Search Results

The Nuubo System is indicated for use on patients who may be asymptomatic or who may suffer from transient symptoms such as palpitations, shortness of breath, dizziness, light-headedness, presyncope, syncope, fatigue, or anxiety.

The Nuubo system continuously records and stores ECG and activity data for up to 30 days at a time. The Nuubo System detects arrhythmias at the end of the monitoring period upon download of the ECG data. The Nuubo System is Rx use device.

The Nuubo System, developed by Smart Solutions Technologies (SST), is a wearable device designed for ambulatory recording electrocardiogram (ECG) up to 30 days. The system is composed of 3 main components:

• Nuubo30 – The Nuubo30 wearable is a single patient textile like a chest-belt that contains 4 textile electrodes in the inner side that are used for sensing patient 's ECG.
• NuuboREC - The Nuubo recorder is a small, lightweight device that records ECG continuously. The device records 2 Leads of ECG data up to 30 days. The device also records data from a 3-axis accelerometer located inside the device.
• Nuubo Leonardo - The Leonardo Software is installed on a computer where the patient´s ECG data stored in the recorder will be downloaded for subsequent analysis and report.

This document is a 510(k) summary for the Nuubo System, submitted to the FDA. It details the device's indications for use, technological characteristics, and performance data to demonstrate substantial equivalence to a predicate device.

Here's an analysis based on the provided information:

1. Table of Acceptance Criteria and Reported Device Performance

The document explicitly states that "The results obtained validate the Nuubo Arrhythmia Algorithms and prove equivalence to Monebo Automated ECG Analysis And Interpretation Software Library, version 3.0 [manufacturer Monebo Technologies, 510(k) number K062282]." It further mentions that "All results are comparable to the results claimed by Monebo."

While specific numerical acceptance criteria (e.g., minimum sensitivity/specificity for each arrhythmia) and the Nuubo System's reported performance for each criterion are not directly listed in the provided text, the document indicates that the device's arrhythmia detection performance was validated and found comparable to the Monebo device, which was referenced by the predicate ZioPatch. The table below lists the arrhythmia functionalities and their detection criteria, implying these are the targets for the algorithm:

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

The arrhythmia detection algorithms were tested using:

• Public Databases: MIT-BIH, AHA, and MITAF. These are well-known, established retrospective ECG databases. The country of origin for these databases is primarily the United States.
• Private Database: Comprised of 90 ECG registries from 58 patients. The provenance (country of origin) is not explicitly stated, but the context of the submission to the FDA suggests it aligns with U.S. medical practice. This data is described as "anonymized," potentially retrospective, selected to contain all arrhythmias detected by the algorithm.

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

For the private database (90 ECG registries from 58 patients), the ground truth was established by "experienced medical professionals." The specific number of experts is not provided, nor are their detailed qualifications (e.g., specific sub-specialty or years of experience). For the public databases (MIT-BIH, AHA, MITAF), ground truth is typically established by consensus of multiple cardiologists or electrophysiologists based on established annotation guidelines, though the document doesn't detail this for these specific databases.

4. Adjudication Method for the Test Set

The document states that the private database "was annotated using experienced medical professionals, using prospectively defined guidelines, consistent with US medical practice." This suggests an expert review process. However, the exact adjudication method (e.g., 2+1, 3+1, or simple consensus from a single annotator if a single expert annotated) is not specified. For the public databases, adjudication methods are usually defined by the database creators, often involving multiple experts to establish consensus.

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

No MRMC comparative effectiveness study involving human readers' improvement with AI vs. without AI assistance is reported in this document. The study focuses on the standalone performance of the algorithm.

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

Yes, a standalone performance study of the algorithm was done. The document states: "The Nuubo Arrhythmia detection algorithm was tested per requirements of AAMI/IEC60601-2-47. To validate the arrhythmia detection algorithms were used three public databases and one private database." The results of this testing were compared to the performance claimed by the Monebo Automated ECG Analysis And Interpretation Software Library (K062282), indicating a standalone comparison. The caveat is that "The Nuubo Arrhythmia Detection Algorithm is not intended to replace the Clinician review of signals. The software menu prompts the Clinician or trained technician to review events prior to generating a report," which implies that the intended use is with a human-in-the-loop for final review, even if standalone performance was tested.

7. The Type of Ground Truth Used (expert consensus, pathology, outcomes data, etc.)

The type of ground truth used for the private database was expert annotation/consensus based on "experienced medical professionals, using prospectively defined guidelines." For the public databases (MIT-BIH, AHA, MITAF), the ground truth is also based on expert annotation (often expert consensus).

8. The Sample Size for the Training Set

The document does not explicitly state the sample size used for the training set for the Nuubo arrhythmia detection algorithm. It only mentions the databases used for validation/testing.

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

Since the training set size and specific databases are not mentioned, how the ground truth for any training set was established is also not described in this document. The description of ground truth establishment is specifically for the test/validation data.

