(567 days)
The H & W Hospital Ethylene Oxide Sterilizer Control System is an accessory in the form of an electronic control system for the retrofit of existing sterilizers to provide operator and information monitoring capabilities of a personal computer (PC) together with the functional reliability, safety, and dependability of control obtained through the use of a programmable logic controller (PLC). The H & W Hospital Ethylene Oxide Sterilizer Control System is intended to retrofit existing Steris (Amsco) Eagle Model 2000 & 3000 series and Getinge (Castle) Model M/C 3500, 3600 or 4200 series Hospital EO Sterilizers only. The automated cycle performs in such a manner as to assure the repeatability of the process. The predetermined cycles duplicate the cycles provided by the original equipment manufacturer and are demonstrated to deliver the biological performance as stipulated in the consensus standard ANSI/AAMI ST-24.
This device is an accessory to an electronic ethylene oxide sterilizer control system. This control system is a modern, programmable logic controller (PLC)-based system with Personal Computer (PC) user interface. This accessory control system uses a PLC and a Visual Basic/Windows user interface. These two features, a PLC for control and a Windows -based user interface provide multiple advantages. First being the inherent safety and reliability of the PLC and second the operator's ease of use of the Windows "based user interface. This accessory control system kit is intended for the retrofit of existing sterilizers or for use on sterilizers remanufactured by original equipment manufacturers, third party remanufacturers or their licensees.
The provided text describes an ethylene oxide sterilizer control system, an accessory for existing sterilizers, and its compliance with standards and safety. However, it does not contain the specific information required to complete a table of acceptance criteria and reported device performance with numerical metrics, nor does it detail a clinical study with elements like sample sizes, expert ground truth establishment, or multi-reader multi-case studies as typically expected for medical imaging or diagnostic AI devices.
The submission is for a control system for a sterilizer, not a diagnostic or imaging device, which explains the absence of the typical "study" information. The "study" here refers to validation against engineering and performance standards for sterilizers.
Here's an attempt to answer based on the provided text, highlighting the limitations due to the nature of the device and the available information:
Acceptance Criteria and Device Performance Study for H & W Hospital Ethylene Oxide Sterilizer Control System
The H & W Hospital Ethylene Oxide Sterilizer Control System is an electronic control system accessory designed to retrofit existing hospital ethylene oxide (EO) sterilizers. The primary objective is to maintain or improve the reliability, safety, and performance of the sterilizer, particularly its biological performance, according to established consensus standards.
Given that this is a control system for sterilizers and not a diagnostic or imaging device, the "acceptance criteria" and "study" details differ significantly from what would be expected for AI-powered diagnostic tools. The "performance" largely revolves around adherence to industry standards, safety regulations, and functional equivalence to predicate devices, rather than statistical metrics like sensitivity, specificity, or AUC from a clinical trial.
1. Table of Acceptance Criteria and Reported Device Performance
| Acceptance Criteria Category | Specific Criteria/Standard | Reported Device Performance | Comments / Reference |
|---|---|---|---|
| Biological Performance | Deliver biological performance as stipulated in ANSI/AAMI ST-24 (3rd edition, 1999). | Predetermined cycles duplicate original equipment manufacturer (OEM) cycles and are "demonstrated to deliver the biological performance as stipulated in the consensus standard." | The H&W system conforms to this standard. The predicate system conformed to an earlier edition, but both met "the same criteria for acceptance" when considering the state of the art at their respective introductions. |
| Functional Reliability, Safety & Dependability | Ensure functional reliability, safety, and dependability of a Programmable Logic Controller (PLC) and PC user interface. | Utilizes a PLC for inherent safety and reliability, and a Windows-based user interface for ease of use. A rigorous Product Development Life Cycle (PDLC) assures integrity of design, controlled fabrication, and testing. | PDLC procedure is intended to assure requirements are met and validation is conducted. |
| Reproducibility | Assure repeatability of the automated sterilization process. | "The automated cycle performs in such a manner as to assure the repeatability of the process." | Duplicate OEM parameters. |
| Mechanical Safety | Meet requirements of ASME Boiler and Pressure Vessel Code; conform to IEC 1010-2-042. | Original equipment manufacturer or qualified third-party remanufacturer inspects and qualifies the vessel suitability to ASME code. H&W system conforms to IEC 1010-2-042. | Predicate device's mechanical safety also assured similarly. |
| Electrical Safety | Conform to IEC 1010-2-042 Particular Requirements for Toxic Gas Sterilizers. | H&W system conforms to IEC 1010-2-042, which is described as more comprehensive and specific than the predicate's UL 3101-1 Part 1 general requirements. | Emphasized as an improvement over the predicate device. |
