Epilepsy affects approximately 50 million people worldwide, and a significant subset of these patients—estimated at about 30%—do not respond to standard antiepileptic drugs. This presents a considerable challenge for both patients and healthcare providers alike, leading to an intensified search for new treatment modalities. Through machine learning, researchers can analyze vast datasets more efficiently, enabling them to discover patterns and features that would be nearly impossible to detect manually. The application of this technology to drug-resistant epilepsy holds the promise of revolutionizing patient outcomes.

The collaborative efforts in this study focused on harnessing the strengths of explainable artificial intelligence (AI) to not only predict but also elucidate the underlying biological mechanisms at play in drug-resistant epilepsy. By leveraging advanced algorithms and vast datasets, the research team aimed to create a model that could not only pinpoint biomarkers but also provide insights into the pathways that govern immune-inflammation interactions in the context of epilepsy. This dual approach could significantly enhance the personalization of treatment plans for affected patients.

One of the critical aspects of this research is the identification of immune-inflammatory biomarkers. These biomarkers are crucial indicators of potential pathological processes that may contribute to the persistence of seizures in drug-resistant epilepsy. By utilizing explainable machine learning models, the researchers successfully delineated specific biomarkers that are associated with inflammatory processes, thus suggesting novel avenues for therapeutic intervention. What sets this study apart is its commitment to transparency and understanding; while traditional machine learning often operates as a ‘black box,’ leaving healthcare providers in the dark, this approach clarifies how each decision is made.

Moreover, the study identifies several promising therapeutic candidates tailored for drug-resistant epilepsy patients. The potential adoptions of these candidates could lead to more effective, individualized treatment options that are based on a patient’s specific biomarker profile. This signifies a monumental step towards not only optimizing existing therapies, but also possibly even developing new drugs that specifically target the identified pathways.

The use of machine learning in the study also underscores a tradeoff that is critical in medical research: interpretability versus predictive power. While many machine learning models excel at generating predictions, their complexity often obscures insights into clinical implications. Ijaz et al.’s commitment to create explainable models bridges this gap, allowing researchers and clinicians to trust the decisions made by these algorithms and paving the way for their integration into clinical practice.

The results presented in this landmark study provide compelling evidence that machine learning applications can foster a deeper understanding of chronic diseases, thus enabling medical professionals to devise better treatment plans. As machine learning continues to evolve, it is imperative for researchers to remain vigilant in developing techniques that ensure transparency, as this may be vital for clinical acceptance and patient safety.

In addition to its immediate implications for epilepsy, this research contributes to a broader conversation about the role of AI in healthcare. As we witness advancements in data science and machine learning, the healthcare community must navigate ethical concerns surrounding the use of AI and ensure that such technologies empower rather than replace human decision-making. This study exemplifies the potential of responsible AI application while maintaining a strong focus on patient welfare.

The significance of this research cannot be overstated. With the identification of immune-inflammatory biomarkers and therapeutic candidates, the groundwork has been laid for future studies that will further explore the intersection of computational techniques and biomedical applications. This represents not just a single breakthrough, but a replicable framework that could be utilized in various disease contexts as we accelerate our understanding of complex medical conditions.

As researchers look to the future, the challenge remains to translate these findings into actionable clinical recommendations and treatments. Scientific discoveries, no matter how groundbreaking, require subsequent studies to validate and refine research results. Nevertheless, the efficacy of machine learning to identify biomarkers and potential therapies for drug-resistant epilepsy marks an exciting advance in the field of neurology.

In conclusion, the work by Ijaz et al. showcases not only the potential of machine learning to revolutionize the approach to drug-resistant epilepsy but also sets a benchmark for future interdisciplinary research. By advocating for explainability within AI applications in healthcare, the authors contribute to a more informed, transparent, and ultimately effective implementation of machine learning in clinical settings.

The integration of AI in medical research harnesses the ability to unpack the complexities of diseases like drug-resistant epilepsy, illuminating new paths for therapies that could fundamentally alter the lives of millions. As healthcare evolves with technological advancements, patient-centered approaches that align machine learning capabilities with ethical research practices will be crucial in tackling the pressing challenge of drug-resistant epilepsy.

