• Sample Size: Not explicitly stated for each test, but implied to be a sufficient number of device units to perform the required non-clinical tests according to the cited ISO/IEC standards.
• Data Provenance: The tests are non-clinical engineering and safety performance tests, not clinical data from patients. The document does not specify the location where these tests were performed, but the manufacturer is based in China. The data would be derived from the physical testing of the device.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

This is not applicable as the "ground truth" for non-clinical performance and safety of a wheelchair is established by the objective measurements against established international engineering and safety standards (ISO, IEC, EN). There are no "experts" establishing a "ground truth" in the way radiologists establish ground truth for an AI diagnostic algorithm; rather, the device's performance is measured and compared against the limits and criteria defined by the standards.

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

This is not applicable for non-clinical engineering and safety tests. Adjudication methods like 2+1 or 3+1 are used for establishing ground truth in clinical imaging studies, particularly for AI device evaluations where human expert consensus is required for labeling data.

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:

This is not applicable. The device is an electrically powered wheelchair, not an AI-powered diagnostic tool, and therefore, no MRMC study or AI assistance evaluation was performed or is relevant.

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

This is not applicable. The device is a physical medical device, not an algorithm, so the concept of standalone algorithm performance is irrelevant.

7. The type of ground truth used:

The "ground truth" for the device's performance and safety is derived from established international engineering and safety standards (ISO, IEC, EN). The device's physical and functional attributes were measured and assessed to ensure they comply with the stipulated requirements of these standards. For biocompatibility, the ground truth is established by adherence to ISO 10993 series standards and related FDA guidance, with specific tests like cytotoxicity, irritation, and skin sensitization. For software, the ground truth for its safety and functionality is established through validation against FDA guidance for "Moderate Level of Concern" software.

8. The sample size for the training set:

This is not applicable. The document does not describe an AI/ML device that requires a training set.

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

This is not applicable, as there is no training set for this device.

Ask a specific question about this device

The device is intended for medical purposes to provide mobility to persons restricted to a sitting position.

The proposed device, Wheelchair model HM303, is traditional manually operated, user propelled, manual, mechanical wheelchairs. The device can be foldable easily for transport.

The main frame is made of magnesium alloy frame, which has a seat base with four-wheeled with a back cover. Upon the outside of this framework, and to the rear, are assembled two axle plates, wheels are connected to the stainless steel axle receivers via stainless steel axles. On the front end of the frame are assembled two caster housings. Caster forks are mounted to these housings via steel axles. On the front end of the frame, upon the caster, two footplates are assembled for foot holding. Two armrests are mounted on the frame for user's arm holding. Two grips mounted on the top end of the frame with a brake, and another brake is designed under the seat base. A folden set is on the frame, under the grips, for folding backrest.

The provided text does not contain information about acceptance criteria and a study that proves a device meets those criteria in the context of a medical device with an AI component or a comparative effectiveness study.

The document is a 510(k) premarket notification for a Wheelchair, which is a mechanical device, not an AI-powered one. The submission focuses on demonstrating substantial equivalence to a predicate mechanical wheelchair through non-clinical performance and biocompatibility testing against established ISO standards.

Here's why the requested information cannot be extracted from the provided text:

• No AI Component: The device described is a "Mechanical Wheelchair" (Model HM303). There is no mention of any artificial intelligence, machine learning, or software algorithm components.
• No Clinical Study for Effectiveness: The document explicitly states "No Clinical Test conducted" in Section 8. The evaluation relies solely on non-clinical tests (physical performance and biocompatibility) and comparison to a predicate device.
• No Comparative Effectiveness Study: Since there's no AI component and no clinical study, there is no multi-reader multi-case (MRMC) comparative effectiveness study or analysis of human reader improvement with AI assistance.
• No Standalone Algorithm Performance: The concept of "standalone (algorithm only without human-in-the-loop performance)" is inapplicable because there is no algorithm or AI involved.
• No Ground Truth for AI Models: The establishment of ground truth for training or test sets is relevant for AI/ML models. For a mechanical wheelchair, performance is assessed against physical standards, not against expert labels or pathological outcomes for diagnostic accuracy.

Therefore, I cannot provide the requested information about acceptance criteria and a study proving the device meets them in the context of AI/ML or comparative effectiveness from the given document.

The document details the following regarding performance and acceptance for a mechanical wheelchair:

1. Table of Acceptance Criteria and Reported Device Performance:

2. Sample size used for the test set and the data provenance: Not applicable. For mechanical device testing against ISO standards, the "sample size" refers to the number of units tested, which is not specified but is typically a small number (e.g., 1-3 units) unless stress testing multiple units. The tests are non-clinical, likely performed in China (country of origin for the manufacturer).

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts: Not applicable. Ground truth for mechanical performance and biocompatibility is established by the specifications of international standards (ISO) and laboratory test results, not by expert interpretation in the medical sense.

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set: Not applicable. Adjudication methods are relevant for subjective interpretations (e.g., image reading). Mechanical and biocompatibility tests have objective pass/fail criteria based on standards.

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: Not applicable, as this is not an AI device.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done: Not applicable, as this is not an AI device.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc): The "ground truth" for this mechanical device is defined by international standards (ISO) for safety, performance, and biocompatibility, as well as FDA regulatory requirements for labeling.

8. The sample size for the training set: Not applicable. This is a mechanical device, not an AI/ML model that requires a training set.

9. How the ground truth for the training set was established: Not applicable.

Ask a specific question about this device