(136 days)
The PhySoft Anemia Management System® (PhySoftAMS®) is a software application used to obtain, track and trend patient data pertaining to the management of anemia, and to provide a schedule of erythropoiesis-stimulating agent (ESA) dosage recommendations to help achieve and maintain target hemoglobin levels in dialysis patients. PhySoftAMS® is intended to help physicians. nurses, clinicians, and anemia managers manage anemia in adult stage 5 chronic kidney disease (CKD) patients.
PhySoftAMS® is not a substitute for, but is rather intended to assist, clinical judgment. The erythropoiesis-stimulating agent (ESA) dosing regimen options calculated by this device are intended to be used by qualified and trained medical personnel to inform the optimization of the dosage of ESAs in accordance with their approved labeling in conjunction with clinical history, symptoms, and other diagnostic measurements, as well as the medical professional's clinical judgment. No medical decision should be based solely on the patient Hgb response to dosing regimen options calculated by this device.
PhySoftAMS® is a software application used to obtain, track and trend patient data pertaining to the management of anemia and to provide a schedule of erythropoiesis-stimulating agent (ESA) dosage recommendations to help achieve and maintain target hemoglobin levels in dialysis patients. PhySoftAMS® is intended to help physicians, nurses, clinicians and anemia managers manage anemia in adult stage 5 chronic kidney disease (CKD) patients.
PhySoftAMS® is intended for use by medical personnel such as clinicians, nurses, and physicians in dialysis clinics or other settings where anemia management for hemodialysis patients is conducted.
Healthcare professionals access PhySoftAMS® directly using a web application graphical user interface (GUI) or indirectly using the drug dosing-related screens of a third party's electronic health record (EHR) system via an application programming interface (API) provided by the PhySoftAMS® application server.
PhySoftAMS® evaluates whether adequate historical data is available to model patient ESA dose-Hgb response dynamics and project future ESA dose-Hgb response. If adequate data is available, PhySoftAMS® enables a physician to model a patient and select from one or more dosing schedule options most likely to result in achieving target Hgb levels or, at the physician's discretion. override the presented dosing schedule options.
The provided text describes a 510(k) premarket notification for a medical device called PhySoftAMS®, a software application for managing anemia in dialysis patients. However, the document does not contain the detailed information necessary to answer all the questions regarding acceptance criteria and the study proving the device meets those criteria.
Specifically, the document states:
- "Bench testing results demonstrate the modified device performance is computationally equivalent to the performance of the predicate device."
- "Software verification and validation of the device modifications that are the subject of this submission demonstrated that the enhancements for the subject (modified) device perform as intended and have no effect on the modeling process or other device functions of PhySoftAMS®."
This indicates that internal testing (bench testing, software V&V) was performed to show computational equivalence and proper function of the modifications, but it does not detail a clinical study or performance study with acceptance criteria, sample sizes, ground truth establishment, or multi-reader multi-case studies as requested.
Therefore, many of the requested details cannot be extracted from this document. I can only provide information directly inferable from the given text.
Here's an attempt to answer based only on the provided text, highlighting what is not available:
Acceptance Criteria and Study for PhySoftAMS®
The provided 510(k) summary (K232283) focuses on demonstrating substantial equivalence of a modified PhySoftAMS® to its predicate device (PhySoft AMS™, K130579), primarily regarding the new integration with EHR systems via an API. The performance data presented relates to computational equivalence and software functionality, rather than a clinical performance study with defined acceptance criteria for diagnostic or clinical accuracy.
1. Table of Acceptance Criteria and Reported Device Performance
Based on the provided text, formal acceptance criteria for a clinical performance study (e.g., sensitivity, specificity, accuracy targets) and corresponding reported performance metrics are not explicitly stated. The "Performance" section focuses on computational and functional equivalence to the predicate.
| Acceptance Criteria | Reported Device Performance |
|---|---|
| Not explicitly stated as numerical performance targets for clinical outcomes. The document states the goal was to demonstrate "computational equivalence to the performance of the predicate device" and that modifications "perform as intended and have no effect on the modeling process or other device functions." | Computationally equivalent to the predicate device. |
| Enhancements perform as intended and have no effect on the modeling process or other device functions of PhySoftAMS®. |
2. Sample Size Used for the Test Set and Data Provenance
The document does not specify a sample size for a "test set" in the context of a clinical or retrospective/prospective performance study. The evaluations described (bench testing, software V&V) typically do not involve patient-level test sets in the same manner as a diagnostic study. The data provenance (country, retrospective/prospective) is also not mentioned.
