Search Results

The APK SpO2 Pulse Oximeter Sensor (Resuable) is intended for spot checking or continuous monitoring of functional arterial oxygen saturation and pulse rate in non-invasive with oximeter equipment, which are intended for adult or pediatric patients (more than 5kg) in hospital-type facilities, and home environments.

The APK SpO2 Pulse Oximeter sensor (Disposable) is intended for single use when continuous noninvasive arterial oxygen saturation and pulse rate monitoring are adults or pediatric patients (more than 5kg) in hospital-type facilities, and home environments.

The applicant sensor APK SpO2 Pulse Oximeter Sensor (Reusable and Disposable) measure, non-invasively, the arterial oxygen saturation of blood. The measurement method is based on the red and infrared light absorption of hemoglobin and oxyhemoglobin. Light of a red and infrared light source is emitted through human tissue and received by a photodiode. The measurement is based on the absorption of light, which is emitted through human tissue (for example through the index finger). The light comes from two sources (red LED and infrared LED) with different wavelengths and is received by a photodiode. Out of the different absorption behavior of the red and infrared light a so-called ratio can be calculated. The saturation value is defined by the percentage ratio of the oxygenated hemoglobin [HbO2] to the total amount of hemoglobin [Hb]. Those sensors contain a red and infrared light source and a photodiode receiving the non-absorbed red and infrared light. The received signals are forwarded to a measurement device that amplifies the acquired signal and an algorithm that calculates the ratio and converts via a validated calibration table the ratio to a saturation value. The device is not software-driven. The applicant sensors are not for implant. Those sensors are not sterile and do not need sterilization or re-sterilization. The device is for prescription. The device does contain drug or biological product.

Here's an analysis of the provided text regarding the acceptance criteria and study for the APK SpO2 Pulse Oximeter Sensor:

Unfortunately, the provided document is a 510(k) summary for a pulse oximeter sensor, which focuses on demonstrating substantial equivalence to a predicate device rather than providing detailed acceptance criteria and a comprehensive study report with all the requested information for a novel AI/software medical device.

Key Missing Information for AI/Software Devices:

• Quantitative Performance Metrics (Acceptance Criteria): The document states "All the information about the device performance were provided," but it does not list specific quantitative acceptance criteria (e.g., sensitivity, specificity, accuracy mean absolute error (RMSE), etc.) or the device's reported performance against such criteria. The focus is on demonstrating equivalence to the predicate, not on setting and meeting new performance thresholds.
• Sample Size and Data Provenance for Test Set: While a "Clinical Test Report following ISO 9919:2005" is mentioned, the crucial details of the test set (number of cases, data origin, retrospective/prospective) are absent in the summary. ISO 9919:2005 outlines requirements for pulse oximeters, but the specifics of this particular study are not elaborated.
• Ground Truth Establishment, Experts, and Adjudication: These details are completely omitted. For a pulse oximeter, "ground truth" for SpO2 is typically established through co-oximetry, not human expert consensus like in image-based AI.
• MRMC Study and AI-Assisted Performance: This is not applicable as the device is a hardware sensor, not an AI or software device. The document explicitly states, "The device is not software-driven."
• Standalone Performance: While a clinical test report is mentioned, it's not described as a "standalone" performance evaluation in the context of an AI algorithm, but rather a performance test of a physical sensor.
• Training Set Details: As the device is not software/AI-driven, there is no "training set" in the context of machine learning. The device relies on a "validated calibration table."

Analysis Based Only on Provided Information:

Given the nature of the device (a physical pulse oximeter sensor) and the document (510(k) summary for substantial equivalence), the requested information pertaining to AI/software device evaluation is largely absent or not directly applicable.

Here's what can be extracted and inferred from the text, with an emphasis on what is not present:

1. Table of Acceptance Criteria and Reported Device Performance

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

• Sample Size (Test Set): Not specified in the summary.
• Data Provenance: Not specified. The document indicates a "Clinical Test Report following ISO 9919:2005" was completed, which typically involves human subject testing, but details like location or whether it was retrospective/prospective are not provided.

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

• Number of Experts: Not applicable. For pulse oximetry, "ground truth" (reference SpO2) is typically established using a laboratory co-oximeter on arterial blood samples from human subjects undergoing controlled desaturation, not expert interpretation of outputs.
• Qualifications of Experts: Not applicable.

4. Adjudication Method for the Test Set

• Adjudication Method: Not applicable. Ground truth for pulse oximetry is instrumental (co-oximetry) rather than based on expert consensus requiring adjudication.

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

• Was a MRMC study done? No.
• Effect size of human readers improval with AI vs. without AI assistance: Not applicable, as this is a hardware sensor, not an AI or software device. The device is explicitly stated as "not software-driven."

6. Standalone Performance Study

• Was a standalone performance study done? A "Clinical Test Report following ISO 9919:2005" was performed, which evaluates the sensor's performance. However, this is not a "standalone (algorithm only without human-in-the-loop performance)" study in the context of an AI algorithm, as the device is not an AI algorithm. It's a performance test of the physical sensor.

7. Type of Ground Truth Used

• Type of Ground Truth: Most likely co-oximetry (measurement of arterial blood gas samples by a laboratory co-oximeter) as per standard pulse oximetry testing guidelines (e.g., ISO 9919). This is inferred, not explicitly stated beyond "Clinical Test Report following ISO 9919:2005."

8. Sample Size for the Training Set

• Sample Size (Training Set): Not applicable. The device is not an AI/ML-driven device that requires a training set. It uses a "validated calibration table."

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

• How Ground Truth Was Established (Training Set): Not applicable, as there is no training set for this type of device. The "validated calibration table" would have been developed and validated through similar clinical testing methods using co-oximetry as ground truth.

Ask a specific question about this device