Search Results

Obsidio GEM™ is indicated for use in the embolization of:

· Hypervascular tumors,

· Blood vessels to occlude blood flow for controlling bleeding/hemostaging in the peripheral vasculature.

GEM™ consists of a pre-hydrated bioresorbable gelatin, layered silicate (Laponite) mixture, and tantalum powder packaged in a 1 ml syringe with a standard luer lock tip. The device is provided ready to use. The device is injected into a target vessel through intravascular catheters to provide a mechanical barrier to blood flow. The device is intended for single use and is provided sterile.

The provided text is a 510(k) summary for a medical device called Obsidio GEM™. It describes the device, its intended use, technological characteristics, and performance data to demonstrate substantial equivalence to a predicate device (EmboCube™ Embolization Gelatin).

However, the document does not describe an AI/ML-driven device or an acceptance criteria table with reported device performance in the context of an AI/ML model's output metrics (e.g., sensitivity, specificity, AUC). Instead, it discusses the performance of a physical vascular embolization device through non-clinical testing and animal studies.

Therefore, I cannot fulfill your request for:

1. A table of acceptance criteria and the reported device performance (for an AI/ML device): The document doesn't contain such a table, as it's not about an AI/ML device.
2. Sample size used for the test set and data provenance: While animal study sizes are mentioned (e.g., "ten test (GEM™) and nine control (EmboCube™) adult swine"), these relate to biological studies, not a test set for an AI/ML model.
3. Number of experts used to establish ground truth & qualifications: N/A for this type of device.
4. Adjudication method for the test set: N/A for this type of device.
5. Multi-reader multi-case (MRMC) comparative effectiveness study: N/A, as it's not an AI-assisted diagnostic device.
6. Standalone (algorithm only) performance: N/A.
7. Type of ground truth used: Animal study outcomes, not expert consensus on images or pathology for an AI system.
8. Sample size for the training set: N/A, as there's no AI model being trained.
9. How ground truth for the training set was established: N/A.

The document focuses on:

• Non-clinical performance testing: Macroscopic inspection, component characterization, rheometry, pH test, injection force, catheter deliverability, sterilization validations, packaging validations, sterility, pyrogenicity, and biocompatibility.
• Animal studies:
  • Hypervascular tumors: Ten GEM™ and nine EmboCube™ swine, across 38 renal arteries and 19 hepatic arteries, evaluated at 4, 30, and 180 days. Endpoints included in-life health, performance, occlusion, arterial flow, regional/downstream response, and tissue analysis.
  • Simulated hemorrhage: Five GEM™ and five EmboCube™ swine, in eight renal and eight splenic arteries, over 7 days. Endpoints included in-life health, performance, hemostatic success, macroscopic regional tissue response, time to hemostasis, and embolization effectiveness.

The conclusion is that "The product met its specifications. All test results confirmed that the technological differences between the subject and the predicate device did not adversely impact its performance." and that "GEM™ is substantially equivalent to the predicate device, EmboCube™ Embolization Gelatin."

In summary, the provided document details the testing and substantial equivalence argument for a physical medical device (vascular embolization device), not an AI/ML-driven diagnostic or assistive device. Therefore, the requested information pertaining to AI/ML model evaluation cannot be extracted from this text.

Ask a specific question about this device