3. Number of Experts Used to Establish Ground Truth and Qualifications
The document does not mention using experts to establish ground truth for a test set, as no such clinical or performance test set is described. The device provides "dosing recommendations," which would typically be compared against clinical outcomes or expert judgment in a performance study, but this is not detailed here.
4. Adjudication Method for the Test Set
Since no test set adjudicated by experts is described, the adjudication method is not applicable/not mentioned.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was Done
No MRMC study is described in the provided text. The device offers dosing recommendations and assists human users, but there is no information on a study comparing human readers with and without AI assistance, or an effect size.
6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done
The text indicates the device "provides a schedule of erythropoiesis-stimulating agent (ESA) dosage recommendations to help achieve and maintain target hemoglobin levels" and is "intended to help physicians, nurses, clinicians, and anemia managers manage anemia." It explicitly states, "PhySoftAMS® is not a substitute for, but is rather intended to assist, clinical judgment." This implies a human-in-the-loop design. No standalone algorithmic performance study results are detailed in the document. The "computational equivalence" refers to the underlying modeling process itself.
7. The Type of Ground Truth Used
For the software verification and validation, the "ground truth" would be the expected behavior and output of the algorithms and software functions, based on predefined specifications and the known performance of the predicate device. For potential clinical application, the ultimate ground truth would relate to actual patient hemoglobin levels and clinical outcomes, but the studies described do not detail the collection or establishment of such ground truth.
8. The Sample Size for the Training Set
The document does not mention a training set size. This device appears to use PK/PD modeling of patient response rather than a deep learning model that typically requires a large training set of annotated data. The system tracks and trends past Hgb and ESA dosages for individual patient modeling, not a general training dataset for an AI model.
9. How the Ground Truth for the Training Set Was Established
As no training set (in the context of machine learning) is explicitly mentioned, the method for establishing its ground truth is not detailed. The modeling is described as "PK/PD modeling of patient response to ESAs," which is typically based on physiological and pharmacological principles, not a data-driven training set with established ground truth labels in the same way an image classification AI might be trained.
§ 876.5820 Hemodialysis system and accessories.
(a)
Identification. A hemodialysis system and accessories is a device that is used as an artificial kidney system for the treatment of patients with renal failure or toxemic conditions and that consists of an extracorporeal blood system, a conventional dialyzer, a dialysate delivery system, and accessories. Blood from a patient flows through the tubing of the extracorporeal blood system and accessories to the blood compartment of the dialyzer, then returns through further tubing of the extracorporeal blood system to the patient. The dialyzer has two compartments that are separated by a semipermeable membrane. While the blood is in the blood compartment, undesirable substances in the blood pass through the semipermeable membrane into the dialysate in the dialysate compartment. The dialysate delivery system controls and monitors the dialysate circulating through the dialysate compartment of the dialyzer.(1) The extracorporeal blood system and accessories consists of tubing, pumps, pressure monitors, air foam or bubble detectors, and alarms to keep blood moving safely from the blood access device and accessories for hemodialysis (§ 876.5540) to the blood compartment of the dialyzer and back to the patient.
(2) The conventional dialyzer allows a transfer of water and solutes between the blood and the dialysate through the semipermeable membrane. The semipermeable membrane of the conventional dialyzer has a sufficiently low permeability to water that an ultrafiltration controller is not required to prevent excessive loss of water from the patient's blood. This conventional dialyzer does not include hemodialyzers with the disposable inserts (Kiil type) (§ 876.5830) or dialyzers of high permeability (§ 876.5860).
(3) The dialysate delivery system consists of mechanisms that monitor and control the temperature, conductivity, flow rate, and pressure of the dialysate and circulates dialysate through the dialysate compartment of the dialyzer. The dialysate delivery system includes the dialysate concentrate for hemodialysis (liquid or powder) and alarms to indicate abnormal dialysate conditions. This dialysate delivery system does not include the sorbent regenerated dialysate delivery system for hemodialysis (§ 876.5600), the dialysate delivery system of the peritoneal dialysis system and accessories (§ 876.5630), or the controlled dialysate delivery system of the high permeability hemodialysis system § 876.5860).
(4) Remote accessories to the hemodialysis system include the unpowered dialysis chair without a scale, the powered dialysis chair without a scale, the dialyzer holder set, dialysis tie gun and ties, and hemodialysis start/stop tray.
(b)
Classification. (1) Class II (performance standards) for hemodialysis systems and all accessories directly associated with the extracorporeal blood system and the dialysate delivery system.(2) Class I for other accessories of the hemodialysis system remote from the extracorporeal blood system and the dialysate delivery system, such as the unpowered dialysis chair, hemodialysis start/stop tray, dialyzer holder set, and dialysis tie gun and ties. The devices subject to this paragraph (b)(2) are exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 876.9.