(29 days)
The ACL Elite and ACL Elite Pro are fully automated, high-productivity analyzers designed specifically for in vitro diagnostic clinical use in the hemostasis laboratory for coagulation and/or fibrinolysis testing in the assessment of thrombosis and/or hemostasis. The systems provide results for both direct hemostasis measurements and calculated parameters.
The ACL Elite and ACL Elite Pro are fully automated, high-productivity analyzers designed specifically for in vitro diagnostic clinical use in the hemostasis laboratory for coagulation and/or fibrinolysis testing in the assessment of thrombosis and/or hemostasis. The systems provide results for both direct hemostasis measurements and calculated parameters.
The analytical modifications to the ACL 8000/9000/10000 introduced with the new family members (ACL Elite and ACL Elite Pro) include the following:
- Modification to the test parameter (revised use of the Secondary Algorithm) for several . PT and APTT assays.
- Addition of dedicated calibration for HemosIL Factor VIII and HemosIL Factor IX . Deficient Plasmas, when used in conjunction with HemosIL APTT-SP.
- Introduction of automated factor parallelism, which is a manual function on the ACL . 8000/9000/10000.
Here's a breakdown of the acceptance criteria and study information for the ACL Elite and ACL Elite Pro as described in the provided 510(k) summary:
1. Table of Acceptance Criteria and Reported Device Performance
The acceptance criteria are implied by the comparison to a legally marketed predicate device (ACL 9000, ACL 8000, and ACL 10000; K000053). The goal is to demonstrate "substantial equivalence" in performance. The summary details precision and method comparison data to support this claim.
Implied Acceptance Criteria (based on predicate equivalence and good laboratory practice):
- Precision: The device should exhibit acceptable within-run and total precision (low %CV) for various coagulation tests (FVIII, FIX, APTT, PT, Fibrinogen), comparable to or better than established laboratory standards and the predicate device.
- Method Comparison: The device should show statistical similarity (e.g., strong correlation, slope near 1, and intercept near 0) when compared to the predicate device for various coagulation assays.
Reported Device Performance (ACL Elite Pro):
| Test | Material | n | Mean | Within Run % CV | Total % CV |
|---|---|---|---|---|---|
| HemosIL Factor VIII (w/ APTT-SP) | Normal Control | 50 | 85.3 | 4.1 | 4.5 |
| Special Test Control Level 1 | 50 | 72.4 | 2.9 | 3.6 | |
| Special Test Control Level 2 | 50 | 36.3 | 4.6 | 5.4 | |
| HemosIL Factor IX (w/ APTT-SP) | Normal Control | 50 | 115.6 | 2.7 | 5.2 |
| Special Test Control Level 1 | 50 | 76.2 | 2.3 | 4.5 | |
| Special Test Control Level 2 | 50 | 41.0 | 3.3 | 5.1 | |
| HemosIL APTT-SP (APTT) | Normal Control | 50 | 29.3 | 1.1 | 2.0 |
| Low Abnormal Control | 50 | 46.1 | 1.9 | 4.1 | |
| High Abnormal Control | 50 | 56.9 | 2.2 | 4.7 | |
| HemosIL SynthASil (APTT) | Normal Control | 50 | 29.0 | 0.8 | 1.8 |
| Low Abnormal Control | 50 | 51.9 | 1.0 | 1.5 | |
| High Abnormal Control | 50 | 60.5 | 1.2 | 1.2 | |
| HemosIL PT-Fibrinogen (PT) | Normal Control | 50 | 12.3 | 0.6 | 1.1 |
| Low Abnormal Control | 50 | 20.7 | 0.9 | 1.1 | |
| High Abnormal Control | 50 | 25.5 | 0.8 | 1.4 | |
| HemosIL PT-Fibrinogen (Fibrinogen) | Normal Control | 50 | 277.9 | 4.4 | 4.6 |
| Low Fibrinogen Control | 50 | 132.9 | 3.6 | 5.5 | |
| HemosIL PT-Fibrinogen Recombinant (PT) | Normal Control | 50 | 10.7 | 1.0 | 1.0 |
| Low Abnormal Control | 50 | 35.0 | 1.1 | 1.3 | |
| High Abnormal Control | 50 | 51.6 | 1.4 | 3.4 | |
| HemosIL PT-Fibrinogen Recombinant (Fibrinogen) | Normal Control | 50 | 297.6 | 3.8 | 6.7 |
| Low Fibrinogen Control | 50 | 113.1 | 5.3 | 6.4 |
Method Comparison Data (ACL Elite/ACL Elite Pro vs. ACL 9000):
| Test | Parameter | n | Slope | r |
|---|---|---|---|---|
| HemosIL Factor VIII | FVIII w/ APTT-SP (% Act.) | 47 | 0.909 | 0.9938 |
| HemosIL Factor IX | FIX with APTT-SP (% Act.) | 49 | 0.948 | 0.9914 |
| HemosIL APTT-SP | APTT (Seconds) | 54 | 1.032 | 0.9990 |
| HemosIL SynthASil | APTT (Seconds) | 52 | 1.009 | 0.9981 |
| HemosIL PT-Fibrinogen | PT (sec) | 49 | 0.989 | 0.9985 |
| Fibrinogen (mg/dL) | 47 | 0.996 | 0.9848 | |
| HemosIL PT-Fibrinogen Recombinant | PT (sec) | 49 | 0.972 | 0.9993 |
| Fibrinogen (mg/dL) | 47 | 0.974 | 0.9992 |
2. Sample Sizes Used for the Test Set and Data Provenance
- Precision Test Set: For each combination of test and control material,
n=50measurements were performed. This involved control plasma samples (Normal Control, Special Test Control Levels, Low Abnormal Control, High Abnormal Control, Low Fibrinogen Control). - Method Comparison Test Set: Sample sizes ranged from
