Publications

Inselspital and UBern studied Neosoma Glioma's auto-contours, comparing time savings and contour accuracy vs. expert manual contours, concluding that the Neosoma Glioma model generates clinically useful postoperative GTV segmentations, with geometric performance comparable to expert variability and dosimetric equivalence to consensus contours, and reducing contouring time by over 50%, enabling faster RT workflows.

A new study published in the Journal of Neuro-Oncology demonstrates how the ATHENA AI model can transform low-resolution sodium MRI into high-quality, clinically useful images at standard 3T field strength. This breakthrough addresses a major limitation in sodium neuroimaging - traditionally requiring expensive ultra-high field systems or suffering from poor image quality. The research team, led by Dr. Benjamin Ellingson at UCLA, used the Neosoma Glioma AI software for precise tumor segmentation, enabling accurate analysis of contrast-enhancing regions, necrosis, and T2-hyperintense areas.

Neosoma's technology was used to support clinical research by Ben Ellingson's lab to characterize advanced imaging features of recurrent glioblastomas showing a survival benefit from anti-VEGF agents. The authors conclude that post-chemoradiation glioblastomas with a diffusion MRI phenotype that is known to predict a favorable response to anti-VEGF (ADCL ≥1240 µm2/s) have distinct biological features, with different perfusion and metabolic characteristics, and T2 relaxation times.

A new study published in CNS Oncology demonstrates how artificial intelligence and advanced imaging analysis can help predict survival in patients with recurrent glioblastoma, the most aggressive form of brain cancer.


The research, led by Dr. Benjamin M. Ellingson and colleagues at UCLA, utilized Neosoma's innovative NSHGlio AI technology to perform precise measurements of tumor volumes across multiple timepoints. This automated segmentation technology, combined with novel radio-pathomic mapping techniques pioneered by Dr. Peter LaViolette at the Medical College of Wisconsin, enabled researchers to track both volumetric changes and cellular characteristics of tumors over time.

Using the Neosoma AI technology for neuro-oncology, a group of prominent researchers at UCLA have demonstrated a correlation between alterations in tumor growth rate (TGR) and overall survival in patients with recurrent glioblastoma (rGBM) undergoing chemotherapy, with or without radiation therapy.

In December, 2022, Neosoma published an article in Neuro-Oncology Advances detailing its core deep learning technology architecture, training, validation, and performance. This patented technology is currently applied to high-grade glioma tumor segmentation for longitudinal assessment, and will be applied to other neuro-oncology indications.

Neosoma's technology was utilized in a clinical study led by Benjamin Ellingson, PhD, to investigate early volumetric, perfusion, and diffusion MRI changes in IDH1-mutant gliomas during IDH inhibitor treatment.