Search Results

Ask a specific question about this device

The Upper Arm Automatic Digital Blood Pressure Monitor, Model BP Progress (BP3T01-1B) is a device intended to measure the systolic and diastolic blood pressure and pulse rate of an adult individual by using a non-invasive oscillometric technique in which an inflatable cuff is wrapped around the upper arm for a circumference range from 22 to 40cm. The device detects the appearance of irregular heartbeat during measurement and gives a warning signal with the reading once the irregular heartbeat is detected. The device can be used in connection with a smart phone via Bluetooth. The measurement data can be transferred to a smart phone running the Microlife Connected Health+ mobile software (App).

The Upper Arm Automatic Digital Blood Pressure Monitor, Model BP Progress (BP3T01-1B) is designed to measure systolic and diastolic blood pressure, pulse rate of an individual with arm circumference sizes ranging from 22 to 40 cm by using a non-invasive technique in which one inflatable cuff is wrapped around the single upper arm. Our method to define systolic and diastolic pressure is similar to the auscultatory method but using a semiconductor sensor rather than a stethoscope and mercury manometer. The sensor converts tiny alterations in cuff pressure to electrical signals, by analyzing those signals to define the systolic and diastolic blood pressure and calculating pulse rate, which is a well - known technique in the market called the "oscillometric method". The device detects the appearance of irregular heartbeat during measurement, and the symbol " " is displayed after the measurement. In addition, the device can be used in connection with smart mobile devices running the Microlife Connected Health+ mobile software (App) via Bluetooth. The blood pressure monitor is a fully automatic digital blood pressure measuring device use by adults on the upper arm at home.

The provided FDA 510(k) clearance letter and summary for the Microlife Upper Arm Automatic Digital Blood Pressure Monitor, Model BP Progress (BP3T01-1B), include information about its testing and equivalence to a predicate device. However, it does not contain explicit acceptance criteria thresholds (e.g., specific accuracy ranges for blood pressure measurements) or detailed results beyond stating that the device "met all relevant requirements" and "results were passing."

Based on the provided text, here's an attempt to extract the requested information, with notable gaps where details are not explicitly stated.

Acceptance Criteria and Study Details for Microlife Upper Arm Automatic Digital Blood Pressure Monitor, Model BP Progress (BP3T01-1B)

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly state numerical acceptance criteria for blood pressure accuracy (e.g., mean difference and standard deviation standards like those in ISO 81060-2). It only mentions that the device "met all relevant requirements" and "results were passing" according to the relevant standards.

2. Sample Size Used for the Test Set and Data Provenance

• Sample Size: The document does not specify the exact sample size for the clinical validation study conducted in accordance with ISO 81060-2.
• Data Provenance: Not specified (e.g., country of origin, retrospective or prospective). Typically, such clinical validations are prospective, but this is not explicitly stated.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications

• The document does not provide details on the number or qualifications of experts used for establishing ground truth, as is typical for an oscillometric blood pressure monitor validation. For ISO 81060-2, readings are independently taken by multiple trained observers, but their specific qualifications (e.g., experience level) are not detailed here.

4. Adjudication Method for the Test Set

• The document does not explicitly state an adjudication method. For ISO 81060-2, the ground truth (reference blood pressure) is established by two or more trained observers using auscultation. Discrepancies between observers are typically handled by a predefined protocol (e.g., averaging, or a third observer if differences are significant), but these specifics are not mentioned in the provided text.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done

• No, an MRMC comparative effectiveness study was not done. This type of study (comparing human readers with and without AI assistance) is not relevant for a standalone blood pressure monitor. The device itself performs the measurement; it does not assist human readers in interpreting complex imagery or clinical data in the way an AI algorithm might.

6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done

• Yes, a standalone performance assessment was effectively done. The clinical validation conducted according to ISO 81060-2 assesses the device's accuracy in measuring blood pressure independently, without human interpretation other than setting up the device and initiating the measurement. The device's measurement algorithm is evaluated directly against a reference standard.

7. The Type of Ground Truth Used

• Expert Consensus (Auscultation): For blood pressure monitor validation per ISO 81060-2, the ground truth for blood pressure measurements is established through simultaneous or closely timed readings by trained human observers using the auscultatory method (stethoscope and sphygmomanometer/mercury manometer), which is considered the clinical reference standard.

8. The Sample Size for the Training Set

• Not Applicable / Not Provided: The summary indicates that the subject device (BP3T01-1B) uses the "same oscillometric method" and "common blood pressure measurement technological architecture and algorithm" as the predicate device (BP3KV1-5K). It does not describe a new algorithm that required a specific training set. Blood pressure algorithms for oscillometric devices are typically developed based on physiological models and empirical data, rather than being "trained" in the machine learning sense with a discrete "training set" in the context described here. If iterative development was done, the size of data used for such development is not provided.

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

• Not Applicable / Not Provided: As in point 8, the concept of a "training set" with established ground truth as it applies to machine learning models is not explicitly relevant or described for this blood pressure monitor's algorithm development in the provided documentation. The algorithm's fundamental principles are stated to be the same as the predicate device, implying established and verified methods rather than a newly trained model requiring specific ground truth for training.

Ask a specific question about this device

Fully Automatic Electronic Blood Pressure Monitor is for use by medical professionals or at home and is a non-invasive blood pressure measurement system intended to measure the diastolic and systolic blood pressures and pulse rate of an adult individual by using a non-invasive technique in which an inflatable cuff is wrapped around the upper arm. The cuff circumference is limited to 15cm-48cm (approx. 5.9"-18.9").

Fully Automatic Electronic Blood Pressure Monitor (BP-300C, BP-300CV, BP-300V, BPM1, BPX1, KD-338N, KD-553, KD-557BR, KD-558, KD-558BR, KD-595, KD-5031N, KD-5810, KD-5810B, KD-5811, KD-5811A, KD-5811V, KD-5815, KD-5920, KD-5920L, KD-5920TL, KD-5923, KN-550LT) is designed and manufactured according to IEC 80601-2-30.

The operational principle is based on Oscillo-metric and silicon integrates pressure sensor technology. It can calculate the systolic and diastolic blood pressure and display the result. The measurements results can also be classified by the function of blood pressure classification indicator. If any irregular heartbeat is detected, it can be shown to the user.

