Search Results

The Zap-X Radiosurgery System is intended to provide treatment planning and image-guided stereotactic radiosurgery and precision radiotherapy for tumors, lesions and conditions in the brain, head and neck when radiation treatment is indicated.

The Zap-X Radiosurgery System ("Zap-X System") is a computer-controlled system for performing non-invasive stereotactic radiosurgery that is self-shielded for ionizing radiation. A gantry-mounted linear accelerator provides the Zap-X System with a source of therapeutic radiation and a kV imaging system is used to accurately locate the treatment target. At the start of treatment, X-ray images of patient skeletal anatomy serve to align the treatment target with respect to the system isocenter. During radiosurgical treatment, the kV imaging system of the Zap-X System tracks patient movement and adjusts the table precisely to compensate for such movement.

The provided text is a 510(k) premarket notification for the Zap-X Radiosurgery System. This document focuses on demonstrating substantial equivalence to previously cleared predicate devices, rather than proving a device meets specific acceptance criteria through a clinical study or a study directly assessing performance against quantitative criteria (e.g., accuracy metrics for an AI algorithm).

The nature of this document (a 510(k) for a radiation therapy system) means it presents a technical and safety comparison to existing devices, not a study evaluating AI performance against predefined acceptance criteria for a diagnostic/AI-driven device. Therefore, many of the requested items (e.g., sample size for test set, number of experts for ground truth, MRMC study, standalone performance) are not applicable or measurable from this document.

Here's an analysis of what can be extracted and what cannot, based on the provided text:

Acceptance Criteria and Reported Device Performance

This document does not present explicit "acceptance criteria" in the format typically seen for AI/diagnostic devices (e.g., "sensitivity must be >X%, specificity >Y%"). Instead, it focuses on demonstrating that the Zap-X Radiosurgery System is as safe and effective as the predicate device(s) and performs as intended through non-clinical testing. The "performance" discussed relates to the engineering and functional performance of the radiotherapy system itself.

• Electrical safety and electromagnetic compatibility testing
• Software verification and validation testing (System and subsystem verification, system validation of commissioning, treatment planning, and treatment delivery)
• Usability testing
• Standards conformance testing (IEC 60601-1, IEC 60601-1-2, IEC 60601-2-1, IEC 60825-1, IEC 61217, IEC 62083) |
  | Radiation Leakage/Protection: The system must protect operators and the general public from radiation. | "The Zap-X System was demonstrated to meet the requirements for radiation leakage and provide protection from radiation to the operator and general public similar to that of CyberKnife within a radiation shielded vault." |
  | Technological Characteristics Comparability: Key technical features should be comparable to predicate/reference devices. | Detailed comparison table provided for: Regulation Number, Classification Product Code, Indications for Use, Accelerator (treatment beam), Dose rate, Depth of Dose Maximum, Treatment Beam, Moveable Treatment Beam, Patient Table/Couch, Shielding for ionizing radiation, Real-Time Dosimetry, Safety subsystem, System console/user interface software, Treatment target tracking software, Treatment planning software, Treatment delivery software. |

1. Sample sizes used for the test set and the data provenance:

• Sample Size: Not applicable. This document describes non-clinical bench testing, software V&V, and physical device performance verification, not an AI model trained or tested on a dataset of patient cases. There is no "test set" of patient data in the context of an AI algorithm described here.
• Data Provenance: Not applicable. The "data" comes from engineering tests, software logs, and measurements on the physical device, not patient data from specific countries or types of studies.

2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

• Not Applicable: This process does not involve establishing ground truth from experts on a patient image or clinical dataset. The "ground truth" for this device relates to engineering specifications, safety standards, and functional performance, which are verified through standard testing procedures by qualified engineers and testers.

3. Adjudication method (e.g. 2+1, 3+1, none) for the test set:

• Not Applicable: No expert adjudication of clinical data or images is involved as this is a physical radiation therapy system, not an AI diagnostic device.

4. If a multi-reader multi-case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improved with AI vs without AI assistance:

• Not Applicable: This is not an AI-assisted diagnostic device; it is a physical radiosurgery system. Therefore, no MRMC study, human reader improvement, or AI assistance is relevant.

5. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:

• Not Applicable: No AI algorithm in the diagnostic sense is described. The "performance" is of the complete physical system.

6. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):

• Not Applicable in the traditional sense for AI/diagnostic devices: The "ground truth" here is based on engineering specifications, international and national standards (e.g., IEC standards), established physics principles of radiation therapy, and the performance characteristics of predicate devices. Verification ensures the device meets these pre-defined, measurable engineering and safety parameters.

7. The sample size for the training set:

• Not Applicable: This device is a physical system with integrated software for control and planning. It is not an AI system trained on a "training set" of patient data like a typical deep learning algorithm. Software verification and validation (V&V) is performed, which involves testing against requirements, but this is distinct from training an AI model.

8. How the ground truth for the training set was established:

• Not Applicable: As there's no "training set" of patient data in the AI sense, there's no ground truth establishment for it.

Summary of Document's Purpose:

This Section 510(k) submission for the Zap-X Radiosurgery System demonstrates substantial equivalence to existing predicate devices (CyberKnife M6 Systems and Leksell Gamma Knife Icon). The "study" proving its capabilities is primarily a comprehensive suite of nonclinical bench testing, software verification and validation, usability testing, and standards conformance testing. The acceptance criteria are implicitly that the device performs as intended, meets its design specifications, and is as safe and effective as the legally marketed predicate devices without raising new questions concerning safety or effectiveness. No clinical trials or AI performance evaluations with patient data are discussed or required for this type of submission for this device.

Ask a specific question about this device