Ask a specific question about this device

Guava II is intended to be used by cardiologists, general practitioners, cardiac or ECG technicians, nurses, monitoring service technicians, and other cardiac related institutions or care givers to store, retrieve, communicate and report ECG and ECG data acquired from a variety of ECG sources including single and multilead ECG devices. Guava II is intended to be electronically interfaced, and perform analysis with data transferred from other computer based ECG systems. Guava II includes an algorithm that will analyze the 12 Lead ECG and produce measurements of the ECG recording as well as textual interpretation. The product also includes an automatic analysis and interpretation software library that provides ECG signal processing and analysis on a beat by beat basis, QRS and Ventricular Ectopic Beat detection, QRS feature extraction, interval measurement, heart rate measurement, and rhythm analysis for up to twelve (12) leads of historic ambulatory ECG data. The Guava II algorithm analysis can detect and classify arrhythmias and may be used to triage data, however, Guava II is not intended to offer independent diagnosis or provide medical alarms. Clinical judgment and experience are used to check and interpret the data as part of a diagnosis. Guava II is not for use in lifesupporting or sustaining systems or EGG monitoring devices. Guava II is intended for use in clinics, hospitals, physician's offices, or anywhere a medical doctor deems appropriate.

The Guava II is a medical device software with a formal Application Programming Interface (API). The API can be invoked from Host Applications to provide services for capturing, storing, retrieving, viewing, editing, and analyzing ECGs (1-, 2-, 3-Channel and 12-Lead) and other Biosignals. Guava II may also be licensed to 3rd party organizations interested in embedding the capabilities within their own products. When the Guava II is used in other medical products, manufacturers will identify the indications for use depending on the application of their device.

The Guava II is not a life-supporting or life-sustaining system. It is intended that competent human intervention be involved before any impact on health occurs. Clinical judgment and experience are used to check and interpret the data.

Here's an analysis of the acceptance criteria and study information for the Guava II device based on the provided 510(k) summary:

Acceptance Criteria and Reported Device Performance

The provided document does not explicitly list specific numerical acceptance criteria (e.g., sensitivity, specificity, accuracy targets) nor provide detailed quantitative performance metrics for the Guava II device's algorithm. Instead, the performance testing section broadly states:

• "The device complies with AAMI EC11 and AAMI EC38 standards."
• "Verification and validation activities related to the device modification were performed on the applicant device, and the predetermined acceptance criteria were met in all cases."

Without specific numerical criteria or results, a table of acceptance criteria and reported performance cannot be generated in the traditional sense of metrics like sensitivity or accuracy. The "acceptance criteria" here are defined by compliance with AAMI standards and the successful outcome of the V&V activities.

If we were to infer the "reported performance" based on the conclusion, it would be "substantially equivalent to the predicate devices" and that its functions are "identical in function" to the QRS Algorithm in the predicate device.

1. Table of Acceptance Criteria and Reported Device Performance

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

• Test Set Sample Size: Not specified in the provided document.
• Data Provenance: Not specified in the provided document.

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

• Number of Experts: Not specified.
• Qualifications of Experts: Not specified.
  • The Intended Use section states that "Clinical judgment and experience are used to check and interpret the data as part of a diagnosis," implying human expertise is involved in the clinical workflow, but this does not directly describe the establishment of a ground truth for a test set in a study.

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

• Not specified.

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

• No MRMC comparative effectiveness study involving human readers with and without AI assistance is mentioned. The study described focuses on the device's own performance and its equivalence to predicate devices, not on human reader improvement.

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

• Yes, a standalone performance assessment of the algorithm appears to have been conducted. The statement "The Guava II Louvain Data Processor performs automated ECG analysis and interpretation in the same manner as GlobalCardio. The two products use the identical machine code library (DLL) to perform the analysis, and testing demonstrates the QRS Algorithm embedded in the Guava II Louvain Data Processor is identical in function to the QRS Algorithm embedded in GlobalCardio" suggests a direct comparison of the algorithm's output.
• "Verification and validation activities related to the device modification were performed on the applicant device, and the predetermined acceptance criteria were met in all cases. The activities included scenario validations, algorithm confirmation testing, and device functional testing." This also points to standalone algorithm testing.

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

• The document does not explicitly state the type of ground truth used for performance testing. However, given the context of ECG analysis and comparison to a predicate device, it is common for the ground truth to be established by expert cardiologist review/consensus of the ECG waveforms, or by comparison to the established accuracy of the predicate's algorithm (implying the predicate’s ground truth method was implicitly used). The closest hint is the reference to "Clinical judgment and experience are used to check and interpret the data."

8. The sample size for the training set:

• Not specified. The document primarily discusses performance testing and equivalence to predicate devices, not the development or training of the algorithm.

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

• Not specified. As no training set information is provided, no details on its ground truth establishment are available.

Ask a specific question about this device