METLiT AI Solution

Faster.   Safer.   Easier.


METLiT Specialized

METLiT can solve pharmaceutical companies' problem
by providing customized MRS analysis solution with
Deep learning.

METLiT Specialized solution helps to determine the
effectiveness of new drugs by acquiring information on
direct metabolite changes in vivo, without relying on the
patient's symptoms.

Our solution provides quantitative and objective data to

optimize clinical studies which helps to ascertain
the most effective dosage and material of new drugs.

METLiT Diagnostics

Just as the development of medical imaging has advanced medicine in the 20th century,

the next innovation for medicine is providing substantial in vivo information without invasive measures.

METLiT Diagnostics

Our company is working to commercialize MRS through our AI Analysis/Diagnostic solution.
If MRS becomes commercialized, it can be performed within 3 minutes following a 20–30-minute MRI scan.

METLiT Diagnostics

Through the addition of MRS, information can be acquired for more accurate diagnostics.
The commercialization of MRS would provide a safer and more innovative medical service for patients. In addition,
it would allow medical staff to access more information for clinical evaluation
and provide hospitals with a diagnostic method that does not require additional equipment or manpower.

METLiT diagnostics solution uses deep learning for metabolite quantification based on in vivo
MRS metabolic information.

We aim to diagnose numerous diseases related to metabolites, such as

  • Brain Tumor, Stroke
  • Multiple Sclerosis
  • Liver fibrosis/cirrhosis/tumor
  • Other diseases linked to metabolites, as well as diseases that do not cause structural changes

Our achievements

Since 2012,

Deep learning technologies have been substantially advanced and our company
has overcome various limitations through the application of this technology.

As a result, we're able to achieve


Fast Data Acquisition

It is possible to obtain MRS data on the lesion and the contralateral normal under 3 minutes (2 voxels)[¹]
~ 46%Competitors

Diagnostic Accuracy Improvement

Diagnostic accuracy improved to 80% compared to competitor's accuracy of 46% [²]
27th ISMRM Summa Cum Laude Merit Award

27th ISMRM Summa Cum Laude Merit Award

(ISMRM 27th Annual Meeting 2019 - Summa Cum Laude Merit Award)

Based on these achievements

our solution has great potential for use in various medical research fields and clinical treatment.


References 1. * Lee, Hyeonghun, and Kim, Hyeonjin. "Intactmetabolite spectrum mining by deep learning in proton magnetic resonance spectroscopy of the brain." Magnetic resonance in medicine 82.1 (2019): 33-48.

검색 닫기