n=47ton=54for each specific parameter compared between the ACL Elite/Elite Pro and the predicate ACL 9000. These were "citrated plasma samples." - Data Provenance: The document does not explicitly state the country of origin. It indicates the submitting company is in Lexington, MA, USA, suggesting the data was likely generated in the US or in compliance with US regulatory standards. The data presented is characteristic of a prospective study, where samples are run on the new device and compared to a predicate or against established control values.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of those Experts
This type of submission for an in vitro diagnostic device (IVD) for coagulation testing does not typically rely on "experts establishing ground truth" in the same way an image analysis AI might.
- Ground Truth for Precision: The "ground truth" for precision is derived from the mean and standard deviation of repeated measurements on stable control materials. These control materials (e.g., Normal Control, Special Test Control) themselves have established target ranges, which are typically determined by the manufacturer of the control material based on extensive testing and/or collaborative studies. There are no "experts" in the sense of human readers adjudicating results.
- Ground Truth for Method Comparison: The "ground truth" for method comparison is the measurement obtained from the predicate device (ACL 9000). The new device's performance is compared against the predicate's measurements. Again, this does not involve human expert adjudication.
4. Adjudication Method for the Test Set
Not applicable for this type of IVD performance study. The data involves quantitative measurements from automated instruments, not qualitative assessments requiring human adjudication.
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
Not applicable. This is an in vitro diagnostic device (automated analyzer) and not an AI-assisted diagnostic tool that involves human readers or interpretation of complex medical images.
6. If a Standalone (i.e. algorithm only without human-in-the loop performance) was done
Yes, the studies reported are standalone device performance studies. The ACL Elite and ACL Elite Pro are automated analyzers. Their precision and method comparison studies evaluate the device's inherent analytical performance without any human intervention in the result generation or calculation process. The studies describe the performance of the "algorithm only" in the sense of the automated analytical process of the instrument.
7. The Type of Ground Truth Used
- For Precision: The ground truth for precision is statistical: the mean values and the expected variability (low %CV) of the control materials, established by the control manufacturer and expected from a well-functioning analytical system.
- For Method Comparison: The ground truth for method comparison is the measurements obtained from the predicate device (ACL 9000). The new device is considered substantially equivalent if its results correlate strongly with, and are statistically similar to, those produced by the predicate.
8. The Sample Size for the Training Set
Not applicable in the context of machine learning model training. This device is an automated analyzer, and its "algorithm" refers to its pre-programmed analytical processes and calculations, not a machine learning model that undergoes a distinct "training phase" on a dataset. The system's operational parameters and calibration procedures are established during its design and manufacturing based on general principles of coagulation testing and validation, rather than iterative training on a specific "training set" of patient data.
9. How the Ground Truth for the Training Set was Established
Not applicable for this type of device. As explained in point 8, there isn't a "training set" in the machine learning sense. The device's calibration and internal algorithms are based on established scientific principles of coagulation measurement, reagent properties, and instrument design, not on a data-driven training regimen with a "ground truth" derived from external data.
§ 864.5425 Multipurpose system for in vitro coagulation studies.
(a)
Identification. A multipurpose system for in vitro coagulation studies is a device consisting of one automated or semiautomated instrument and its associated reagents and controls. The system is used to perform a series of coagulation studies and coagulation factor assays.(b)
Classification. Class II (special controls). A control intended for use with a multipurpose system for in vitro coagulation studies is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 864.9.