SaMD for Computer-Aided Detection of Neurological Disorders


Software DevelopmentImage Processing


The work conducted by Innolitics to complete a complex software port of a core module of our clinical product was outstanding. The clinical product must by cleared by the FDA, and as such, we needed the highest quality workmanship, and close collaboration with a dispersed team. Innolitics performed exceedingly well in gathering requirements, researching, developing and implementing a design, and providing test modules and documentation. Their team's understanding of medical imaging analysis software, and FDA quality processes, was key to the success of the project.

Nick Schmansky

Nick Schmansky

Co-Founder and CEO of CorticoMetrics

The Problem

CorticoMetrics LLC, was dealing with efficiency issues related to their neuroimage analysis software 'FreeSurfer,' which had a dependency on MATLAB. This software was composed of various command-line tools to process MRI scan data of the human brain and convert it into quantitative measures. However, its reliance on MATLAB led to a slow runtime. The primary challenge was to increase the software's speed and reduce dependency on MATLAB. To do this, the client required the Matlab portion of the software to be ported over to Python, a programming language chosen for its advantages in this application. In addition, the client sought to ensure that the newly converted Python software maintained compatibility with commercial usage and did not violate any licensing regulations.

The Solution

The AutoRegister project is: A software-based system for an MRI scanner to reduce the error in tumor measurement.

Overview of the AutoRegister system

MRI assessments are extremely sensitive to the subject’s position inside the machine. This sensitivity can be mitigated by using image data from the subject’s previous MRI sessions. AutoRegister aligns acquired imaging data to previous data from the same subject to improve the accuracy of tumor growth detection with an MRI.

CorticoMetrics developed a working prototype of this software but needed a partner to bring it to market. The company hired us because of our experience in medical imaging, especially with DICOM, image registration, and IEC 62304. Our engineers were able to re-implement their prototype and get it ready for a 510(k).


The engineer at CorticoMetrics was not familiar with integrating software into an MRI machine and hospital PACS. We were contracted to get AutoRegister integrated with these systems. By hiring us, CorticoMetrics was able to integrate its product into industry testing quicker than if they were to rely only on in-house engineers.

Diagram of AutoRegister's PACS server configuration

On-site Testing

Reserving a time to use an MRI in a hospital can be costly and hard to acquire due to busy hospital schedules. In order to make the most use of their MRI appointments, CorticoMetrics needed to be able to make on-site adjustments to the software during testing. Our team ensured the engineer on staff was familiar with the configuration of AutoRegister to the Hospital IT so he could make the most use of his time testing the software.

Machine Simulation

CorticoMetrics needed a way to test AutoRegister without having to buy an MRI machine. To solve this issue, we developed an MRI emulator to test AutoRegister as if it were integrated with an MRI machine.

Integrated Testing

In order to get AutoRegister approved by the FDA, CorticoMetrics needed to ensure the product was tested. Our team delivered integrated tests to ensure the product was ready for FDA approval.


Submitting a 510(k) for a product requires documentation. Our team provided thorough documentation for our services, delivering software that is suited for 510(k) review.

After completing our contract for the AutoRegister product, the prototype of the software was rewritten, validated, documented, and they were on track to submit their first 510(k). In all, we brought more robustness to the software team at CorticoMetrics and helped them expedite the process to bring their product to the market.

About Nick Schmansky

Nick Schmansky
Co-Founder and CEO of CorticoMetrics

Nick Schmansky is the Co-Founder and CEO of CorticoMetrics LLC, a position he has held since June 2012. With his extensive background in Cognitive and Neural Systems and Artificial Intelligence, Schmansky is a prominent figure in his field. He earned his Master of Science degree in Artificial Intelligence from the University of Edinburgh in 1999, followed by a Master of Arts degree in Cognitive and Neural Systems from Boston University in 2005. His solid academic background is backed up by an impressive professional track record. Before launching CorticoMetrics, he worked for over a decade as a Software Engineer at the Massachusetts General Hospital, where he made significant contributions in the field of medical imaging.


  • Initially, we were grappling with the slow runtime of our 'FreeSurfer' software due to its dependency on Matlab, but the Innolitics efficiently ported the Matlab code to Python, significantly enhancing the overall runtime speed and directly improving our processing capabilities.
  • The complexity and lack of accessibility of our software were major challenges due to extensive use of Matlab; however, transitioning to Python, undertaken by Innolitics, made the software more comprehensible and accessible, especially to users with less programming experience, thereby expanding our user base.
  • We were concerned about potential licensing issues due to the dependency on Matlab, but by eliminating this dependency, Innolitics ensured that our 'FreeSurfer' software remained compatible with licenses like MIT, BSD, or Apache 2.0, mitigating any risk of violating licensing regulations.