The provided document is a 510(k) clearance letter for various blood pressure monitors. It outlines the regulatory approval process and compares the new devices to a predicate device. However, it does not contain the detailed acceptance criteria and study results in the format typically used for AI/software devices.

Specifically, this document describes validation against standards for medical electrical equipment (IEC 60601-1, IEC 60601-1-2, IEC 60601-1-11) and automated non-invasive sphygmomanometers (IEC 80601-2-30, ISO 81060-2). It focuses on the substantial equivalence of the physical blood pressure monitors and their underlying oscillometric and pressure sensor technology, rather than the performance of an AI algorithm based on a test set, ground truth, and expert interpretations.

Therefore, many of the specific questions about AI/software device validation (e.g., sample size for the test set, data provenance, number of experts for ground truth, MRMC studies, standalone performance, training set details) cannot be answered from this document.

However, I can extract information related to the performance of the blood pressure monitors themselves, based on the included standards.

Acceptance Criteria and Device Performance (for Blood Pressure Monitor functionality, not AI):

Since this is a blood pressure monitor, the primary performance criteria relate to its accuracy in measuring blood pressure and pulse rate, and compliance with relevant safety and performance standards for automated non-invasive sphygmomanometers.

• IEC 60601-1:2005+AMD1: 2012+AMD2: 2020
• IEC 60601-1-2:2014+AMD1: 2020
• IEC 60601-1-11: 2015+AMD1: 2020
• IEC 80601-2-30: 2018 | All listed standards were met, demonstrating basic safety, essential performance, EMC, and home healthcare environment compliance. |

Unable to Answer from Document (Common for AI/Software Device Submissions, but not for this type of device):

The following questions are not applicable or cannot be answered from this 510(k) summary because the device described is a physical blood pressure monitor, not an AI/software device that interprets medical images or other complex data requiring expert adjudication, training sets, or MRMC studies.

• Sample size used for the test set and the data provenance:
  • Test Set Size: "A total of 231 patients (107 males and 124 females) were enrolled in the study." This is the clinical study population for blood pressure measurement accuracy.
  • Data Provenance: Not explicitly stated (e.g., country of origin). The study is described as a "clinical study," which implies prospective data collection for the purpose of the study.
• 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):
  • Not applicable. Ground truth for blood pressure measurement is established through a standard auscultation method (manual measurement by medical professionals using a stethoscope and sphygmomanometer), not by interpretation of images by experts.
• Adjudication method (e.g. 2+1, 3+1, none) for the test set:
  • Not applicable. Ground truth is direct measurement by a reference method.
• 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.
• If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
  • The device is a standalone blood pressure monitor. No human-in-the-loop interaction for interpretation (as in AI devices) is relevant. Its performance is its direct measurement accuracy.
• The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
  • Ground Truth: "Standard auscultation method was used as the reference blood pressure monitor measuring." This is the established clinical standard for direct comparison.
• The sample size for the training set:
  • Not applicable. This is not an AI/machine learning device requiring a training set.
• How the ground truth for the training set was established:
  • Not applicable.

Ask a specific question about this device

ArteVu is intended to noninvasively and continuously measure a patient's blood pressure and pulse rate, which are derived from the pulse pressure waveform using the scientific method of pulse waveform decomposition, for use on adult patients aged between 50 and 86 years who are resting in a supine or similarly reclined position.

ArteVu is calibrated using an ISO 81060-2 compliant sphygmomanometer. All parameters derived by ArteVu are shown on a compatible remote display monitor (RDDS) via wired transmission. The device is intended for use by clinicians or other properly trained medical personnel in professional healthcare facilities.

ArteVu is an automatic, continuous, and non-invasive blood pressure (CNBP) monitoring system designed for adult patients at rest and intended for use by medical professionals. The device features a disposable Finger Clip containing a tactile sensor that detects pulse pressure waveforms at the fingertip. ArteVu utilizes the scientific method of pulse waveform decomposition to derive blood pressure and pulse rate, with initial calibration performed using a non-invasive upper arm cuff. These measurements are displayed on a compatible remote monitor, updated every two seconds via wired transmission. ArteVu incorporates technical and physiological alarms to enhance reliability, providing continuous and accurate monitoring while alerting users to abnormal conditions.

The provided FDA 510(k) clearance letter for ArteVu does not contain the detailed information necessary to fully answer all aspects of your request regarding acceptance criteria and the study proving the device meets these criteria. Specifically, it lacks a table of acceptance criteria with reported device performance metrics and explicit details on how ground truth was established for training and testing sets.

However, based on the information available, here's a structured response addressing the requested points to the best of what the document provides:

Acceptance Criteria and Study for ArteVu

The document states that ArteVu's safety and effectiveness were validated through a clinical study that adhered to the acceptance criteria of ISO 81060-2 for substantial equivalence to the predicate device, CareTaker4. It also incorporated elements from IEEE 1708, ISO 81060-3, ISO 80601-2-61, and IEC 60601-2-27. While it doesn't provide a specific table of numerical acceptance criteria or reported values for ArteVu, it implicitly relies on the standards set by ISO 81060-2 for non-invasive sphygmomanometers. This standard typically defines accuracy requirements for blood pressure measurements.

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

The document does not provide an explicit table with numerical acceptance criteria and ArteVu's reported performance metrics against those criteria. It only states that the study design "adhered to the acceptance criteria of ISO 81060-2."

If this were a complete submission, such a table would typically include:

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

• Sample Size for Test Set: 109 subjects. The document states, "ArteVu's safety and effectiveness have been validated through a clinical study conducted in Taiwan involving 109 subjects." Since this is the primary validation study mentioned, it serves as the test set for the device's performance claims.
• Data Provenance: The clinical study was "conducted in Taiwan." The data is prospective, as it was collected as part of a clinical study to validate the device.
• Subject Recruitment: Subjects were recruited from "operating rooms and intensive care units."

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

The document states that ArteVu is calibrated using an "ISO 81060-2 compliant sphygmomanometer" and that the study design "adhered to the acceptance criteria of ISO 81060-2." This strongly implies that the ground truth for blood pressure measurements was established using a reference standard device (the compliant sphygmomanometer) and not necessarily by a panel of human experts. Therefore, the concept of "number of experts" for establishing ground truth via consensus (as might be seen in image-based AI studies) does not directly apply here. The "experts" would be the clinical personnel performing the reference measurements according to the ISO standard.

