Search Results

The EON Portable Reverse Osmosis Water Purification System is intended to be used as a dialysis accessory to produce water through reverse osmosis for use with hemodialysis equipment. EON can be connected to hemodialysis equipment used in hospitals, clinics and in home environments, in conjunction with the appropriate pre and post treatment units, as part of a water treatment system designed to meet current AAMI and Federal (U.S.) standards.

EON has optional heat disinfection cycles intended to disinfect the reverse osmosis (RO) machine and product loop, and connection tubing to the hemodialysis machine. EON's heat disinfect the connection tubing (heat forward cycle) is intended to be used only with hemodialysis machines which contain their own heat disinfection cycles and hence are able to tolerate high temperatures. EON is not intended to heat disinfect the hemodialysis machine.

The device is a portable water purification system which uses reverse osmosis to remove contaminants from water that is used to dilute dialysis concentrate to form dialysate for use in hemodialysis equipment. Feed water enters the unit and is directed through a pump into a RO membrane. The pump applies a high hydrostatic pressure that forces water from the concentrated (feed) side to the dilute (product) side of the RO membrane. As water flows across the membrane purified water is produced. Both subject and predicate devices are designed to maintain low microbiological levels in the flow pathway by using optional cycles which perform heat disinfection on the entire RO machine and loop. The subject device also has an optional Heat Forward cycle which is intended to heat disinfect the connection tubing to the hemodialysis machine.

The EON is capable of generating purified water that meets AAMI water quality requirements for hemodialysis. It must be used with appropriate pre and post treatment units, including at a minimum carbon adsorption media pretreatment in order to remove chlorine/chloramines. Additional pre and post treatment requirements may vary and are dependent on the quality of the local feed water supply and individual facility requirements.

The EON system is designed to maintain low microbiological levels in the flow pathway through regular heat disinfection and chemical sanitization. Notable components and features of the EON include:

• RO membrane .
• System pump .
• Water quality monitoring system .
• Operating panel and programmable logic controller (OPLC) .
• Heat disinfection and chemical sanitization capability .
• Audible and visual alarms .
• Automatic divert to drain mode upon start-up and anytime product water . TDS is above the quality set-point
• System control via a touch-screen user interface .
• Heat forward cycle intended to heat disinfect connection tubing from the . Portable Reverse Osmosis Water Purification System to the hemodialysis machine.

This document is an FDA 510(k) clearance letter for a medical device, specifically a water purification system. As such, it describes the device itself and its comparison to predicate devices, but it does not contain detailed information about acceptance criteria or a study proving the device meets those criteria in the context of an Artificial Intelligence/Machine Learning (AI/ML) enabled device.

The provided text focuses on:

• Device Name: EON Portable Reverse Osmosis Water Purification System
• Indication for Use: To produce water through reverse osmosis for use with hemodialysis equipment, and optional heat disinfection cycles.
• Regulatory Classification: Class II, Product Code FIP (Water purification system for hemodialysis).
• Comparison to Predicate Devices: EON Portable Reverse Osmosis Water Purification System (K171099) and WRO 300H (K093608). The key difference is the membrane used and minor component changes.
• Non-Clinical Performance Data: Mentions "System and RO Membrane Performance Flow and product water quality verification over range of operating conditions" and "Software Verification."

Therefore, I cannot provide the requested information about acceptance criteria and a study proving the device meets those criteria as it pertains to an AI/ML device. The document describes a physical water purification system, not an AI/ML diagnostic tool.

If this were an AI/ML device, the detailed information requested (sample sizes, expert qualifications, etc.) would typically be found in a separate clinical study report or a more detailed submission summary, not typically in the top-level 510(k) clearance letter itself, which primarily confirms substantial equivalence.

However, if we were to hypothesize what such information would look like for an AI/ML device related to this context (e.g., an AI assessing water purity from sensor data), here's an example of how the requested table and study description might be presented. This is purely illustrative and not based on the provided document.

Hypothetical AI/ML Device for Water Purity Assessment in Hemodialysis

Let's imagine a hypothetical AI/ML device that analyzes sensor data from the EON Portable Reverse Osmosis Water Purification System to predict water purity levels and flag potential issues (e.g., microbial contamination risk, inadequate filtration) to assist hemodialysis technicians.

1. Table of Acceptance Criteria and Reported Device Performance (Hypothetical)

Study Proving Device Meets Acceptance Criteria (Hypothetical)

2. Sample Size and Data Provenance for Test Set:

• Sample Size: 10,000 operational hours of sensor data, corresponding to approximately 500 distinct water purification cycles.
• Data Provenance: Retrospective data collected from 15 hemodialysis clinics across the United States (70%), Germany (20%), and Japan (10%) over a 2-year period. Data anonymized and de-identified.

3. Number of Experts and Qualifications for Ground Truth:

• Number of Experts: 5
• Qualifications: Three board-certified Nephrologists with an average of 15 years of experience in hemodialysis and water quality management. Two clinical microbiologists with over 10 years of experience in water quality analysis for medical applications.

4. Adjudication Method for Test Set:

• Method: 3+2 Adjudication. For each data segment requiring ground truth labeling (e.g., "contaminated" or "clean"), three primary experts (2 Nephrologists, 1 Microbiologist) independently reviewed sensor data, lab results (if available), and clinical logs to make an initial determination. If there was a disagreement among these three, the two remaining experts (1 Nephrologist, 1 Microbiologist) were brought in for a consensus discussion until a final label was agreed upon by at least 4 out of 5 experts. If consensus couldn't be reached, the data segment was excluded.

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

• Was it done? Yes.
• Effect Size of Human Improvement (AI-assisted vs. without AI):
  • Study Design: 10 hemodialysis technicians (readers) independently reviewed the 500 water purification cycles from the test set for potential issues.
  • Phase 1 (Without AI): Technicians reviewed raw sensor data and log files. The average accuracy for identifying critical water purity issues was 78%.
  • Phase 2 (With AI-Assistance): After a washout period, the same technicians reviewed the same cases, this time with the AI system providing alerts and risk scores. The average accuracy increased to 98%.
  • Effect Size: The AI assistance led to an average 20% absolute improvement in the accuracy of human readers for identifying critical water purity issues (from 78% to 98%). The relative improvement was approximately 25.6% (20/78). A statistically significant difference (p < 0.001) was observed.

6. Standalone Algorithm Performance:

• Was it done? Yes.
• The standalone algorithm achieved a sensitivity of 96.2% and a specificity of 91.5% for detecting critical water purity issues on the test set, corresponding to the "Reported Device Performance" in the table above.

7. Type of Ground Truth Used:

• Expert Consensus: The primary ground truth was established through expert consensus based on a comprehensive review of sensor data, real-world laboratory water quality reports (e.g., TOC, conductivity, microbial counts if available), and clinical documentation of any related adverse events or interventions. For verification, a subset of the cases with confirmed lab-diagnosed contamination was cross-referenced.

8. Sample Size for Training Set:

• Sample Size: 50,000 operational hours of sensor data, corresponding to approximately 2,500 distinct water purification cycles.

9. How Ground Truth for Training Set was Established:

• Initial ground truth for the training set was established primarily through automated labeling based on predefined thresholds from AAMI water quality standards, combined with routine laboratory analysis results for water quality parameters (e.g., TDS, conductivity, bacterial counts).
• A subset of 10% of the training data was manually reviewed and verified by internal subject matter experts (engineers with hemodialysis system expertise) to correct any noisy or erroneous automated labels, employing a simpler 2-expert review process. This iterative approach helped refine the training data quality.

Ask a specific question about this device