Ultimately, the synergy of machine learning and biomedical sciences holds the promise of more accurate diagnoses, innovative treatments, and improved patient outcomes. The future of epilepsy treatment may very well lie in the insights that arise from the marriage of data-driven research with a keen understanding of biological systems, bringing hope to those suffering from this debilitating condition.

Subject of Research: Drug-Resistant Epilepsy and Machine Learning

Article Title: Explainable Machine Learning Identifies Immune-Inflammatory Biomarkers and Therapeutic Candidates in Drug-Resistant Epilepsy

Article References:

Ijaz, T., Maqsood, H., Rehman, A. et al. Explainable machine learning identifies immune-inflammatory biomarkers and therapeutic candidates in drug-resistant epilepsy.
Sci Rep (2025). https://doi.org/10.1038/s41598-025-30401-x

Image Credits: AI Generated

DOI: 10.1038/s41598-025-30401-x

Keywords: Machine Learning, Drug-Resistant Epilepsy, Biomarkers, Therapeutics, Immunology, AI in Healthcare

The core ambition of the GAINS project is to introduce biologically realistic temporal dynamics into neuromorphic systems, resulting in platforms capable of more sophisticated and nuanced computations. This systematic incorporation of brain-inspired characteristics is expected to offer substantial advancements in the performance of computational tasks, making these systems not only faster but also more reliable and adaptable to varying conditions encountered in real-world applications. The implications of such advancements could be profound, potentially revolutionizing how machines learn from and interact with the world around them.

Through their collaborative initiative, PI Parsa, along with co-PIs from the University of Wisconsin–Madison and Northwestern University, aims to tackle the current inadequacies present in neuromorphic hardware platforms. Neuromorphic computing, a field inspired by the neural architecture of the human brain, is often limited by its inability to replicate complex brain dynamics effectively. GAINS endeavors to bridge this gap, facilitating a significant leap towards hardware that is not only efficient but adequately mirrors the intricate workings of the brain.

At the heart of GAINS is the utilization of Izhikevich-based models, which are fundamental in achieving biologically plausible neural dynamics. This modeling allows the system to harness the rich dynamism exhibited by neurons under varying stimuli and conditions, thus ensuring that the resulting neuromorphic architectures are responsive and adaptable. The project promises not just enhanced computational capabilities but also a new paradigm for energy-efficient computing solutions.

The funding awarded to Parsa and Ascoli amounts to $156,667 for the first year, launching this ambitious two-year project with a total financial backing of $500,000. As such, the resources allocated will support not only the development of cutting-edge technology but also the research needed to explore the various dimensions of brain-like computations through advanced spintronic elements. Such spintronics facilitate the merging of traditional electronics with quantum effects, potentially yielding unprecedented efficiency gains.

As we stand on the brink of a new era in computing, the GAINS initiative lays the foundation for hardware solutions that nurture the replication of essential brain functions. The transition to energy-efficient and biologically plausible computing systems is crucial for the sustainability of digital technology as we know it today. Researchers are optimistic that the outcomes of this project will contribute significantly to a spectrum of applications ranging from artificial intelligence to advanced manufacturing processes and edge computing technologies.

The influence of GAINS transcends academia, with potential ramifications for the industry at large. As businesses increasingly seek innovative ways to harness data, the ability of neuromorphic systems to offer superior privacy, robustness, and generalizability could provide a competitive edge. With enhanced computing power, organizations will be able to derive valuable insights from complex datasets, leading to more informed decision-making processes.

Moreover, this project underscores the collaborative spirit of modern scientific inquiry, bringing together experts from diverse disciplines. With co-PIs like Akhilesh Jaiswal and Pedram Khalili contributing their knowledge from different institutions, the project encapsulates interdisciplinary collaboration as a critical ingredient for success in advancing neuromorphic technologies. Such partnerships are becoming increasingly vital in a world where diverse challenges demand comprehensive solutions derived from varied expertise.

As research progresses, the team anticipates drawing insights from their work that may inspire future innovations beyond the scope of GAINS. The methodologies and findings could stimulate further exploration into the realms of cognitive computing, enhancing our understanding of how machines might emulate not just the workings of the brain but also the subtleties of human thought and behavior.

Life-like performance in computing could redefine the boundaries of what is computationally possible. The promise of GAINS lies not just in its technical prowess but in its capacity to address the ethical and operational challenges posed by advanced AI systems. By creating more intuitive and ‘human-like’ computing environments, the project also raises important questions about the implications of integrating such technology into daily life.