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

Since the ground truth for blood pressure measurements in this context is established by a reference device (ISO 81060-2 compliant sphygmomanometer) and not by subjective interpretation of medical images or conditions by multiple human readers, a numerical adjudication method (like 2+1 or 3+1) is not applicable or mentioned. The accuracy of the sphygmomanometer itself is the standard.

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, an MRMC comparative effectiveness study was not done as described in the document. This type of study is commonly used for AI in diagnostic imaging (e.g., radiology) where AI assists human interpretation. ArteVu is a continuous non-invasive blood pressure monitoring system, so its primary function is measurement, not assisting human readers in interpreting complex medical data.
• Therefore, there is no mention of effect size related to human readers improving with or without AI assistance.

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

• Yes, in essence, a standalone performance assessment was conducted for the device's core function. ArteVu is described as an "automatic, continuous, and non-invasive blood pressure (CNBP) monitoring system." Its performance (accuracy against a reference standard) is evaluated directly, implying it operates as an algorithm determining BP from the pulse waveform. The clinical study validated the device's ability to "achieve comparable safety and effectiveness" to the predicate device, which is a standalone measurement. While it displays measurements for "clinicians or other properly trained medical personnel," the core measurement derivation is done by the device itself, making it a standalone function in terms of its output.

7. The type of ground truth used

• The primary ground truth used is a reference standard measurement from an ISO 81060-2 compliant sphygmomanometer. This standard specifies the requirements for non-invasive sphygmomanometers and their clinical validation. The document explicitly states, "ArteVu is calibrated using an ISO 81060-2 compliant sphygmomanometer" and that the "study design adhered to the acceptance criteria of ISO 81060-2."

8. The sample size for the training set

• The document does not specify the sample size for the training set used to develop or train the ArteVu algorithm. The 109 subjects mentioned are for the validation/test set. Typical 510(k) summaries often do not disclose detailed training set information unless it's critical to the novelty or specific performance claims of an AI/ML device. While ArteVu uses a "scientific method of pulse waveform decomposition," it's unclear if this involves a machine learning model that requires a dedicated training set as opposed to an algorithm based on established physiological models.

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

• Since the training set size is not provided, the method for establishing its ground truth is also not described in this document. If ArteVu's algorithm involved machine learning, it's highly probable that similar methods (i.e., reference standard measurements from compliant sphygmomanometers) would have been used for training data as for the test data.

Ask a specific question about this device

The Automatic Blood Pressure Monitor is designed to measure blood pressure (systolic and diastolic) and pulse rate in adult patients with arm circumference range between 7.1 inches (18.0 cm) to 16.5 inches (42 cm).

The Automatic Blood Pressure Monitor (models: CH-S693L, CH-B601L, CP-B01, CH-S603, CH-S691L, CH-B607, CH-B606, CH-S692L, CH-S602, CH-W701L) is a kiosk-type, automated, single upper-arm cuff oscillometric BP monitor developed for measurement of BP and pulse rate in healthcare facility/hospital or at home.

It is designed for BP measurements on either the right or left upper arm and has a fixed tubular (Arm barrel) opening to insert the user's arm, with an integral single-arm cuff, which when inflated surrounds the upper arm. It is suitable for arm circumference range 18~42 cm. The device has an elbow groove to ensure correct positioning of the arm and measures BP during inflation. A wide LED screen presents systolic and diastolic BP, heart rate and time of measurement. After the user pushes the start button, the cuff is inflated automatically by an internal pump, the systolic and diastolic blood pressures are determined by oscillometric method.

Principle of operation:

The product uses the Oscillometric Measuring method to detect blood pressure.

When the user presses the "START" button to initiate the measurement, the winding mechanism, driven by a geared motor, begins to operate. It stops winding when it starts to encounter resistance from the arm. At this point, the cuff is adapted to the arm size. Subsequently, the cuff is automatically inflated by an internal pump to reach a pressure above systolic pressure, no blood flow occurs through the artery. As the cuff is deflated below the systolic pressure, the reducing pressure exerted on the artery allows blood to flow through it and sets up a detectable vibration in the arterial wall. When the cuff pressure falls below the patient's diastolic pressure, blood flows smoothly through the artery in the usual pulses, without any vibration being set up in the wall. Vibrations occur at any point where the cuff pressure is sufficiently high that the blood has to push the arterial wall open in order to flow through the artery. The vibrations are transferred from the arterial wall, through the air inside the cuff, into a transducer in the monitor that converts the measurements into electrical signals. Hence when it starts inflating the arm cuff, meanwhile, the unit detects pressure oscillations generated by beat-to-beat pulsatile, which is used to determine the systolic and diastolic pressure, and pulse rate.

This document is a 510(k) clearance letter for an Automatic Blood Pressure Monitor, which means it describes a non-invasive blood pressure measurement system, not an AI/ML device in the context of the requested questions. The provided text is a standard FDA 510(k) summary for a traditional medical device and does not contain information about AI model performance, expert ground truth adjudication (for AI), or MRMC studies.

Therefore, many of the requested fields are not applicable or cannot be extracted from the provided text because they pertain to AI/ML device studies, which is not what this document describes.

However, I can extract information related to the device's performance based on the provided clinical accuracy testing for the Automatic Blood Pressure Monitor.

Acceptance Criteria and Device Performance (Non-AI/ML Device):

The device's clinical accuracy was tested according to ISO 81060-2 Third edition 2018-11 Amendment 1 2020-01. The document explicitly states:

"The test data showed the clinical accuracy of the subject device complied with the requirements of ISO 81060-2 Third edition 2018-11 Amendment 1 2020-01."

This standard outlines the requirements for clinical investigations of intermittent automated measurement type non-invasive sphygmomanometers. While the specific numerical performance metrics (e.g., mean difference and standard deviation of differences between device and reference measurements) are not quantitatively stated in the summary table, the statement of compliance confirms that the device met the acceptance criteria defined by this international standard for blood pressure measurement accuracy.