| Human Factors/Ease of Use | Incorporate user expectations and consensus standard elements for operator interface. | Utilizes a PC as the operator control interface, allowing for effective information presentation (sound, color, animation) to gain attention as appropriate. | Acknowledged as an improvement/specific consideration beyond the predicate. |
| On-site Performance Validation | Validate system performance on-site in a sterilizer with known and defined characteristics as per reference standard. | The "bench mark to be used was and is the sterilizer retrofitted." The retrofitted system "demonstrates equal or better performance." | Performance is validated on-site. |
| Emission Controls | Accommodate emission controls. | Accommodated in both predicate and H&W systems, whether integral or external to the sterilizer. | Addressed. |
| Compatibility | Intended for specific sterilizer models. | Intended to retrofit existing Steris (Amsco) Eagle Model 2000 & 3000 series and Getinge (Castle) Model M/C 3500, 3600 or 4200 series Hospital EO Sterilizers only. | Specific statement of intended use. (K014013 Indications for Use Statement) |
2. Sample size used for the test set and the data provenance
The document does not specify a "test set" in the context of data for an algorithm. Instead, the validation appears to be centered on the system itself when installed on existing sterilizers.
- The "test set" is implied to be actual sterilizers where the system is retrofitted and its performance validated on-site.
- Sample Size: Not explicitly stated as a number of devices. The validation process refers to "a sterilizer" (singular) but implies this process is to be followed for each retrofitted unit: "Performance is assured... by validating the performance of the system, on-site in a sterilizer with known and already defined characteristics."
- Data Provenance: This is an industrial control system, not a data-driven AI. "Data provenance" as typically understood (e.g., country of origin of patient data) is not applicable. The performance is validated against the physical characteristics and operational output of the sterilizer itself. The context is practical engineering validation rather than clinical data analysis.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts
This information is not explicitly provided. The "ground truth" for the biological performance is the successful sterilization outcomes as stipulated by consensus standard ANSI/AAMI ST-24. This standard itself is developed by experts in the field. The operational validation (on-site testing) would likely involve qualified technicians or engineers, but their number and specific qualifications are not detailed in this submission.
4. Adjudication method for the test set
Not applicable. There is no mention of "adjudication" in the context of panel reviews or conflict resolution for expert opinions, since the "test set" is the biological and mechanical performance of the sterilizer, measured against established standards.
5. If a multi-reader multi-case (MRMC) comparative effectiveness study was done
No. This type of study is typically performed for diagnostic devices (e.g., medical imaging AI) to compare human reader performance with and without AI assistance. This device is an industrial control system accessory, not a diagnostic tool, so an MRMC study is not relevant or described.
6. If a standalone (i.e., algorithm only without human-in-the-loop performance) was done
The device is a control system, which functions "standalone" once programmed, but always within the context of physically controlling a sterilizer. The "algorithm" here refers to the PLC-based control logic, which operates independently to execute sterilization cycles. The validation focuses on whether these automated cycles achieve the required biological performance and safety standards, which is essentially its standalone or automated performance. Human interaction is primarily for monitoring and initiation through the PC interface.
7. The type of ground truth used
The primary ground truth is compliance with consensus standards and demonstrated biological performance as stipulated by ANSI/AAMI ST-24. This includes the successful inactivation of biological indicators, which is an objective measure of sterilization efficacy. Other ground truths include adherence to engineering and electrical safety standards (ASME, IEC).
8. The sample size for the training set
Not applicable. This device is not an AI/machine learning model in the sense of requiring a "training set" of data to learn from. Its logic is programmed based on engineering principles and industry standards, not data-driven model training.
9. How the ground truth for the training set was established
Not applicable, as there is no "training set" in the context of machine learning. The "ground truth" for the design and programming of the control system is based on the requirements defined by the Product Development Life Cycle (PDLC) procedure, industry best practices, and the consensus standard ANSI/AAMI ST-24. These requirements are established by experts and regulatory bodies in the field of sterilization.
§ 880.6860 Ethylene oxide gas sterilizer.
(a)
Identification. An ethylene gas sterilizer is a nonportable device intended for use by a health care provider that uses ethylene oxide (ETO) to sterilize medical products.(b)
Classification. Class II (performance standards).