Rayscape | Frequently Asked Questions

Product Information

Who benefits from Rayscape's services?

Rayscape's solution caters to a wide range of medical conditions commonly found across various healthcare settings, including private chains of clinics, outpatient centers, and public hospitals.

What types of medical imaging does Rayscape support?

Rayscape can analyze chest X-rays, both PA and AP views. For CT scans, Rayscape supports lung CT scans, TAP scans with lung slices, 3D reconstructions, both contrast and non-contrast scans, multi-slice and single-slice scans, with a slice thickness of less than 3mm.

Does Rayscape provide real-time analysis of medical scans?

Rayscape offers real-time analysis of medical scans in under 1 minute (the processing time for X-rays is < 30 seconds while for CT is < 1 minute). The seamless integration of these results directly into the patient's file means there's no need to rely on external applications or screens to access the valuable findings.

What distinguishes Rayscape from other AI-powered radiology platforms?

Rayscape CXR stands out by ensuring advanced features like bone suppression and subtractions, as well as innovative measurements such as the cardiothoracic index. Additionally, Rayscape Lung CT enhances diagnostic accuracy by facilitating comparative analysis, a valuable tool for conducting nodule follow-up investigations, Fleischner follow-up recommendations and nodules attributes.

Data Security and Compliance

How does Rayscape ensure the confidentiality and security of patient data while preventing unauthorized data access?

Rayscape takes stringent measures to safeguard patient privacy and data integrity. Patient information and scans are anonymized from the moment of extraction, remaining within the hospital infrastructure throughout the analysis process until they are safely returned. Notably, access to this data is restricted solely to authorized personnel and is granted only upon request. This multi-layered approach ensures that patient confidentiality remains paramount throughout the entire process.

Is our company compliant with relevant data protection regulations, such as GDPR?

Our company takes data protection seriously. We have taken the necessary steps to ensure that we are fully compliant with all relevant regulations, including the General Data Protection Regulation (GDPR).

Is patient data shared with third parties or other organizations?

No, the strict protection of patient data is a fundamental principle in our practices. We uphold the highest standards of confidentiality and adhere to a strict policy of not sharing any patient data with third parties or external organizations.

Which regulatory approvals does Rayscape have?

Rayscape has successfully secured multiple regulatory approvals, notably including ISO 13485, CE class I, and MDR certifications. These endorsements underscore our commitment to meeting stringent quality and safety standards.

AI Technology and Training

How does AI validate its algorithms' performance and safety?

Safety validation research is at the core of our work. This is why we conducted two clinical trials and an additional 10 peer-reviewed studies that dissect any potential risks associated with our AI. Moreover, over the years, we have built a robust pre-deployment validation dataset against which we measure each algorithm update, in accordance with our ISO 13485 compliant quality management system. Nonetheless, the external audits required for CE marking ensure an additional safety layer.

What is the "accuracy" of Rayscape's algorithms? What value do they add?

Different measures provide various perspectives, and the multitude of medical and computer vision metrics can be challenging to contextualize. The most effective way to gauge our algorithm's proficiency is by considering the tangible benefits they offer to radiologists. Validatory studies have demonstrated that the utilization of Rayscape results in a 12% increase in sensitivity for lung nodule detection, a 30% reduction in turnaround time, and reduce human bias by up to a factor of ten.

How does Rayscape handle and mitigate potential algorithm bias?

Careful collection, curation and distribution of patient subpopulations are paramount in ensuring our algorithms' fairness. Moreover, in depth quantitative analyses of each subpopulation, at validation time, prevent any skeweness the AI might usually be inclined to model.

How well do Rayscape's AI algorithms generalize across devices, geography, subpopulations, and data?

Generalization challenges are a common concern in AI, often resulting in notable performance declines in real-world scenarios. At Rayscape, we take immense pride in our dedicated approach to addressing this issue. Not only have we meticulously researched and published insights into the mechanisms contributing to such challenges in medical imaging, but we've also developed practical techniques to effectively mitigate their impact. Moreover, our robust dataset of 15 million images sourced from 26 countries across 5 continents, captured using different acquisition devices, positions us uniquely with unparalleled generalization capabilities within the industry.

Integration and Support

What is the timeline for complete integration and deployment of Rayscape, and which radiology systems does it seamlessly integrate with?

The integration timeline is notably efficient; gaining access to a virtual machine (VM) within a hospital, for instance, can take less than a week. Rayscape seamlessly integrates with any Picture Archiving and Communication System (PACS) compliant with the DICOM standard.

Does integrating Rayscape necessitate hardware upgrades or extra infrastructure? What types of integrations are supported by Rayscape?

The integration process requires specific hardware specifications. For instance, a server or VM with 4 CPU cores, 8 GB RAM, and 500 GB storage, or more powerful hardware such as 16 CPU cores, 32 GB RAM, 1 GPU with at least 12 GB VRAM, and 1000 GB storage for local processing. Rayscape supports both on-premise, hybrid cloud, or pure cloud integrations to cater to different infrastructural needs.

What maintenance is involved in using Rayscape, and how frequently can we expect updates to be released?

Following the initial installation, which is facilitated with the assistance of clinic/hospital IT staff, all subsequent configuration and updates are conducted remotely by Rayscape personnel. However, if any hardware or networking issues arise, local IT staff may be required to assist.

How can Rayscape be scaled to accommodate the expansion of our organization? What are the constraints or limitations of Rayscape's scalability?

Scaling Rayscape is a flexible process. For local deployments, expansion can be achieved by adding more servers. In hybrid deployments, most scalability enhancements can be executed remotely by augmenting the processing power of our cloud processor. The scalability of Rayscape is primarily influenced by the networking infrastructure of the hospital or clinic.