AI-based MRI auto-segmentation of brain tumor in rodents, a multicenter study

LYNSU: automated 3D neuropil segmentation of fluorescent images for Drosophila brains

The brain atlas, which provides information about the distribution of genes, proteins, neurons, or anatomical regions, plays a crucial role in contemporary neuroscience research. To analyze the spatial distribution of those substances based on images from different brain samples, we often need to warp and register individual brain images to a standard brain template. However, the process of warping and registration may lead to spatial errors, thereby severely reducing the accuracy of the analysis. To address this issue, we develop an automated method for segmenting neuropils in the Drosophila brain for fluorescence images from the FlyCircuit database. This technique allows future brain atlas studies to be conducted accurately at the individual level without warping and aligning to a standard brain template. Our method, LYNSU (Locating by YOLO and Segmenting by U-Net), consists of two stages. In the first stage, we use the YOLOv7 model to quickly locate neuropils and rapidly extract small-scale 3D images as input for the second stage model. This stage achieves a 99.4% accuracy rate in neuropil localization. In the second stage, we employ the 3D U-Net model to segment neuropils. LYNSU can achieve high accuracy in segmentation using a small training set consisting of images from merely 16 brains. We demonstrate LYNSU on six distinct neuropils or structures, achieving a high segmentation accuracy comparable to professional manual annotations with a 3D Intersection-over-Union (IoU) reaching up to 0.869. Our method takes only about 7 s to segment a neuropil while achieving a similar level of performance as the human annotators. To demonstrate a use case of LYNSU, we applied it to all female Drosophila brains from the FlyCircuit database to investigate the asymmetry of the mushroom bodies (MBs), the learning center of fruit flies. We used LYNSU to segment bilateral MBs and compare the volumes between left and right for each individual. Notably, of 8,703 valid brain samples, 10.14% showed bilateral volume differences that exceeded 10%. The study demonstrated the potential of the proposed method in high-throughput anatomical analysis and connectomics construction of the Drosophila brain.

Predicting histological grade in symptomatic meningioma by an objective estimation of the tumoral surface irregularity

INTRODUCTION

Predicting the histopathologic grade of meningioma is relevant because local recurrence is significantly greater in WHO grade II-III compared to WHO grade I tumours, which would ideally benefit from a more aggressive surgical strategy. It has been suggested that higher WHO grade tumours are more irregularly-shaped. However, irregularity is a subjective and observer-dependent feature. In this study, the tumour surface irregularity of a large series of meningiomas, measured upon preoperative MRI, is quantified and correlated with the WHO grade.

METHODS

Unicentric retrospective observational study of a cohort of symptomatic meningiomas surgically removed in the time period between January 2015 and December 2022. Using specific segmentation software, the Surface Factor (SF) was calculated for each meningioma. SF is an objective parameter that compares the surface of a sphere (minimum surface area for a given volume) with the same volume of the tumour against the actual surface of the tumour. This ratio varies from 0 to 1, being 1 the maximum sphericity. Since irregularly-shaped meningiomas present proportionally greater surface area, the SF tends to decrease as irregularity increases. SF was correlated with WHO grade and its predictive power was estimated with ROC curve analysis.

RESULTS

A total of 176 patients (64.7% females) were included in the study; 120 WHO grade I (71.9%), 43 WHO grade II (25.7%) and 4 WHO grade III (2.4%). A statistically significant difference was found between the mean SF of WHO grade I and WHO grade II-III tumours (0.8651 ± 0.049 versus 0.7081 ± 0.105, p < 0.0001). Globally, the SF correctly classified more than 90% of cases (area under ROC curve 0.940) with 93.3% sensibility and 80.9% specificity. A cutoff value of 0.79 yielded the maximum precision, with positive and negative predictive powers of 82.6% and 92.6%, respectively. Multivariate analysis yielded SF as an independent prognostic factor of WHO grade.

CONCLUSION

The Surface Factor is an objective and quantitative parameter that helps to identify aggressive meningiomas preoperatively. A cutoff value of 0.79 allowed differentiation between WHO grade I and WHO grade II-III with high precision.