The summary also specifies instrumental accuracy for blood pressure and pulse rate:

• Blood Pressure Measurement Accuracy: ±3 mmHg
• Pulse rate measurement accuracy: ±5%

Table of Acceptance Criteria and Reported Device Performance (as inferred from the document):

Regarding the specific questions about AI/ML studies:

1. A table of acceptance criteria and the reported device performance: See table above. More specific quantitative clinical performance results (e.g., mean difference and standard deviation of differences of BP readings compared to reference) are not explicitly detailed in the provided summary but are implied by the compliance statement.
2. Sample size used for the test set and the data provenance:
   • Test Set Sample Size: 85 subjects.
   • Data Provenance: Not specified (e.g., country of origin). It's a clinical trial, implying prospective data collection for the validation study.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts: Not applicable. This is for a traditional blood pressure monitor validated against a reference sphygmomanometer (aneroid/auscultation method), not an AI/ML device requiring expert consensus for ground truth. The reference device was a "CM-BPM-D Aneroid sphygmomanometer" by Shanghai Caremate Medical Device Co. Ltd.
4. Adjudication method (e.g., 2+1, 3+1, none) for the test set: Not applicable for this type of device and study. The accuracy is determined by comparison to the reference measurements.
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 a traditional medical device, not an AI/ML product assisting human readers.
6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done: The device itself is a standalone automatic blood pressure monitor. Its performance is measured directly, not as an algorithm's output to be interpreted by a human.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc): The ground truth for the clinical accuracy study was established by measurements from an "Aneroid sphygmomanometer" using the "Aneroid/auscultation method." This is the established reference method for validating automatic BP monitors per ISO 81060-2.
8. The sample size for the training set: Not applicable. This is a traditional device; there is no "training set" in the AI/ML sense. The device's measurement algorithm is predetermined and fixed.
9. How the ground truth for the training set was established: Not applicable. No AI model training set.

Ask a specific question about this device

Electronic Sphygmomanometers is intended to measure the systolic and diastolic blood pressure of adult person and adolescents age 18 through 21 years of age. It can be used at home. It is contraindicated in pregnant women, including those with preeclampsia.

The Electronic Sphygmomanometers, including ZH-X9, ZH-X12, ZH-X16, ZH-X17, ZH-X18, ZH-X19, ZH-X23, ZH-X24, is suitable for measurement of systolic blood pressure and diastolic blood pressure of adult person and adolescents 18 to 21 years old with arm circumference ranging from 22 cm to 42 cm by the oscillometric technique. The error is controlled within the range specified in IEC 80601-2-30 Non-invasive automated monitor. User can select the blood pressure unit mmHg or kPa. The initial inflation pressure of the cuff is zero pressure. When start the device, the cuff will be inflated and deflated.

The device consists of the microprocessor, pressure sensor, operation keys, pump, deflation control valve, LCD screen and cuff. And all models are powered by 4 AAA dry batteries (DC 6V).

The device has a memory function that automatically stores 2*99 sets data of the latest measurements. It can also display the latest measurement result. Additionally, the device also can read the data through voice broadcast function.

The seven models have the same intended use, working principle, measuring range, accuracy, cuff, and conformance standard; only appearance have some difference.

The provided FDA 510(k) clearance letter and synopsis for the Electronic Sphygmomanometers (K250185) primarily focus on demonstrating substantial equivalence to a predicate device, rather than detailing a specific clinical study with granular acceptance criteria for an AI-powered diagnostic device. The application is for a standard medical device (blood pressure monitor), not an AI/ML-driven diagnostic tool. Therefore, many of the requested points for an AI study (e.g., number of experts for ground truth, adjudication methods, MRMC studies, effect size of AI assistance, training set details) are not applicable or provided in this document.

However, I can extract and infer information relevant to the device's accuracy and performance validation using the provided text, particularly concerning the mentioned standards and the comparative clinical study.

Here's an attempt to describe the acceptance criteria and study as best as possible given the provided non-AI-specific document:

Device: Electronic Sphygmomanometers (ZH-X9, ZH-X12, ZH-X16, ZH-X17, ZH-X18, ZH-X19, ZH-X23, ZH-X24)

Study Type: Comparative Clinical Study (for substantial equivalence to a predicate device and mercury sphygmomanometer) and Non-Clinical Performance Testing.

Regulation/Standard for Accuracy: IEC 80601-2-30, ISO 81060-2 (including Amendment 1(2020)), and FDA Guidance No-Invasive Blood Pressure (NIBP) Monitor Guidance.

1. Acceptance Criteria and Reported Device Performance

The core acceptance criterion for non-invasive blood pressure monitors, as per ISO 81060-2 (which is explicitly listed as a standard the device complies with), relies on statistical analysis of difference between the device's readings and a reference standard (usually mercury sphygmomanometer readings or an established method). The key metrics are:

• Mean Difference (Bias): The average difference between the device reading and the reference reading.
• Standard Deviation of the Differences: A measure of the spread of these differences.

The ISO 81060-2 standard (for clinical validation of automated measurement type) typically requires:

• Mean difference between the test device and reference measurement: $\leq \pm 5$ mmHg
• Standard deviation of the differences (SD): $\leq 8$ mmHg

Note: The reported accuracy of $\pm 3$ mmHg or 2% of the reading for pressure is a general accuracy specification, not necessarily the mean difference and standard deviation from the clinical validation as per ISO 81060-2. However, achieving $\pm 3$ mmHg as a general accuracy implies that the device is likely to meet or exceed the ISO 81060-2 criteria. The document states compliance with ISO 81060-2, meaning it passed these criteria.

2. Sample Size and Data Provenance

• Test Set (Clinical Study): The document states, "We conducted a comparative clinical study to verify the performance of the proposed device and predicate device as well as a mercury sphygmomanometer according to ISO 81060-2."
  • Sample Size: ISO 81060-2 typically requires a minimum of 85 subjects for clinical validation. The document does not explicitly state the exact sample size used in their study, but implies it met the requirements of the standard it followed.
  • Data Provenance: Not specified (e.g., country of origin). The study is described as a "comparative clinical study," which generally implies a prospective data collection for the purpose of the validation.