As interest in neuromorphic computation continues to rise, both researchers and industry leaders are keenly focused on the advancements heralded by GAINS. By addressing the dual challenge of performance and biological realism, this project has the potential to reshape not only academic research but also commercial products and services in the coming years. The future of computing could be brighter, driven by machines that think more like us.

In conclusion, Maryam Parsa and Giorgio Ascoli’s work on the GAINS project symbolizes a significant leap forward in the race to develop neuromorphic computing systems that mirror the brain’s complexity. Their endeavor promises to deliver not only enhanced computational performance but a path forward for technology that respects and replicates the intricacies of human cognition. The impact of their research will likely resonate across multiple sectors, influencing how we interact with technology in the future, and revealing new frontiers in our understanding of both computing and the human brain.

Subject of Research: Neuromorphic Computing
Article Title: GAINS: A Leap Toward Brain-Like Computing
News Publication Date: October 2023
Web References: N/A
References: N/A
Image Credits: N/A

Keywords

Neuromorphic Computing, Izhikevich Models, Spintronics, Brain Dynamics, Energy Efficiency, Artificial Intelligence, Cognitive Computing, Interdisciplinary Research, George Mason University, U.S. Department of Energy.

The team’s findings were published on July 23 in the prestigious journal Fluids and Barriers of the CNS. This collaborative effort involved researchers from TTUHSC’s Jerry H. Hodge School of Pharmacy in Amarillo, and the Graduate School of Biomedical Sciences. Helmed by principal investigator and senior author, Dr. Ulrich Bickel, with lead author Ehsan Nozohouri — a TTUHSC graduate research assistant — the study also included contributions from other graduate researchers within the institution. Their combined expertise and innovative methodologies culminated in a transformative understanding of the BBB in the context of Alzheimer’s pathology.

Historically, scholars have debated whether Alzheimer’s disease results in the impairment of the BBB. Traditionally perceived as a protective barrier, the BBB is made up of tightly connected endothelial cells that serve as a gate, determining what enters the brain from the bloodstream. The prevailing theory suggested that the onset of Alzheimer’s could compromise the integrity of this barrier, leading to a situation where harmful compounds could enter the brain and exacerbate the disease.

The implications of this study are profound for the field of Alzheimer’s research. Ehsan Nozohouri elucidated that understanding the integrity of the BBB becomes critical, particularly for drug delivery systems intended to treat Alzheimer’s disease. Since the BBB effectively blocks the majority of medications, revealing its actual condition in Alzheimer’s provides essential insights for future therapeutic strategies.

In their investigation, Nozohouri and his colleagues employed the Tg2576 mouse model, which is well-documented for its propensity to develop amyloid beta plaques synonymous with Alzheimer’s pathology. To probe the integrity of the BBB, the research team injected a harmless test molecule, [¹³C₁₂]sucrose, a compound known for its poor ability to cross the BBB. By utilizing advanced analytical techniques, including liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS) and laser microdissection for precise tissue sampling, they meticulously monitored the presence of sucrose in various brain regions.

The results were striking. The examination revealed no significant leakage of sucrose into the brain of either the Alzheimer’s-afflicted Tg2576 mice or their healthy counterparts, at varying ages. This indicates that the BBB is not compromised on a broad scale in this model, and highlights the need for re-evaluation of the theories surrounding BBB permeability in Alzheimer’s disease.

Furthermore, the study suggested that within critical regions of the brain associated with memory and cognitive function, there were no significant differences present between the Alzheimer’s models and healthy mice. This included the examination of tight junction proteins, which function as the “mortar” that holds BBB cells together, again showing largely preserved structures even in proximity to amyloid plaques.

As the scientists contemplate these results, they emphasize that their findings challenge the widespread assumption of extensive BBB leakiness in Alzheimer’s disease. Such revelations could catalyze a paradigm shift in how drugs are designed for effective treatment, as the understanding needs to pivot towards recognizing that the BBB may not universally be compromised in the disease.

Despite the promising nature of the findings from the Tg2576 model, Nozohouri cautioned against a sweeping application of these conclusions to human physiology. The researchers stressed the necessity for additional models that may more accurately reflect human brain responses and the intricacies of Alzheimer’s disease. As the field looks forward, there exist FDA-approved monoclonal antibody treatments showing potential in slowing cognitive decline. Further examination of these therapeutics in relevant models could illuminate pathways whereby localized changes affect drug effectiveness, especially concerning concerns of micromorphological alterations like microhemorrhages.

