Behind the patent: A data storage breakthrough

 By Pramod V R

If your healthcare organization is like most, you’re probably facing a data dilemma.

You need as much patient information as possible to deliver exceptional care, but electronically storing that data for future use can expose your company to hefty fines and other penalties if the information isn’t properly protected.

This is the challenge several colleagues and I set out to solve more than two years ago, when we began exploring ideas that ultimately led to the development of record-level sensitivity-based data storage in a hybrid multi-cloud environment (US11455416)¹, or the ‘416 Patent for simplicity.

Now, using the ‘416 Patent, you can safeguard your patients’ most confidential, or sensitive, data while allowing authorized individuals to access less-sensitive information when needed.

In the process, you may also be able to reduce costs, improve efficiencies, and reinforce compliance across your enterprise, converting your data storage obstacles of today into business opportunities for tomorrow.

How the ‘416 Patent works

The ‘416 Patent describes a data segregation process—implemented as a system—that breaks apart clusters, or rows, of data into individual records, or fields, and analyzes the information within each record. The system then partitions the data within the fields to a networked on-premises, public cloud, or private cloud data storage center based on the sensitivity of the information.

So, if an organization implements the process claimed in the ‘416 Patent and identifies information in a data record as extremely confidential, the system can direct the data within that field either to the organization’s on-premises or private cloud data storage center. On-site or private cloud data stores are typically the most secure places to store data, making them ideal for personally identifiable information (PII)² and other highly sensitive information.

On the other hand, if the system identifies a select data field as less sensitive, it can assign information from that record to an interconnected public cloud data center. Data stored on a public cloud hosted by a hyperscaler or another trusted cloud provider remains secure, but the data center likely has different storage parameters, making it more cost efficient than using on-premises or private cloud data stores.

The system described in the ‘416 Patent also allows users to define the parameters that measure the sensitivity of each data record. That means an organization can adjust values for privacy thresholds, which could help in administering the organization’s privacy policies and complying with HIPPA requirements and other regulatory mandates.  

Where the ‘416 Patent may eventually be used

Unlike solutions in search of a problem, the ‘416 Patent process could be applied across a wide range of regulated industries. Healthcare represents a logical starting point.

Hospitals and health systems already rank among the most popular targets for cybercriminals.³ In fact, healthcare organizations in the United States averaged more than 1,400 cyberattacks per organization each week during 2022, an increase of 86% from the year before.

Cyberattacks will continue to escalate as larger swaths of the healthcare industry undergo digital transformation and greater amounts of protected health information (PHI)⁴ are stored electronically.

The need for advanced and configurable cybersecurity measures⁵ in general and HIPPA-compliant data centers⁶ in particular will likely follow suit, challenging both the physical and financial resolve of even the most forward-leaning healthcare companies.

A potential role in healthcare data storage

Given the volume and pace of change taking place within the industry, healthcare organizations must reimagine all areas of care delivery. Data storage and protection will play a pivotal role in this process, meaning implementation of the ‘416 Patent may eventually help providers:

• Reduce storage costs. Using the ‘416 Patent process to send less-sensitive data to the cloud will free up space on on-premises and private cloud data centers and should lower costs associated with acquiring, managing, updating, and optimizing these systems.
• Increase data security. Leveraging the system described in the ‘416 Patent to analyze data at the record level and then partition it to the appropriate data center could offer a more robust solution than either on-premises or cloud-based storage options can provide independently. 
  
•  Reinforce regulatory compliance. The ‘416 Patent system could aid in designing compliant systems for use in other regulated environments that have different storage requirements based on the types of information being housed and protected.

In time, less-sensitive data stored in cloud data centers of healthcare companies could even be aggregated and anonymized with other less-sensitive patient information and used for industry-wide forecasting, disease modeling, and treatment protocols.

How the patent differs from other data partitioning solutions

One aspect of the ‘416 Patent process that truly distinguishes it from other data partitioning approaches—and the feature that may one day revolutionize information storage and protection—is how granular the system can get when reviewing data.

For illustrative and comparison purposes, consider Acme Health Co. and Futuristic Healthcare Inc., two fictional yet representative organizations that each operate numerous hospitals and medical clinics across the US.

Acme uses a traditional segregation process to manage patient information, while Futuristic recently deployed a system as described in the ‘416 Patent to automate its data protection program. Both companies utilize hybrid multi-cloud platforms to store data across their enterprises.

When a health care provider enters a patient’s information into Acme’s data storage center, its system treats multiple rows of information as a single data cluster. The system then examines the data set to determine if it contains sensitive information tags like name, age, gender, and a specific disease diagnosis.

If a single sensitive term is found anywhere within the data record, the entire data set is deemed sensitive and isn’t allowed to be stored in a public cloud. Instead, it must be housed in Acme’s on-premises data center, which is expensive to operate, maintain, and scale.

Futuristic’s health information management system uses similar parameters to classify the sensitivity of patient data. However, rather than viewing rows of data as a cluster, its system uses the process described in the ‘416 Patent to separate each data set into a single record and evaluate it independently.

If Futuristic’s system detects confidential information within a record, it will only send the sensitive data within that record to the company’s on-premises data store. The remaining information in the data record can be stored in the public cloud, where it can be housed and protected more cost-effectively.

Shifting focus from data storage to patient outcomes

Data challenges within your organization shouldn’t disrupt your operation or interfere with delivering quality care.

By embracing innovations like the ‘416 Patent and other proven and emerging technologies, you can worry less about data storage and focus more on patient outcomes.

Within the medical community and beyond, that type of prescription should prove transformative.

Pramod V R is a Master Inventor and an Associate Director on Kyndryl’s Solutioning team.

¹ Record-level sensitivity-based data storage in a hybrid multi cloud environment, Google Patents
² Guidance on the Protection of Personal Identifiable Information, US Department of Labor
³ Keeping Hospitals and Patients Safe against Cyberattacks, American Hospital Association, February 2023
⁴ What is protected health information?, The HIPPA Journal, March 2023
⁵ Data security must be prioritized in the healthcare industry, Forbes, May 2022
⁶ HIPPA-compliant data centers, The HIPPA Journal