3. Number/Qualifications of Experts for Ground Truth

• Not Applicable / Not Specified: For a blood pressure monitor, the "ground truth" for blood pressure measurements is typically established using a reference standard like a mercury sphygmomanometer or an auscultatory method performed by trained observers, as outlined in ISO 81060-2. This is a measurement comparison, not an expert-based image interpretation or diagnosis that would require multiple reading experts in the AI sense.

4. Adjudication Method

• Not Applicable / Not Specified: Since this is a direct physiological measurement comparison, adjudication methods commonly seen in AI/image interpretation studies (e.g., 2+1, 3+1 consensus) are not relevant here. The ISO 81060-2 standard defines precise measurement protocols by trained observers for comparison.

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

• Not Applicable: This type of study is relevant for AI-assisted diagnostic aids where human readers' performance (e.g., accuracy, efficiency) is evaluated with and without AI. This device is a standalone measurement tool, not an AI diagnostic aid for human readers.

6. Standalone (Algorithm Only) Performance

• Implicitly Done: The "Performance Data" section explicitly states compliance with "IEC 80601-2-30" and "ISO 81060-2." These standards define the requirements for the automated (standalone) measurement of non-invasive blood pressure, requiring the device algorithm (oscillometric technique) to provide accurate pressure readings independently. The clinical study compares the device's automated readings against a reference standard.

7. Type of Ground Truth Used

• Reference Standard Measurement: The ground truth for blood pressure measurements was established using a mercury sphygmomanometer (or an equivalent valid reference method) during the comparative clinical study, as per the ISO 81060-2 standard. This is a direct, objective physiological measurement.

8. Sample Size for the Training Set

• Not Applicable: This is a traditional medical device (blood pressure monitor) that utilizes a deterministic oscillometric algorithm, not a machine learning or AI model that requires a "training set" in the context of deep learning. Its algorithm is based on established engineering principles for detecting pressure oscillations.

9. How Ground Truth for Training Set Was Established

• Not Applicable: As there is no "training set" for a machine learning model, this point is not relevant for this device.

Ask a specific question about this device

The Aktiia G0 Blood pressure monitoring system consists of a wrist worn monitor and an oscillometric cuff. The system is intended for measuring blood pressure and pulse rate based on using a Pulse Wave technique in adults aged from 22 to 59 years old with wrist circumference ranging from 14 cm to 21 cm following a calibration process (every 24 hours) using the oscillometric blood pressure cuff.

The Aktiia G0 Blood pressure monitoring system is intended for spot-checking of adult patients for home use.

The Aktiia G0 Blood Pressure Monitoring System measures blood pressure and pulse rate (spot check) based on the analysis of Photoplethysmography (PPG) signals. Aktiia G0 Blood Pressure Monitoring System consists of the following components:

• Bracelet
• Charger (for the Bracelet)
• Aktiia Init I1 (cuff)
• Mobile App
• Backend Software and Algorithm (in the Cloud)

The Aktiia G0 Blood Pressure Monitoring System hardware component, referred to as the bracelet, is responsible for PPG data acquisition on the user's wrist. It is composed of a data logging unit called the pod and of a detachable strap that is intended to secure the pod to the user's wrist. The pod uses the bracelet's electronics and PPG sensor. The bracelet's internal battery is located inside the pod and is recharged using a pin to USB docking station using a provided charger.

A calibration process (also referred to as initialization) is required prior to converting blood pressure values from optical data. This calibration process uses reference blood pressure values measured with an oscillometric blood pressure monitor (cuff) also referred to as Aktiia Init I1.

The Aktiia G0 Blood Pressure Monitoring System includes a mobile application for displaying data to the user and uses a cloud server referred to as the backend for data storage. The backend also hosts the algorithm which converts optical data generated by the bracelet into blood pressure and pulse rate data.

The FDA 510(k) clearance letter for the Aktiia G0 Blood Pressure Monitoring System (K250415) provides details on the acceptance criteria and the study conducted to prove the device meets these criteria.

Here's a breakdown of the requested information:

1. Table of Acceptance Criteria and Reported Device Performance

The core acceptance criterion for blood pressure measurement accuracy for both the G0 System and the Aktiia Init I1 cuff is compliance with ISO 81060-2. For pulse rate, a specific RMSE threshold is provided.

2. Sample Sizes and Data Provenance

• Test Set Sample Size:
  • Blood Pressure: The document states that "Clinical validations of blood pressure per ISO 81060-2:2018 were performed." ISO 81060-2:2018 generally requires a minimum of 85 subjects for accuracy testing, evenly distributed across specific blood pressure ranges. While the exact number isn't explicitly stated for the BP validation itself, the pulse rate study involved 85 participants, and it's highly likely that the blood pressure validation used the same or a very similar cohort given it was also a clinical investigation aiming for ISO compliance.
  • Pulse Rate: 85 participants were included in the final dataset for the pulse rate validation study (NCT06565780), with 1273 simultaneous pulse rate measurements collected.
• Data Provenance: The study aimed for "a representative US population cohort," which implies the data originated from the United States. The study was a prospective clinical investigation (NCT06565780).

3. Number of Experts and Qualifications for Ground Truth

The document does not explicitly state the number of experts or their specific qualifications for establishing the ground truth for the blood pressure measurements. However, for blood pressure, the ground truth reference method was double auscultation. This method inherently involves human observers (typically trained medical professionals) taking readings.

For pulse rate, the ground truth was an electrocardiogram (ECG), which is a gold standard instrumental measurement and does not inherently require expert human interpretation for establishing pulse rate.

4. Adjudication Method for the Test Set

• Blood Pressure: The ground truth was established by "double auscultation." This method typically involves two independent observers taking blood pressure readings, and potentially a third if there is a significant discrepancy. The document does not specify the exact adjudication protocol (e.g., 2+1, 3+1), but conformity to ISO 81060-2 implies a standardized, typically adjudicated, process for reference measurements.
• Pulse Rate: The ground truth was an ECG. Adjudication for instrumental measurements like ECG usually refers to the process of interpreting or reviewing the ECG waveform, but for simple pulse rate determination, the automated measurement from the ECG device is considered the ground truth. There is no mention of human adjudication for the ECG-derived pulse rate.

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

No, a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not explicitly mentioned for the Aktiia G0 Blood Pressure Monitoring System. The device is a "blood pressure monitoring system," not an interpretation device like an AI for medical imaging. The evaluation focuses on the accuracy of the device's measurements against established reference methods, not on how human readers' performance is improved by AI assistance in diagnosis or interpretation.

