No abstract

Deep learning (DL) is a widely applied mathematical modeling technique. Classically, DL models utilize large volumes of training data, which are not available in many healthcare contexts. For patients with brain tumors, non-invasive diagnosis would represent a substantial clinical advance, potentially sparing patients from the risks associated with surgical intervention on the brain. Such an approach will depend upon highly accurate models built using the limited datasets that are available. Herein, we present a novel genetic algorithm (GA) that identifies optimal architecture parameters using feature embeddings from state-of-the-art image classification networks to identify the pediatric brain tumor, adamantinomatous craniopharyngioma (ACP). We optimized classification models for preoperative Computed Tomography (CT), Magnetic Resonance Imaging (MRI), and combined CT and MRI datasets with demonstrated test accuracies of 85.3%, 83.3%, and 87.8%, respectively. Notably, our GA improved baseline model performance by up to 38%. This work advances DL and its applications within healthcare by identifying optimized networks in small-scale data contexts. The proposed system is easily implementable and scalable for non-invasive computer-aided diagnosis, even for uncommon diseases.

Malignant brain tumors including primary brain tumors (e.g., glioblastoma multiforme) and metastases, are aggressive and lethal entities for the majority of affected patients. Current standard treatments involving combinations of surgery, radiotherapy and systemic chemotherapy offer only modest improvements in survival. Faced with dismal survival, great efforts are deployed to find interesting treatment alternatives. However, the blood-brain barrier (BBB) and the blood-tumor barrier (BTB) remain great obstacles to significant drug delivery to brain tumors. The need to optimize delivery strategies for better patient outcome in the treatment of malignant brain tumors is well acknowledged. Certain interesting strategies use surgical or physical techniques to enhance the distribution of therapeutic agents to the central nervous system. The following strategies will be discussed in this review: intra-arterial delivery, osmotic BBB disruption, intranasal delivery, convection-enhanced delivery and osmotic pumps, implanted polymers, magnetic microspheres and ultrasound BBB disruption. The purpose of this paper is to review the importance of the BBB and the BTB and to review the current status and future perspectives of these delivery procedures.

No abstract