OnQ Neuro

OnQ Neuro is a medical image processing software designed to automatically analyze neurological MR images by segmenting and quantifying brain tumors, performing diffusion-weighted imaging analysis, and enabling longitudinal comparisons. Its methodology relies on automated algorithms validated against expert clinicians to provide quantitative metrics and visualizations intended to assist in the evaluation of pathologically confirmed brain cancers.

OnQ Neuro is a fully automated medical image post-processing software designed for the analysis and interpretation of neurological MR images in adult patients. Its primary functions include the automatic segmentation and volumetric quantification of pathologically confirmed brain tumors, post-acquisition analysis of diffusion-weighted MRI (DWI) data, and the automated fusion of derived image data with anatomical MR images. The software accepts DICOM images and can compare derived image metrics from multiple time-points to assist clinicians like radiologists and oncologists. OnQ Neuro operates as a standalone package, installable in the cloud or within a hospital's IT infrastructure, processing images automatically from MRI scanners or PACS without requiring a user interface post-installation. Outputs are provided in standard DICOM format as image series, numerical volumes, grayscale intensity maps, graphical color overlays, and reports, serving as complementary information for clinical review. The scientific basis of OnQ Neuro involves advanced image processing techniques for neurological MR data. It employs algorithms for automatic tumor segmentation, quantification of tumor volume, and analysis of DWI to generate metrics such as Apparent Diffusion Coefficient (ADC) maps and Restricted Signal Maps, which characterize water diffusion in brain tissue. The system also supports longitudinal analysis by comparing imaging data over time. Performance testing demonstrated that OnQ Neuro's automated segmentation accuracy, evaluated against expert manual segmentations using Dice similarity coefficient and R2 for volumes, met predefined acceptance criteria, showing consistency across different scanner types, field strengths, tumor characteristics, and patient demographics. Validation studies confirmed the correct generation and clinical utility of its outputs, including tumor segmentation overlays, reports, and diffusion maps, affirming its suitability for clinical use as intended.

AI Regulatory Concerns

Powered by

Jev reads the full FDA letter and returns calibrated concern probabilities per regulatory category. These are AI estimates, not regulatory advice.

• Q1. How does the algorithm for automatic brain tumor segmentation in OnQ Neuro account for variability in tumor morphology, location, and post-treatment changes?; What specific diffusion models are employed for the DWI analysis, and how does the 'Restricted Signal Map' enhance diagnostic information compared to standard ADC maps alone?; What were the quantitative performance metrics (e.g., mean Dice coefficient, R2 values) achieved in the validation studies comparing OnQ Neuro's automated segmentation to expert manual segmentations across different tumor types?; How does OnQ Neuro handle potential image artifacts or variations in MR acquisition protocols from different scanners when performing automated analysis?; For longitudinal analysis, what specific image registration techniques are used to align images from multiple time-points, and what is their reported accuracy?