CHICAGO, USA

318 West Adams Street, Chicago
IL 60606
USA
Tel: +1 512 999 5796

CLEVELAND, USA

675 Alpha Dr, Suite E, Highland Heights,
OH – 44143
Cleveland, Ohio
Tel: +1 512 999 5796

KOCHI, INDIA

#102, First Floor Lulu Cyber Tower
Infopark, Kochi – 682 042
Kerala, India

×

CONTACT US

Contact

This Blog is part of our continuing blog series on Interopability.In the previous article we spoke about interoperability and AI that are deemed to be a symbiotic relationship for the healthcare industry.You can check it out here. In this blog we speak about How Cabot helped an AI startup achieve interoperability.

Business Problem

Over six months back, an Artificial Intelligence Company approached Cabot with a need, which is, of late, very common in the healthcare provider organizations - to be able to pass information to and from third party EMRs, billing software, PMS, Laboratory systems, pharmacy inventory systems etc at the clinics where their product is implemented.

The client’s flagship product is currently the only comprehensive Oncology A.I Solution which is enabled by a clinical inference engine, supported by a leading and codified content library, and actioned through an embedded smart EMR. The platform helps oncologists generate patient-specific treatment plans in real-time and facilitates clinical decision support at point-of-care. To achieve this, the system maintained a library of diagnostic and treatment plans, where each plan includes a set of codified rules that are used by the engine to constantly review and assess each patient and their clinical data. The system has been designed to go a step beyond existing EMR offering the ability to manage and inference across the patient genomic data.

The proprietary platform provides all standard EMR capabilities one would expect (Scheduling, Labs, Progress Notes, Demographic and ADT data capture, etc) and would need to be accessing a patient’s complete medical history from multiple disparate EMR and other related systems. The system would also ingest a cross-section of different documents from genomic labs and a variety of hospital/clinical systems using PDF, XML and CCD document types.

The underlying basis for most learning systems is that the more the data, the better would be the end results be it making recommendations or forecasts. As you would expect, each implementation of the platform in an oncology office will require significant integration with existing systems and processes. Peer-to-peer knowledge transfer between this platform and EMRs at different healthcare organizations enable physicians to get real-time access to personalized care plans for their patients.The Cabot team was entrusted with the task of providing custom, interoperability services to the client to enable the seamless exchange of patient data with multiple healthcare organizations.

Challenges

Cabot had two ways forward to choose from - establish point-to-point interfaces or use an integration engine.

Point-to-point vs Interface engine

It was clear that given the business model of our client, point-to-point custom interfaces would be totally unfeasible. On a regular basis, many systems would have to be interfaced with client’s platform either/both as message sender and receiver. There will be differences in data structures, formats, transmission protocols and environments to combat during each instance of interfacing and the number of interfaces to be created will also be huge. For example, bi-directional ADT interfacing between three systems would require up to 6 interfaces; for six systems, there will be up to 30.

An interface engine, on the other hand would eliminate the need for individual connections between systems. Being specifically built to connect systems, such an engine manages the message workflow, performing any necessary message transformations, and ensures message delivery. Workflow Integration functionalities were not required for the current scenario and therefore integration engines were not studied.

While commercial off-the-shelf integration engines provide the ability to manage multiple interfaces between systems, messaging standard and protocol support is limited. This meant that if the client had their own proprietary formats that are not compatible with formats designed by interface engines, it could aggravate the problem further. Hence, Cabot proposed to customize messaging interfaces in order to demonstrate successful transmission of messages, customized according to client's requirements, between a previously specified sources / destinations.

Solution proposed

After a thorough research on the interface engines available on the market, Cabot and the client together decided to choose Mirth Connect. Mirth Connect is a cross-platform interface engine used in the healthcare industry for bi-directional sending of messages and can handle multiple sources, destinations, standards/protocols and message formats. It was a unanimous decision to customize the interface engine’s i.e. Mirth Connect’s messaging standards as per client’s requirements. The rest of this article describes how this project helped the client in establishing interoperability between the oncology platform and popular software systems by programming, customizing and configuring Mirth Connect interface engine.

Our first step was to represent the elements of a messaging interface. Sample production messages/message definitions were procured from the proposed source and destination systems to perform gap analyses. Connection types were decided upon and endpoint details received. Proper project team communication protocol was put in place with the points of contact of both the sending/receiving parties. These contacts would be engaged with closely during the development and testing of the interfaces.

In the first phase, support was added for transmission of patient details from disparate EMR systems to our client’s platform. This would involve such details as the demographics & photo, visit details, allergies, insurance, family history and so on as well as details of the admit/discharge/transfer as the case may be. In some cases, the client was receiving some information as separate custom messages and these had to be merged into the standard HL7 message format.

In the next phase, the transmission of prescription details from the client’s platform to a popular Pharmacy Inventory Management was undertaken. Pharmacy Order messages are used by clinical applications to send an order to the pharmacy and/or dispensing systems. It may be sent as an order containing a single pharmacy/treatment order either for a patient or as an order containing multiple pharmacy/treatment orders for a patient. Cabot helped client in accomplishing their needs by creating medication orders from client’s EMR and sending those orders to Oncology Supply’s Inventory Management System. The details sent included prescription details - components /drugs, administration instructions, route, dose, observation and so on, along with patient and visit identification details.

Conclusion

As Pioneers in Healthcare Software Solutions, Cabot develops innovative solutions for aiding organizations to rapidly exchange and acquire health data. At Cabot, we help our customers to promote and scale interoperability for a greater and more efficient access to clinical data by patients and payers.

Interested in Integrating EMR Systems for exchanging healthcare information? Contact Us today

SHARE THIS BLOG ON

test

STAY UP-TO-DATE WITH US

Subscribe to our newsletter and know all that’s happening at Cabot.

SHARE