platoseedplatoseed
Higher resolution MRI with faster scan times.
MICSI is introducing AI software that doubles resolution and halves scan time. This breakthrough enables imaging centers to significantly enhance their capacity and patient throughput, potentially saving countless lives and generating an additional $2 million of revenue per MRI scanner. Our initial offering serves as a stepping stone toward the company’s larger vision of transforming the MRI into a truly quantitative instrument that is capable of providing highly reproducible data for more accurate diagnoses and patient management.
MICSI is a software company that upgrades existing MRI and PET scanners with AI-powered denoising and image enhancement to achieve higher throughput and improved image quality. Their MICSI-RMT software suite uses MP-PCA-based denoising and supports quantitative diffusion imaging, aiming to make MRI more accessible and cost-effective. They have FDA clearance for MRI image enhancement and are pursuing FDA submission for MICSI-PET.
MICSI-RMT is a software suite that enhances diffusion and functional MRI image quality using MP-PCA denoising, leveraging redundancy across MRI measurements to remove noise. It includes processing modules for weighted linear least squares and Bayesian fitting to facilitate quantitative diffusion analysis, plus DICOM data routing for secure, seamless integration into medical workflows. MICSI-PET is under FDA 510(k) submission. The platform aims to upgrade existing MRI and PET scanners to higher-throughput, higher-value imaging assets, reducing the number of images required and supporting fewer repeat scans while improving resolution and diagnostic sensitivity.
Who it’s for: Medical imaging providers, radiology centers, hospitals, and MRI/ PET imaging facilities seeking higher throughput and better image quality from existing scanners.
FDA clearance for MICSI-RMT; FDA submission for MICSI-PET; emphasis on clinical validation and broad industry adoption
Gregory Lemberskiy (CEO) co-founded MICSI with goal of bringing cutting edge image processing tools, that he encountered in the research setting into clinical practice. His PhD work (2019) focused on algorithm development for image enhancement and biophysical modeling of the MRI signal. His methods have been used to enable high-end quality on low field MRI systems through noise reduction (MICSI’s first product) and use MRI to characterize the physical properties of prostate glandular lumen.
I got my PhD in electrical engineering in 2022 from NYU, where I specialized in developing machine learning based computer vision algorithms for the visualization of MRI data. I specialize in developing software to process and route diffusion and functional MRI data, with a goal of improving image signal, reducing noise, and producing clinically viable diagnostic imaging biomarkers.
Higher resolution MRI acquired faster
MICSI introduces MICSI-RMT, a self-supervised AI image processing algorithm that reduces MRI scan time and increases image resolution by denoising across the exam, enabling microstructure imaging. Targeted at MRI centers for higher throughput and potential diagnostic improvements.
From the original launch (Aug 2023) — may be outdated.
Cancer screening without barriers
AI Diagnosis of Heart Disease
