Automatic brain tumor segmentation with subregions

automatic brain tumor segmentation patient's MRI that was automatically segmented 3 tumor subregions (edema, enhancing tumor and necrosis) and provide volume and RANO (Response assessment in neuro-oncology) measurements, DIOM, regression

Our models automatically perform brain tumor segmentation and segment 3 tumor subregions (edema, enhancing tumor and necrosis) and provide volume and RANO (Response assessment in neuro-oncology) measurements. It can be a useful tool for monitoring patients in clinical studies.

We are able to develop custom automated and semi-automated algorithms that analyse the brain and brain lesions, according to your need. The following example illustrates how our algorithms have been used to assess brain tumor segmentation, treatment and progression.

OUR WORK 

AI algorithm for brain tumor segmentation

The following example demonstrates the limitations of RANO measurements when compared with full volumetric measurement of a segmented image based on Brain-Tumor-Progression dataset where we show how our models can be used to assess brain tumor treatment and progression. In two examples below, we vizualize how our models can be used for fully-automated tumor monitoring in longitudinal studies and what information can be extracted.

For each patient, we present the most representative slice for the tumor (usually the one selected for RANO calculation), and compare the results of our model in two time points. Our models require MRI studies with 4 sequences (T1-pre, T1-post, T2, FLAIR) but for visualization purposes, we present only T1-post and FLAIR series. The segmentations presented in the examples were also approved by our cooperating senior radiologist.

VISUALIZATION

Brain tumor segmentation: regression

Patient responded well for the treatment, and we observe a strong enhancing tumor (ET) regression for this patient a year after the treatment. Automatic RANO measurement (marked with blue line segments in both scans) and ET volume both significantly decreased (-78% and -93%, respectively). The region of edema became much smaller, and necrosis is nearly absent.

brain tumor segmentation without using AI

Edema: 97 cm3
Enhancing (ET): 28 cm3
Necrosis: 19 cm3

Tumor regression
RANO -78%
ET volume -93%

Edema: 39 cm
Enhancing (ET): 2 cm3
Necrosis: 1 cm3

VISUALIZATION

Brain tumor segmentation: progression

Our model showed a doubled tumor burden in just 2 months between scans for this patient. Two separate enhancing tumor regions have grown and merged into one big lesion. It is reflected in RANO measurement marked in the later scan (RANO for the earlier scan was measured on a different slice). All the subregions have expanded what confirms the rapid tumor progression.

Edema: 82 cm3
Enhancing (ET): 20 cm3
Necrosis: 2 cm3

Tumor progression
RANO +100%
ET volume +85%

Edema: 136 cm3
Enhancing (ET): 36 cm3
Necrosis: 20 cm3

Contact our team

Let’s talk about the details of your AI algorithm for the liver tumors and liver segmentation. We can create and develop it together.

We are ready to answer any questions.

You can submit our contact form or send us an email at contact@graylight-imaging.com

Contact our team

Let’s talk about the details of your AI algorithm for brain tumors segmentation. We can create and develop it together.

You can submit our contact form or send us an email at contact@graylight-imaging.com

Learn more about our competences