Frequently Asked Questions

Ferrum organizes the deployment and workflow of artificial intelligence through our end-to-end platform that can run multiple AI algorithms across different service lines. 

A recent Health It Analytics article noted that while 90% of hospitals have an AI strategy in place, only 25% of hospitals are implementing AI. When asked why the low adoption, the reasons were clear – cost, security issues, staff resources, algorithm selection, clinical efficacy, and the ability to scale. Deploying AI is a complicated process for a healthcare system. 

Ferrum Health addresses all these challenges with our turnkey AI platform, enabling your clinicians to focus on their patients and the care they need, not complicated technology deployment.

Your health system can deploy as many AI algorithms as needed to meet your clinical and administrative needs. Typically, health systems launch with the deployment of four AI algorithms, adding additional use cases over time.

The most common AI algorithms deployed one the Ferrum Platform include:

  • Lung nodule
  • Fracture detection
  • Pulmonary embolism
  • Pneumothorax 
  • Liver lesion
  • Breast lesion

The Ferrum Platform is available globally. Currently, we work with leading health systems across the United States, Australia, New Zealand, and Asia.

Ferrum is a platform company with a growing catalog of healthcare AI algorithms available for use by our health systems. Ferrum has developed/co-developed some of the algorithms in our catalog, but the majority come from best-in-class 3rd party vendors. 

Ferrum’s goal is to provide our health systems with access to validate AI algorithms that meet a variety of clinical needs.

This is a common question we are asked and one that is often explored in the media. The answerer is no, AI will not replace clinicians. There is a radiology professor who says, ‘I don’t think AI will replace doctors, but doctors who use AI will replace doctors that don’t.’

AI is a tool that augments the work of the clinician, helping to reduce their workload. Using radiology as an example, AI provides a set of ‘machine learning’ eyes for a second review of all images. This creates an opportunity for learning while improving patient care.  

The Ferrum implementation process includes algorithm validation on your health system’s patient population, ensuring the applications selected work correctly for your patients. 

The Ferrum Platform supports the evaluation of AI and ML applications on your specific patient population. This allows you to know for sure the algorithm works as promised on your patients and eliminates the need to rely on validation data from whitepapers or testing on a random demographic. You can use this information to evaluate additional ML Applications to see which one works best for your health system.

After going live, the Ferrum Platform provides continuous monitoring capabilities to identify and warn if an ML application exhibits data drift or other anomalous behavior. This ensures the ML application continues to perform as you expect.

Your data is hosted on-premises and is encrypted both at rest, with full disk encryption, and in transit, using the industry standard AES cipher and TLS protocol. 

The Ferrum Platform is further locked down by software firewalls that prevent unwanted network traffic from entering the Ferrum Appliance, and unauthorized network traffic from leaving the system. 

Built into the Ferrum Platform is a comprehensive SIEM solution that provides file integrity monitoring, intrusion detection, container security, and more. This allows for easy incorporation into an IT team’s existing monitoring solutions.

Your health system PHI stays on site and is deleted when no longer needed.

The Ferrum Platform is integrated with the health system RIS and PACS devices using standard protocols such as HL7 and DICOM. Using these healthcare-specific protocols, patient data is sent to the Ferrum Platform and immediately de-identified and anonymized before being sent over to ML applications for further processing. This prevents any machine learning from operating on unnecessary PHI and protects you patient data.

A patient safety organization (PSO) is a group, institution, or association that drives quality care and patient safety by reducing medical errors. PSOs collect and analyze data voluntarily reported by healthcare providers to help improve the care process. PSOs provide feedback to healthcare providers aimed at promoting learning and preventing future patient safety events. 

Working with a PSO makes it possible for information from healthcare providers to receive certain legal protections and allows the data to be contributed to the Network of Patient Safety Databases (NPSD). These protections are in place to facilitate peer learning as a tool to improve patient care.

Ferrum Health’s National Patient Safety Organization (PSO) is a leading advocate in the adoption of technology to advance patient safety. The PSO enables peer learning through the sharing and protection of data, allowing the quality checks we conduct to be non-discoverable. 

The mission of the Ferrum PSO is to guide and educate healthcare systems in the application of technology that drives quality, safety, and excellence in patient care.

The Ferrum PSO is in place to encourage data sharing and peer learning as ways to improve patient care.

Contact Us


ARA Health Specialists

Use the button below to download your free case study and learn how our approach to validation has improved the number of clinically significant findings in AI software.


Sutter Health

Use the button below to download your free case study and learn how our approach to validation has improved the number of clinically significant findings in AI software.