The team at TTUHSC envisions this research as a starting point for expanded inquiries into the dynamics of Alzheimer’s disease and the BBB. They are committed to further studying how Alzheimer’s may impact the brain’s protective mechanisms and ultimately aspire to hone drug development strategies that can effectively navigate the complexities of the BBB and deliver meaningful therapeutic ramifications to patients suffering from this debilitating condition.

This investigation underscores an urgent need for more refined understanding and innovative strategies in combating Alzheimer’s disease. With the stakes high, researchers are poised to explore the intersections of neuroscience, pharmacology, and therapeutic innovation, all with the overarching goal of improving the lives of those impacted by this formidable disease.

The work done by this team moves us a step closer to unraveling the complexities of Alzheimer’s disease and the mechanisms underpinning its effects on the brain. By shedding light on the actual state of the BBB, they pave the way for future advancements in treatment strategies and foster hope for more effective interventions in the ever-challenging landscape of Alzheimer’s research.

Subject of Research: Animals
Article Title: Assessing blood-brain barrier (BBB) integrity in an Alzheimer’s disease mouse model: is the BBB globally or locally disrupted?
News Publication Date: 23-Jul-2025
Web References: DOI
References: None available
Image Credits: Credit: TTUHSC

Keywords

Biomedical engineering, Clinical medicine, Diseases and disorders, Epidemiology, Health care, Human health, Medical specialties, Pharmaceuticals, Pharmacology.

Difficulties in early detection of Parkinson’s disease have long hampered efforts for timely intervention. Traditional diagnostic methods often lack the sensitivity required to identify subtle changes in motor function during the initial stages of the disease. In response to these challenges, KAIST researchers have harnessed the power of AI alongside optogenetic techniques to create a more refined diagnostic framework. This innovative combination not only facilitates early detection but also provides an avenue for more effective therapeutic evaluation.

The research team, which included experts from various divisions within KAIST, conducted extensive studies using a mouse model of Parkinson’s disease. The model incorporated male mice that exhibited abnormalities in alpha-synuclein protein, a hallmark of the disease often used to simulate its progression in humans. Within this context, the consortium implemented AI-driven 3D pose estimation to analyze over 340 distinct behavioral features related to the mice’s motor functions.

By distilling these complex data into a singular Parkinson’s disease score (APS), the researchers established a quantifiable metric that indicated the severity of the disease. Remarkably, this score was able to demonstrate significant differentiation from control subjects as early as two weeks post disease induction. The APS proved to be a more sensitive measure than traditional motor function tests, identifying key diagnostic features such as altered stride length, asymmetrical limb motion, and tremors.

In an effort to establish the specificity of the APS to Parkinson’s disease, the researchers extended their analysis to a mouse model of Amyotrophic Lateral Sclerosis (ALS). Given that both diseases can result in motor dysfunction, it was critical that the APS score did not reflect general motor decline but rather highlighted unique indicators pertaining to Parkinson’s. The findings confirmed that the APS score remained low in the ALS model, reinforcing that the observed behavioral alterations were characteristic of Parkinson’s alone.

Beyond diagnosis, the research team’s contributions extended into therapeutic interventions. Utilizing optogenetics technology known as optoRET, they employed light to modulate neurotrophic signals in the brain of the affected mice. This groundbreaking approach allowed for precise management of movement disorders associated with Parkinson’s. Specifically, when the light was applied in a regimen of alternating days, notable improvements in gait, limb movement, and tremor severity were recorded. Moreover, there was evidence suggesting that this method may offer neuroprotection to dopamine-producing neurons, a critical factor in the pathology of Parkinson’s.

In sharing insights from this transformative research, Professor Won Do Heo emphasized that the study represents an unprecedented achievement in preclinical research frameworks. The integration of AI-based behavioral analysis with optogenetics characterizes a significant leap toward the establishment of personalized medicine strategies for Parkinson’s patients, which could potentially revolutionize treatment paradigms in the realm of neurodegenerative disorders.

