Artificial intelligence prediction of non-enhancing brain tumor malignancy based on in vivo confocal laser endomicroscopic imaging

Jiuxu Chen¹Yuan Xu²Irakliy Abramov²Carlos E. Calderon Valero²Thomas J. On²Jennifer M. Eschbacher²Baoxin Li¹Mark C, Preul^2*

• ¹Arizona State University, Tempe, United States
• ²Division of Neurological Surgery, Barrow Neurological Institute (BNI), Phoenix, Arizona, United States

Background: Although non-enhancing tumors are often thought to be lower grade, malignant regions can be missed on conventional magnetic resonance imaging. Fluorescein-based confocal laser endomicroscopy (CLE) enables real-time, cellular-resolution imaging of brain tissue during tumor resection. It is particularly valuable for evaluating non-enhancing brain tumors. However, CLE interpretation remains subjective. Although CLE has high sensitivity, it is less specific than standard histology. Existing artificial intelligence (AI) models process CLE images as independent frames, neglecting the temporal context that human experts use during interpretation. Methods: A novel sequence-based deep learning model was developed to classify tumor grade on the basis of CLE image sequences, mimicking the visual reasoning process of expert neuropathologists. CLE images were collected from 16 patients with non-enhancing brain tumors. Each sequence was labeled as high-grade or low-grade based on neuropathologist interpretation, blinded to final histopathology findings. Visual features were extracted using pretrained backbones (vision transformer, VGG16, ResNet50), followed by temporal modeling with a transformer encoder and temporal convolution. This model was compared with conventional frame-based classification across 3 random train-test splits. Results: The dataset included 105 CLE sequences (3173 images, 40 regions of interest). The sequence-based model achieved top-1 classification accuracies of 93% (vision transformer), 88% (VGG16), 74% (ResNet50), and 67% (Inception-ResNet-V2), outperforming corresponding frame-based models (78%, 74%, 55%, and 50%). Diagnostic performance was comparable to expert neuropathologist interpretation (87%). The model demonstrated robustness in artifact-affected sequences and improved interpretability by incorporating temporal progression. Conclusions: AI models that integrate both visual and temporal information from CLE digital imaging sequences can effectively classify brain tumor grade with accuracy comparable to that of expert neuropathologists, outperforming frame-based models. Such a system reduces interpretive subjectivity and holds promise as an intraoperative decision CLE support tool for non-enhancing brain tumor resection.