6. Standalone (Algorithm Only) Performance

Yes, a standalone performance evaluation was conducted. The clinical validation studies (ISO 81060-2 for BP, NCT06565780 for PR) assessed the accuracy of the Aktiia G0 system's measurements (algorithm output) against reference methods (double auscultation for BP, ECG for PR).

7. Type of Ground Truth Used

• Blood Pressure: Expert consensus via double auscultation (a highly standardized and accepted clinical method for reference blood pressure measurement).
• Pulse Rate: Instrumental data (electrocardiogram - ECG), which is considered a gold standard for heart rate measurement.

8. Sample Size for the Training Set

The document does not provide information regarding the sample size used for the training set of the algorithm. It only details the clinical validation (test) set.

9. How Ground Truth for the Training Set Was Established

The document does not provide information on how the ground truth for the training set was established. It only details the establishment of ground truth for the test set used in the clinical validation studies. Given that the algorithm is described as "Pulse-Wave Analysis (PWA)," it would conceptually involve matching optical signal data to simultaneously acquired reference blood pressure measurements (likely oscillometric cuff or arterial line) to train the model, but these specifics are not in the provided text.

Ask a specific question about this device

The Arm-type Fully Automatic Digital Blood Pressure Monitors are intended to measure blood pressure (systolic and diastolic) and pulse rate of adults and adolescents over 12 years of age with circumference ranging from 22cm to 42cm and 40cm-56cm.

The Arm-type Blood Pressure Monitor (BPM) series is an automatic, non-invasive, blood pressure measurement system for over-the-counter (OTC) use in home and clinical environment. The systolic and diastolic pressures are determined using the Oscillometric method, where the cuff is inflated with an integral controllable piezoelectric pump and deflates via an electric automatic rapid deflation valve. During measurements, an electric pump within the main unit slowly inflates the arm cuff, generating cuff pressure which is monitored and from which pulse waveform data is extracted. This waveform data is analyzed by software algorithms within the microprocessor to determine pulse rate, systolic pressure, and diastolic pressure. The cuff can measure pressure range from 0 to 299mmHg, and the pulse rate range from 30 to 180 beats per minute.

The DBP-6286B embed bluetooth module to transfer data to APP. With the use of software (including APP) and Bluetooth communication module, the wireless software function and hardware function are solely intended to transfer, store, convert formats, or display medical device data and results (blood pressure and pulse rate readings), without controlling or altering the functions or parameters of any connected medical devices, which is not be intended for active patient monitoring, therefore, based on the FDA guidance titled "Medical Device Data Systems, Medical Image Storage Devices, and Medical Image Communications Devices" (issued on September 28, 2022.), this software function is belong to Non-device-MDDS.

Based on the provided FDA 510(k) clearance letter for the JOYTECH Healthcare Co., Ltd. Arm-type Fully Automatic Digital Blood Pressure Monitors (DBP-6286B, DBP-6186), here's a description of the acceptance criteria and the study proving the device meets them:

Acceptance Criteria and Device Performance for Blood Pressure Monitors

The acceptance criteria for blood pressure monitors are typically defined by recognized international standards, primarily ISO 81060-2:2018+AMD2020, which pertains to the clinical investigation of intermittent automated measurement type for non-invasive sphygmomanometers. This standard specifies the accuracy requirements for blood pressure measurements.

The key acceptance criterion is that the accuracy of the blood pressure monitor must be within the acceptable scope specified in ISO 81060-2:2018+AMD2020. While the exact numerical thresholds for group means and standard deviations aren't explicitly stated in this clearance letter, the general requirement is clear.

The reported device performance confirms that the device meets these criteria.

Study Details Proving Device Meets Acceptance Criteria:

1. A table of acceptance criteria and the reported device performance: (See table above)

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

   • Cuff (22-42cm): 91 valid subjects' data were used for analysis.
   • Cuff (40-56cm): 85 valid subjects' data were used for analysis.
   • Data Provenance: The document does not explicitly state the country of origin of the data, but it implies a prospective clinical study ("All the subjects were volunteer to take part in the clinical study, all the subjects completed the clinical study without any AE or side-effect").

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

   • The document implies the use of human observers for the reference measurements, as it states: "The manual Mercury Sphygmomanometer was used as a reference device." This suggests that the ground truth was established by human operators taking readings from a mercury sphygmomanometer against which the automated device's readings were compared. However, the exact number and qualifications of these "experts" (human observers) are not specified in the provided document.

4. Adjudication method for the test set:

   • The document states, "Same arm sequential method was adopted during clinical testing." This method involves comparing simultaneous or near-simultaneous measurements from the test device and a reference device on the same arm. While this describes the measurement procedure, it does not describe an adjudication method for discrepancies between readings, as would be typical for expert consensus. Given that the ground truth is a direct measurement from a mercury sphygmomanometer, an adjudication process in the traditional sense (e.g., for image interpretation) is not directly applicable.

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. This is a medical device clearance for a blood pressure monitor, not an AI-assisted diagnostic imaging device. Therefore, an MRMC study and analysis of human reader improvement with AI assistance are irrelevant and were not performed.

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

   • The study design (clinical validation against a reference standard) inherently evaluates the "standalone" performance of the automated blood pressure monitor's algorithm. The device itself is designed to operate without continuous human intervention during the measurement process, making its performance essentially "algorithm only" in terms of blood pressure determination from the oscillometric data.

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

   • The ground truth was established by direct measurement using a manual Mercury Sphygmomanometer. This is considered the clinical gold standard for non-invasive blood pressure measurement in this context.

8. The sample size for the training set:

   • The document does not specify a separate training set or its sample size. For traditional medical devices like blood pressure monitors, the "training" (development and calibration) often occurs during the device engineering and manufacturing phases, and clinical validation is then performed on a separate, independent test set as described.

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

   • As no specific "training set" is mentioned in the context of a separate clinical dataset, the method for establishing ground truth for such a set is not provided. The device's underlying algorithms would have been developed and refined through engineering and calibration processes, likely using extensive internal testing and validation against standardized pressure sources and reference measurements, rather than a distinct "clinical training set" in the AI/ML sense.

