Grade prediction of lesions in cerebral white matter using a convolutional neural network
Noriaki Takemura1, Yuya Shinkawa2, Kazuo Ishii1,3
1Department of Applied Information Engineering, Faculty of Engineering, Suwa University of Science, Chino, Nagano, Japan.
Abstract:
We established a diagnostic method for cerebral white matter lesions using MRI images and examined the relationship between the MRI images and the medical checkup data. There were approximately 25 MRI images for each patient's head, from the top of the head to near the eyes. To order these images, we defined the unit of axial for convenience. We varied conditions, such as the location and extent of the images to be loaded, into a convolutional neural network model and verified the changes in discrimination performance on the test data. Co-occurrence network diagrams were also used to determine the relationship between the grade of cerebral white matter lesions and the biochemical test items, which were treated as categorical variables, the progression of cerebral white matter lesions, and patient health status. The convolutional neural network showed the highest discrimination performance when the images were loaded into the model with 80 pixels per side, axial from 9 to 15, along with FLAIR and T1-weighted images. The area under the curve for each grade was 0.9814 for grade 0, 0.9800 for grade 1, 0.9905 for grade 2, 0.9977 for grade 3, and 0.9998 for grade 4. In the co-occurrence network diagram, patients with no or mild cerebral white matter lesions, such as grade 0 and grade 1, had near normal blood pressure, whereas grade 2 patients were closer to (isolated) systolic hypertension. This indicates that patients with higher-grade cerebral white matter lesions tend to experience more severe hypertension.
Insights
This study introduces a new MRI-based diagnostic method for cerebral white matter lesions. The findings reveal a correlation between lesion severity and hypertension, suggesting higher lesion grades indicate increased hypertension risk.
Area of Science:
- Medical Imaging
- Neurology
- Artificial Intelligence
Background:
- Cerebral white matter lesions are common and associated with various health conditions.
- Accurate diagnosis and understanding their relationship with other health markers are crucial.
Purpose of the Study:
- To develop a novel diagnostic method for cerebral white matter lesions using MRI.
- To investigate the relationship between lesion characteristics and medical checkup data, including hypertension.
Main Methods:
- Utilized a convolutional neural network (CNN) model with MRI images (FLAIR and T1-weighted) to classify lesion grades.
- Optimized CNN performance by varying image parameters (pixel dimensions, axial slice range).
- Employed co-occurrence network diagrams to analyze associations between lesion grades, biochemical markers, and blood pressure.
Main Results:
- The CNN achieved high discrimination performance, with Area Under the Curve (AUC) values ranging from 0.9814 to 0.9998 across lesion grades 0-4.
- Optimized CNN performance was achieved with 80 pixels/side and axial slices 9-15.
- Co-occurrence network analysis indicated a trend of increasing hypertension severity with higher cerebral white matter lesion grades.
Conclusions:
- The developed MRI-based CNN method is highly effective for diagnosing cerebral white matter lesions.
- A significant association exists between the severity of cerebral white matter lesions and the degree of hypertension.
