Precision neuro-imaging in both research & clinical care has the potential to deliver personalised diagnostic and treatment approaches. Artificial intelligence is expected to play a huge role in transforming neuro-imaging, through automation and increased productivity, enhanced reporting accuracy and the rapid identification of critical abnormalities.
In SNAC, we are committed to develop artificial intelligence algorithms and software platforms that seamlessly integrate with radiology clinical workflows. Key methods in deep learning technology, including classification, segmentation and augmentation, are employed to solve specific problems in the domain of neuro-imaging.
The number and spread of lesions in the patient’s parenchyma has become crucial information about the patient’s disease status. However, a manual operation by radiologist is a tedious and time-consuming task.
A deep learning model with segmentation techniques can shrink the time needed to determine lesion volume to just 3 seconds. This is a significant saving in time and cost, and makes it possible for using such quantitative analysis result in clinical practice.
This image shows a side-by-side comparison of multiple sclerosis lesion segmentation. Left image shows manual lesion segmentation, while right shows fully automated lesion segmentation, performed by AI algorithms developed by SNAC.
Intracranial hemorrhage is a critical abnormality requiring urgent and intensive medical treatment. However, reviewing the medical images of the patient’s cranium is a complicated process and often time consuming. AI algorithm with deep learning models, trained with dataset that is golden labelled by certified radiologists, is able to detect acute intracranial hemorrhage in CT images within seconds.
Automatic triage results are reflected on RIS system, informing radiologists to review critical cases as a priority.
Multiple sclerosis white matter lesions have a heterogeneous appearance on T1-weighted MRI images that reflects the severity of tissue injury. Hypointense white matter lesions are often misclassified as grey matter or cerebrospinal fluid by automated segmentation algorithms, resulting in biased brain substructure volume assessment.
Using inpainting techniques by imaging augmentation based on AI algorithm can mitigate this issue, thus improve the accuracy and consistency of brain volume measurement results from automatic analysis pipelines.