The remarkable synergy between AI and bioengineering showcased in this research underscores not just the scientific rigor but also the collaborative ethos driving the work at KAIST. Relying on interdisciplinary input from teams specializing in biological sciences, cognitive neuroscience, and basic science, the project epitomizes the power of teamwork in advancing medical science.

As the project moves forward, researchers are exploring avenues for expanding the applicability of their findings to human subjects. Dr. Bobae Hyeon, the lead author of the study, is currently undertaking additional research to further the potential of cell therapy for Parkinson’s at Harvard Medical School’s McLean Hospital. Supported by initiatives like the Global Physician-Scientist Training Program, this ongoing research aims to bridge the gap between preclinical findings and clinical applications.

The implications of these findings are far-reaching. Parkinson’s disease affects millions of individuals worldwide, and the contributions from KAIST pave the way for future innovations in diagnostic and therapeutic approaches. Stakeholders in the health industry will undoubtedly keep a keen eye on how these developments evolve and the potential they hold for improving patient outcomes in the battle against neurodegenerative diseases.

As the research landscape continues to evolve with technological advancements, the fusion of artificial intelligence with biological intervention stands to redefine the boundaries of what is possible in disease management. Future studies are anticipated to refine these methodologies, pushing towards enhanced precision in both diagnosis and therapeutic effectiveness.

In summary, the efforts made by KAIST researchers not only enrich the scientific community’s understanding of Parkinson’s disease but also ignite hope for those affected by this challenging condition. The proven capability to utilize AI for enhanced detection and optogenetics for therapeutic intervention signals a new frontier in medical research and provides a template for future studies aimed at elucidating complex neurological disorders.

Subject of Research: Not applicable
Article Title: Integrating artificial intelligence and optogenetics for Parkinson’s disease diagnosis and therapeutics in male mice
News Publication Date: September 22, 2023
Web References: http://dx.doi.org/10.1038/s41467-025-63025-w
References: Not available
Image Credits: KAIST

The blood-brain barrier represents one of the foremost challenges in neurotherapeutics and brain disease management. This complex, selectively permeable membrane shields the brain’s delicate microenvironment from harmful agents such as toxins and pathogens, but at the cost of limiting access to potentially life-saving medications. Consequently, delivering chemotherapy agents or novel therapeutics to brain tissue in adequate concentrations has long been stymied by the BBB’s formidable protective function. Focused ultrasound, in conjunction with microbubble contrast agents, offers a non-invasive avenue to transiently and locally open this barrier, facilitating the controlled passage of drugs without compromising overall cerebral protection.

To rigorously characterize how focused ultrasound can enable this process with precision and reproducibility, Dr. Woodworth and his team conducted an extensive study involving 34 glioblastoma patients. These participants underwent up to six monthly cycles of treatment, culminating in an impressive dataset of 972 individual sonications—targeted ultrasound pulses aimed at specific brain regions. This large-scale effort allowed the team to meticulously analyze how different ultrasound parameters correlate with successful BBB disruption. Vital to this endeavor was the use of acoustic emissions monitoring: the capture of sound waves emitted by microbubbles oscillating in response to the ultrasound field, which serves as a real-time biomarker correlating with the degree of BBB opening.

Acoustic emissions, generated as microbubbles respond to the ultrasound energy, provide a novel, quantitative feedback mechanism enabling clinicians to fine-tune the treatment dose and target. As Dr. Woodworth explains, these signals allow for reliable prediction of BBB opening events, fostering safer and more effective therapeutic delivery. By correlating the acoustic signatures with MRI imaging and clinical outcomes, the team developed dosing guidelines that transcend individual device differences and patient variability, establishing a unifying framework for blood-brain barrier modulation through focused ultrasound across diverse clinical environments.

The study’s technical rigor is complemented by its translational importance. Previously, the lack of standardized protocols and monitoring impeded broader clinical adoption of ultrasound-mediated BBB opening. This research, therefore, marks an essential milestone, elucidating the spatial control and dosing strategies necessary for consistent, reproducible BBB disruption. Consequently, this advancement promises to accelerate the integration of ultrasound-facilitated drug delivery into routine neuro-oncological care, ultimately enhancing therapeutic efficacy for glioblastoma and potentially other neurological conditions.

