(111 days)
Topical control of bleeding from minor cuts and abrasions of the skin surface
BloodSTOP and BloodSTOP iX Hemostatic Gauze are made from regenerated cotton cellulose, chemically treated to become water-soluble. When contacting blood and exudates, they expand into clear gel, thereby adhering and creating pressure to seal the wound.
The provided document is a 510(k) premarket notification for the BloodSTOP and BloodSTOP iX Hemostatic Gauze. It is not a study that describes acceptance criteria and proves the device meets them in the context of an AI/ML medical device.
The 510(k) submission primarily focuses on demonstrating substantial equivalence to a legally marketed predicate device (Seal-On Hemostatic Powder Spray, K010933) for its intended use, which is "Topical control of bleeding from minor cuts and abrasions of the skin surface."
Therefore, based on the provided document, I cannot fulfill your request for the following reasons:
- No Acceptance Criteria for AI/ML performance: The document does not define performance acceptance criteria for an AI/ML device, as it is not an AI/ML device. It's a wound dressing.
- No Reported Device Performance relevant to AI/ML: The performance reported is related to hemostasis time in animal studies, comparing it to the predicate device. This is not performance data for an AI/ML system (e.g., sensitivity, specificity, accuracy).
- No Test Set/Data Provenance for AI/ML: There is no mention of test sets, data provenance, or ground truth for an AI/ML model.
- No Expert Ground Truth establishment: The document does not discuss experts establishing ground truth for a test set.
- No Adjudication Method: Adjudication methods are not applicable here.
- No MRMC Comparative Effectiveness Study: There is no mention of a multi-reader multi-case study, as this device does not involve human readers interpreting outputs.
- No Standalone Algorithm Performance: The concept of standalone algorithm performance is irrelevant for a hemostatic gauze.
- No Ground Truth Type relevant to AI/ML: The "ground truth" for this device would be direct observation of hemostasis, not expert consensus, pathology, or outcomes data in the context of an AI/ML model.
- No Training Set Size/Ground Truth for Training Set: Since it's not an AI/ML device, there are no training sets or ground truth established for training.
What the document does describe (and how it relates to your request's concepts, even if not directly fulfilling it):
- Device Performance (not for AI/ML): "Animal studies included in this submission show that BloodSTOP and BloodSTOP iX are equivalent or faster than the predicate Seal-On Powder Spray in time to hemostasis." This is the primary "performance" discussed.
- Ground Truth (implicitly for animal studies): The concept of "time to hemostasis" in animal studies implies direct observation or measurement of when bleeding stops, which would serve as the "ground truth" for that specific performance metric.
- Comparison to Predicate Device: The study's "acceptance criteria" can be implicitly understood as demonstrating substantial equivalence to the predicate device, particularly regarding safety and effectiveness (hemostasis time and biocompatibility).
In summary, the provided FDA 510(k) notification for BloodSTOP and BloodSTOP iX Hemostatic Gauze is for a physical medical device and therefore does not contain information about acceptance criteria, study design, or performance metrics relevant to an AI/ML-driven device.
N/A