Ask a specific question about this device

The Wrist Blood Pressure Monitor is intended to measure the systolic pressure and diastolic pressure, as well as the pulse rate of adult person via non-invasive oscillometric technique by an inflatable cuff wrapped around the wrist at medical facilities or at home.

The Wrist Blood Pressure Monitor is designed as a battery driven automatic non-invasive blood pressure monitor. It can automatically complete the inflation, deflation and measurement, which can measure systolic and diastolic blood pressure as well as the pulse rate of adult person at wrist within its claimed range and accuracy via the oscillometric technique. The result will be displayed in the international unit mmHg or Kpa.

All the models included in this submission follow the the same intended use, same measurement principle, same blood pressure core algorithm and similar product design. All the models can be used with one cuff size 13.5~19.5 cm (5.3-7.7inches).

The main differences are appearance, Dimensions and some specifications which will not affect the safety and effectiveness of the device.

The provided FDA 510(k) clearance letter and associated summary pertain to a Wrist Blood Pressure Monitor, which is a medical device for measuring blood pressure and pulse rate. It is not an AI/Software as a Medical Device (SaMD). Therefore, many of the typical acceptance criteria and study designs associated with AI/SaMD (such as multi-reader multi-case studies, ground truth establishment by experts, training set details, or effect sizes of AI assistance) are not applicable to this device.

The acceptance criteria and study details provided are tailored to the performance of a non-invasive blood pressure measurement system (hardware device), focusing on accuracy, safety, and effectiveness.

Here's a breakdown of the requested information based on the provided document, addressing the device's specific characteristics as a hardware blood pressure monitor:

Acceptance Criteria and Device Performance (Wrist Blood Pressure Monitor)

1. Table of Acceptance Criteria and Reported Device Performance

As per the 510(k) summary, the device's accuracy is a key performance metric. The acceptance criteria are based on the international standard ISO 81060-2 Third edition 2018-11 [Including AMD1:2020].

2. Sample Size Used for the Test Set and Data Provenance

• Sample Size for Test Set (Clinical Accuracy Study):
  • Three groups of clinical accuracy research were conducted. Each group included 100 subjects, for a total of 300 subjects across the 13 models.
  • Group 1: 100 subjects (47 Male, 53 Female)
  • Group 2: 100 subjects (54 Male, 46 Female)
  • Group 3: 100 subjects (44 Male, 56 Female)
  • Minimum subjects for each group was 85, as per ISO 81060-2.
• Data Provenance: The document does not explicitly state the country of origin for the clinical data. However, the manufacturer is "Shenzhen AOJ Medical Technology Co., Ltd." in Shenzhen, Guangdong, China. It is highly probable the data was collected in China.
• Retrospective or Prospective: The clinical accuracy study, designed to meet ISO 81060-2, is typically conducted prospectively to collect new data for device validation. The wording "clinical accuracy research" and "clinical accuracy test report and data analysis followed the requirements" implies a prospective study.

3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications of Those Experts

• This question is not directly applicable in the context of this device because "ground truth" for a blood pressure monitor's accuracy is established against a reference standard, not through expert consensus on interpretations of images or signals (as would be the case for AI/SaMD).
• For blood pressure monitors, the "ground truth" or reference measurement is typically taken by trained medical professionals using a standardized reference sphygmomanometer (e.g., mercurial or auscultatory method), as per the ISO 81060-2 standard. The document states "The Same Arm Sequential Method was chosen for all studies," which is a standard procedure comparison method against a reference device. The qualifications of the individuals performing these reference measurements would be trained clinicians (e.g., physicians, nurses).

4. Adjudication Method for the Test Set

• This question is not applicable for a blood pressure monitor's accuracy testing. Adjudication methods (like 2+1 or 3+1) are used to resolve discrepancies in human expert interpretations, especially in image-based diagnostics.
• For blood pressure accuracy, deviations are quantified statistically between the device reading and the reference measurement, not through an adjudication process among multiple "readers."

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, this was not done. This type of study is relevant for AI/SaMD devices where AI assists human interpretation and is a key component for assessing the AI's clinical utility. The Wrist Blood Pressure Monitor is a standalone hardware device that provides a measurement; it does not involve human "readers" interpreting data or AI assistance.

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

• Yes, in essence. The entire clinical accuracy study described in Section 8 ("Clinical data") evaluates the device's performance (which incorporates its internal oscillometric algorithm) in a standalone manner against a reference standard. The "algorithm" here refers to the embedded software that processes the oscillometric signals to derive blood pressure and pulse rate. The study directly assesses how accurately the device (with its integrated algorithm) measures blood pressure readings compared to the reference.
• Performance Metrics: The evaluation was based on the "mean error and standard deviation of differences for systolic, diastolic pressure" as per ISO 81060-2.

7. The Type of Ground Truth Used

• The ground truth for the clinical accuracy testing was established through concurrent measurements using a standardized reference method (e.g., auscultatory method with a mercurial sphygmomanometer or another validated reference device) on the same arm, sequentially with the test device. This is the standard for blood pressure monitor validation as per ISO 81060-2.
• It is not "expert consensus" in the sense of subjective medical interpretation, but rather an objective, standardized measurement performed by trained personnel using a calibrated reference instrument.

8. The Sample Size for the Training Set

• This concept is not applicable to this type of medical device clearance. The Wrist Blood Pressure Monitor is a hardware device with an embedded algorithm (oscillometric technique) that is based on established physiological principles. It doesn't use machine learning or deep learning in a way that requires a separate "training set" of patient data for an AI model to learn from, as would be the case for AI/SaMD devices. The device's "training" (development and calibration) would involve engineering principles and laboratory testing, rather than a data-driven machine learning process.

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

• As the concept of a "training set" in the context of machine learning is not applicable here (see point 8), the establishment of ground truth for such a set is also not applicable. The device's underlying measurement principle is well-established oscillometric technology. Development and calibration rely on physical models and engineering validation.

Ask a specific question about this device

The Arm Blood Pressure Monitor is intended to measure the systolic and diastolic blood pressure as well as the pulse rate of adult person via non-invasive oscillometric technique in which an inflatable CUFF is wrapped around the upper arm at medical facilities or at home.

