Volker Steuber | Searcharxiv

arXiv2026

Multimodal Brain Tumour Classification Using Feature Fusion

Clinicians diagnose brain tumors by synthesizing patient symptoms, medical history, and quantitative imaging data from modalities such as MRI and CT scans into a unified clinical judgement. However, most deep learning models rely on MRI/CT images alone, failing to replicate the clinicians multimodal reasoning. We explore a two-branch multimodal network combining raw MRI scans with 91 extracted radiomic features (intensity, texture, shape, and boundary descriptors) to classify brain tumors into glioma, meningioma, pituitary, and no-tumor. A pre-trained CNN backbone encodes the image stream, whereas a dedicated MLP encodes the radiomic stream. Both streams are fused via concatenation, gated, or bidirectional cross-modal attention strategies. Across nine experimental runs on a balanced 7,200 image dataset, all multimodal configurations outperform unimodal baselines with gated fusion achieving the best accuracy of 96.13%.

eess.IV↗

arXiv2017

Modules for Automated Validation and Comparison of Models of Neurophysiological and Neurocognitive Biomarkers of Psychiatric Disorders: ASSRUnit - A Case Study

The characterisation of biomarkers and endophenotypic measures has been a central goal of research in psychiatry over the last years. While most of this research has focused on the identification of biomarkers and endophenotypes, using various experimental approaches, it has been recognised that their instantiations, through computational models, have a great potential to help us understand and interpret these experimental results. However, the enormous increase in available neurophysiological and neurocognitive as well as computational data also poses new challenges. How can a researcher stay on top of the experimental literature? How can computational modelling data be efficiently compared to experimental data? How can computational modelling most effectively inform experimentalists? Recently, a general scientific framework for the generation of executable tests that automatically compare model results to experimental observations, SciUnit, has been proposed. Here we exploit this framework for research in psychiatry to address the challenges mentioned above. We extend the SciUnit framework by adding an experimental database, which contains a comprehensive collection of relevant experimental observations, and a prediction database, which contains a collection of predictions generated by computational models. Together with appropriately designed SciUnit tests and methods to mine and visualise the databases, model data and test results, this extended framework has the potential to greatly facilitate the use of computational models in psychiatry. As an initial example we present ASSRUnit, a module for auditory steady-state response deficits in psychiatric disorders.

q-bio.NC↗