The procedural basis for this treatment uses microbubbles—microscopic, inert gas-filled spheres introduced intravenously—which underlie the focused ultrasound technique. When exposed to low-intensity ultrasound waves, these microbubbles oscillate rhythmically within the cerebral vasculature. Their mechanical activity induces transient, microscopic disruptions in the tight junctions of the endothelial cells that compose the BBB, creating temporary pores through which therapeutic agents can pass. Crucially, this process is reversible and highly localized, minimizing off-target effects and preserving overall brain function while improving drug penetration.

Dr. Pavlos Anastasiadis, Assistant Professor of Neurosurgery at UMSOM and a co-author on the study, highlights the mechanistic underpinnings of this phenomenon. The oscillation of microbubbles within the ultrasound energy field leads to subtle mechanical perturbations of the blood vessel walls in the brain, enabling a safe and reversible opening of the BBB. These events can be monitored in real time using advanced imaging and acoustic emission technologies, allowing clinicians to control the extent and location of barrier disruption with unprecedented precision.

The lineage of this work extends back to seminal experiments in the early 1990s at Brigham and Women’s Hospital’s Focused Ultrasound Lab, where microbubbles were first explored as agents for BBB modulation. Building on these foundational discoveries, senior author Dr. Alexandra J. Golby, Director of Image-Guided Neurosurgery at Brigham and Women’s Hospital, emphasizes that the present study validates a clinically feasible approach to repeatedly open the BBB in glioblastoma patients ahead of chemotherapy cycles. This iterative opening holds promise to vastly improve therapeutic accumulation within tumors, potentially enhancing survival and quality of life.

Data from this investigation were derived from a subset of patients enrolled in ongoing clinical trials spearheaded by Dr. Woodworth. These trials are critically assessing the clinical impact of ultrasound-facilitated BBB opening for enhancing the delivery of standard-of-care chemotherapy in glioblastoma. The research team plans to publish detailed clinical outcomes from these broader trials imminently, promising further valuable insights into safety, efficacy, and patient benefit.

Dr. Taofeek K. Owonikoko, Executive Director of the University of Maryland Marlene and Stewart Greenebaum Comprehensive Cancer Center, noted the far-reaching implications of these findings for the field of neuro-oncology and beyond. The study’s data offer the first detailed technical description of acoustic emissions dosing, a cornerstone for clinical and regulatory progress in the adoption of focused ultrasound as a precision treatment modality. This work solidifies the foundation on which larger pivotal trials and multi-center studies can build.

One such major trial underway is LIBERATE (NCT05383872), a diagnostics-focused study in glioblastoma patients. Co-led by Dr. Woodworth, this trial leverages MRI-guided focused ultrasound to assess not only therapeutic delivery but also diagnostic enhancement capabilities, representing a frontier in personalized medicine for brain cancer. The consortium ReFOCUSED—encompassing over 20 research sites across North America—collaborates on these efforts, aiming to harness focused ultrasound technology to transform clinical outcomes in brain disease through improved drug delivery and imaging.

This research was generously supported by Insightec Inc., the manufacturer of the focused ultrasound devices utilized, along with funding from the Focused Ultrasound Foundation. Their combined support underscores the growing momentum behind ultrasound-enabled therapies, fostering innovation at the intersection of technology, engineering, and clinical neuroscience.

In summary, this detailed elucidation of acoustic emissions-guided dosing and spatial control of BBB opening ushers in a new era in neurotherapeutics. Through meticulous technical exploration, this research offers a blueprint for safely breaching the brain’s protective barrier on demand, thereby expanding the armamentarium against formidable brain cancers such as glioblastoma. As standardized protocols permeate clinical practice, focused ultrasound’s promise as a non-invasive, targeted, and controllable delivery mechanism nears clinical reality, unlocking potential not just in oncology but across the landscape of neurological diseases.

Subject of Research: Focused ultrasound-mediated blood-brain barrier opening for enhanced drug delivery in glioblastoma
Article Title: Acoustic emissions dose and spatial control of blood-brain barrier opening with focused ultrasound
News Publication Date: 25-Aug-2025
Web References:

• University of Maryland School of Medicine
• University of Maryland Marlene and Stewart Greenebaum Comprehensive Cancer Center
• Clinical Trial LIBERATE (NCT05383872)
• Journal Device DOI
  Image Credits: University of Maryland School of Medicine
  Keywords: Blood brain barrier, Glioblastomas, Cancer