The Arm Blood Pressure Monitor is designed as a battery driven automatic non-invasive blood pressure monitor. It can automatically complete the inflation, deflation and measurement, which can measure systolic and diastolic blood pressure as well as the pulse rate of adult person at upper arm within its claimed range and accuracy via the oscillometric technique. The result will be displayed in the international unit mmHg or Kpa.

The device has the data storage function in order for data reviewing, including the systolic pressure, diastolic pressure, pulse rate and measurement time. The device also has low voltage indication, which will be triggered when the battery is low.

The FDA 510(k) clearance letter for the Arm Blood Pressure Monitor does not explicitly state acceptance criteria in a quantitative table or the specific performance metrics achieved in a typical "device performance" section as one might expect for a software-based or diagnostic device. Instead, it refers to compliance with established medical device standards, particularly ISO 81060-2:2018 (and its amendment from 2020) for clinical accuracy.

However, based on the context of the clinical data and the requirements of ISO 81060-2, we can infer the acceptance criteria for accuracy.

Here's a breakdown of the requested information based on the provided document:

Implied Acceptance Criteria and Reported Device Performance

The device's performance is demonstrated by its adherence to the clinical accuracy requirements of ISO 81060-2 Third edition 2018-11 (including AMD1:2020).

Implied Acceptance Criteria of ISO 81060-2:2018 (and AMD1:2020):

The ISO 81060-2 standard consists of two parts for accuracy assessment:

• Criterion 1 (Mean Difference and Standard Deviation):
  • The mean difference between the device and the reference measurement for both systolic and diastolic blood pressure shall be $\leq \pm 5$ mmHg.
  • The standard deviation of these differences shall be $\leq 8$ mmHg.
• Criterion 2 (Cumulative Percentage of Differences):
  • For each subject, the absolute difference between the device and the reference measurement should be calculated.
  • The cumulative percentage of devices should have absolute differences within certain thresholds:
    • $\leq 5$ mmHg for at least 65% of measurements.
    • $\leq 10$ mmHg for at least 85% of measurements.
    • $\leq 15$ mmHg for at least 95% of measurements.

Table of Acceptance Criteria and Reported Device Performance (Inferred from ISO 81060-2)

Note: The document explicitly states, "All data's mean error and standard deviation of differences for systolic, diastolic pressure is not over the limits of ISO 81060-2 Third edition 2018-11 [Including AMD1:2020]." This confirms the device met the numerical precision requirements of the standard for mean difference and standard deviation. While it doesn't provide the exact numbers for each group, it confirms compliance with the key statistical thresholds.

Study Details

1. Sample Size Used for the Test Set and Data Provenance:

   • Total Sample Size: The 28 models were grouped into 7 categories for clinical accuracy testing. Each of these 7 groups had a test set of 100 subjects, except for Group 7, which had 92 subjects.
     • Group 1: 100 subjects
     • Group 2: 100 subjects
     • Group 3: 100 subjects
     • Group 4: 100 subjects
     • Group 5: 100 subjects
     • Group 6: 100 subjects
     • Group 7: 92 subjects
   • Male/Female Distribution: Each group had a participant distribution of at least 30% male and at least 30% female.
   • Age Range: Subjects were "adult person" with an age range of "> 12 years old," specifically noted ranges from 15 to 80 years old across the groups.
   • Data Provenance: The document does not explicitly state the country of origin for the clinical data. It also does not explicitly state if the study was retrospective or prospective, but clinical accuracy studies adhering to ISO 81060-2 are typically prospective and involve simultaneous, sequential measurements.

2. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications of Those Experts:

   • The document does not specify the number or qualifications of experts (e.g., radiologists) for establishing ground truth in this context. For blood pressure monitors, "ground truth" is typically established by trained technicians or clinicians using a validated reference method (e.g., mercury sphygmomanometer) following a rigorous protocol (like the auscultatory method per ISO 81060-2). The standard requires independent observers for reference measurements.

3. Adjudication Method for the Test Set:

   • The document states, "The clinical accuracy test report and data analysis followed the requirements of the ISO 81060-2 Third edition 2018-11 [Including AMD1:2020]. The Same Arm Sequential Method was chosen for all studies."
   • ISO 81060-2 outlines a specific method for ground truth establishment for blood pressure. It typically involves multiple independent observers (at least two trained observers) who simultaneously or sequentially measure blood pressure using a validated reference device. Discrepancies between these observers are resolved in accordance with the standard, which usually involves specific rules for averaging or discarding measurements based on predefined variability criteria, rather than a formal "adjudication panel" in the sense of image interpretation.

4. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done:

   • No, an MRMC comparative effectiveness study was not done. This type of study is specifically relevant for diagnostic imaging AI devices where human readers interpret medical images with and without AI assistance. For a non-invasive blood pressure monitor, the study focus is on the device's measurement accuracy against a reference standard, not on how it assists human interpretation.

5. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done:

   • Yes, in essence, a "standalone" or "algorithm only" performance was evaluated. The device (Arm Blood Pressure Monitor) is an automated system that measures blood pressure. Its clinical accuracy study measures its direct output (systolic and diastolic pressure) against a reference standard. There isn't a "human-in-the-loop" component in the operational measurement process that would affect its core accuracy, as it's an automated device.

6. The Type of Ground Truth Used:

   • The ground truth was established by comparison with a validated reference measurement method, as prescribed by ISO 81060-2. This typically involves the auscultatory method using a mercury sphygmomanometer or an equivalent validated reference device, with measurements performed by trained observers.

7. The Sample Size for the Training Set:

   • Not applicable / Not provided. This device is a hardware-based blood pressure monitor that uses an oscillometric technique and a "blood pressure core algorithm." It is not described as a machine learning or AI algorithm that requires a separate "training set" in the conventional sense of deep learning. Its performance is based on its engineering design and the embedded validated algorithm.

8. How the Ground Truth for the Training Set Was Established:

   • Not applicable / Not provided. As mentioned above, there isn't a "training set" in the common understanding for a typical medical AI device. The core algorithm is likely developed and validated internally by the manufacturer through engineering principles and extensive testing, not necessarily through a separate clinical "training set" with ground truth in the AI context. The clinical data presented is for validation (test set) of the final device, demonstrating its accuracy.

Ask a specific question about this device