FDA 510(k) Search - By Innolitics (SaMD and AI Experts)

To spot check or monitor Oxygen saturation of arterial hemoglobin (SpO2%) and pulse rate.
To analyze the pulse waveform (Photoelectrical Plethysmography or PP) provided by the oximeter. It only provides mathematical analyses of the input of the SpO2 measurement.
To analyze the basic rhythms of the NN or RR intervals in heart rate, both in the time domain and in the frequency domain (short time 5 minutes). It only provides mathematical analyses of the input of the heart rate variability.
The mathematical analysis of Photoelectrical Plethysmography and HRV are not intended use for diagnosis.
The software provides a visual alarm for the values of the heart rate and/or SpO2 percent out of the normal range and for the bad quality signal transmission.
The data are stored in PC in the Backup system of the ESO software.
The device is intended use only for adult subjects (> 20 years old)
This Oximeter is intended to be used in spot-checking (5 minutes)
The device is intended use in licensed practitioner's office
This device is no intended to be used at home, in hospital or out-of-hospital transport
The device is not intended use in support life and not for continuously monitoring
The system will be use by practitioner.

Device Description

USB plug and play hardware device including an electronic box and reusable Adult SpO2 probe
Software installed on a computer.
Description of the features
Displays SpO2%, pulse rate value and vertical bar graph pulse amplitude.
Mathematical analysis of the pulse waveform (photoelectrical Plethysmograph' feature).
Mathematical analysis of the Heart Rate Variability (HRV 'feature).

AI/ML Overview

The provided 510(k) submission for the Electro Sensor Oxi (E.S.O) does not contain detailed acceptance criteria or a study proving that the device meets specific performance criteria related to accuracy of SpO2, pulse rate, or the mathematical analyses for Photoelectrical Plethysmography (PP) and Heart Rate Variability (HRV).

Instead, the submission largely focuses on establishing substantial equivalence to a predicate device (Electro Sensor Teck Model PEMS, K083229) by demonstrating similar intended use, technological characteristics, and safety/effectiveness, with the main new characteristic being the derivation of heart rate from oximeter waveform analysis rather than an EKG device.

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

1. Table of Acceptance Criteria and Reported Device Performance

Acceptance Criteria	Reported Device Performance
No explicit numerical acceptance criteria for SpO2 accuracy, pulse rate accuracy, or the mathematical analyses (PP, HRV) are provided in this document.	- Pulse Oximeter Functions (SpO2, Pulse Rate): The submission states that the OEM subcontractor provided the oximeter, sensor, cable connections and software code source, and "they were cleared for use together." It does not provide specific accuracy data for SpO2 or pulse rate against a reference standard or a specified range of acceptance.
	- Photoelectrical Plethysmography (PP) Analysis & Heart Rate Variability (HRV) Analysis: These are described as "only mathematical analyses" and are "not intended use for diagnosis." The document states "no clinical validation is required for E.S.O." for these features. It further mentions "peer reviews for the photoelectrical plethysmography mathematical analysis" and "comparative testing between EKG RR intervals (from predicate device) and SP02 RR intervals (from E.S.O) in one hand, and in second hand, comparative HRV analysis results from the predicate device and E.S.O." However, no quantitative performance metrics or acceptance criteria for these comparisons are provided.
General Safety & Effectiveness: Device is substantially equivalent to the predicate device in terms of performance, technology, safety, and efficacy.	- General Safety: "The fact to remove the EKG PCB does not change the general safety to the legally marketed predicate device as shown in the laboratory tests reports of the E.S.O. system (IEC 60601-1-2 and IEC 60601-1-1)." (This implies meeting these general safety standards, but these are not performance criteria for diagnostic accuracy).

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

Test Set Sample Size: Not specified for any performance evaluation.
Data Provenance:
- The document mentions "input data from the MIT-BIH database" for "Software algorithms tests." This suggests a retrospective dataset, likely from North America (MIT-BIH is a US-based database). However, it's not explicitly stated if this database was used for the "test set" in the context of device performance, but rather for software verification.
- For the comparative testing of RR intervals and HRV, no details about where this data came from are provided.
- The document references "published clinical investigations" for interpretation of PP and HRV results, stating these investigations "had been made with materials using the same technology (Pulse oximeter) and the same anatomical site (fingertip)." This implies reliance on existing literature rather than new clinical data from the E.S.O. itself. There is no information on the country of origin or whether these studies were retrospective or prospective, as they are external references not directly part of the E.S.O.'s submission data.

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

Not Applicable / Not Specified. The document does not describe a process of establishing ground truth for a test set using experts for performance evaluation of the E.S.O. device.
"Peer reviews" are mentioned for the mathematical analyses of PP and HRV, but the number or qualifications of these reviewers are not detailed, nor is it explicitly stated that they established a "ground truth" for a test set. Their role seems to be more about assessing the mathematical soundness of the algorithms.

4. Adjudication Method for the Test Set

Not Applicable / Not Specified. No formal adjudication method is mentioned as a test set for performance evaluation by human experts was not described.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and the Effect Size of How Much Human Readers Improve with AI vs. Without AI Assistance

Not Applicable. This device is a pulse oximeter with mathematical analysis features, not an AI-assisted diagnostic tool that human readers would use to improve their performance with or without AI. Therefore, an MRMC study and effect size of human improvement with AI assistance are not relevant to this submission.

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

Partially Applicable, but with no detailed results. The "Software algorithms tests with input data from the MIT-BIH database" could be considered a form of standalone testing. The "comparative testing between EKG RR intervals (from predicate device) and SP02 RR intervals (from E.S.O)" and "comparative HRV analysis results from the predicate device and E.S.O." also suggest standalone comparison of the algorithm's output to a reference, but no specific performance metrics like accuracy, sensitivity, or specificity are provided. The document explicitly states that the PP and HRV analyses "are not intended use for diagnosis" and "no clinical validation is required for E.S.O" for these features.

7. The Type of Ground Truth Used (Expert Consensus, Pathology, Outcomes Data, etc.)

For Software Algorithms Tests: The use of the "MIT-BIH database" implies that the ground truth for software verification would be derived from the annotations and established physiological parameters within that database (e.g., expert-annotated heartbeats, rhythm classifications from EKG).
For RR Intervals and HRV Comparisons: The ground truth or reference standard used for the comparative testing between EKG RR intervals and SpO2 RR intervals would have been the EKG RR intervals from the predicate device. For HRV, the comparison was against HRV analysis results from the predicate device. This essentially uses an existing, cleared device as the reference standard.
For SpO2/Pulse Rate: No explicit ground truth method is described for verifying the accuracy of the SpO2 and pulse rate measurements. Reliance seems to be on the OEM subcontractor's clearance and general safety standards rather than detailed clinical accuracy studies within this submission.

8. The Sample Size for the Training Set

Not Specified. The document does not mention a training set or its sample size. The device relies on established algorithms for oximetry and mathematical analyses, rather than a machine learning model that would typically require a training set. The "Software algorithms tests with input data from the MIT-BIH database" is referred to for verification, not training.

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

Not Applicable. As no specific training set for a machine learning model is mentioned, the method for establishing its ground truth is also not applicable.

Summary

Regulation Number and Section

§ 870.2700 Oximeter.

(a)
Identification. An oximeter is a device used to transmit radiation at a known wavelength(s) through blood and to measure the blood oxygen saturation based on the amount of reflected or scattered radiation. It may be used alone or in conjunction with a fiberoptic oximeter catheter.(b)
Classification. Class II (performance standards).