As somebody well into middle age, I can’t deny that the burgeoning Artificial Intelligence (AI) field spurs both excitement and anxiety in me. I am a professional writer, so I actively resist AI’s suggested email replies to my colleagues. I assess my work for original analysis and detail I don’t believe AI could reproduce. A computer couldn’t possibly generate a report or proposal as specific and comprehensive as mine, I reason. My job is safe for now. Probably….
That said, when I read the recent story about how the new AI-powered Meld Graph can analyze MRI images, I was intrigued. The tool was developed at University College London (UCL) and King’s College, London. A study showed that the Meld Graph led to quicker diagnoses, increased access to surgical treatment for epilepsy, and lower healthcare costs. My typical AI skepticism kicked in. Can a neurosurgeon really trust a computer for results, I wondered.
The study tested 703 patients with epilepsy due to Focal Cortical Dysplasia (FCD). Those patients whose test results were interpreted by the Meld Graph detected 64% of lesions previously missed by radiologists with a high level of accuracy. The test results indicate lesions’ size and location, providing neurosurgeons with critical data to guide their patients’ treatment plans. While my epilepsy wasn’t due to FCD, my pre-surgery tests showed mesial temporal sclerosis, which presented as scarring on the right side of my brain. The post-surgery analysis of my excised temporal lobe tissue showed abnormal cortical neuronal clustering, a broader term describing any situation where neurons are unusually grouped together. I suspect the AI-powered Meld Graph would detect all these aberrations and more.
So, again, it’s the 21st century, time to put my AI apprehensions aside and cheer on the developing new technologies now available to people with epilepsy. Today’s epilepsy treatment toolbox is far more robust compared to 2014, when I had the surgeries I describe in my book, Seizing Control. I genuinely enjoy learning about the new neuro-scientific developments. I find the innovations reassuring: Thanks to them, brain surgery options are becoming less invasive and more accurate.
Another striking aspect of this study is the global collaborative effort among the researchers. According to the study’s co-lead, Dr. Sophie Adler of UCL, “This type of research is only possible with international collaboration. We were privileged to work with 75 researchers and clinicians towards this common goal of ‘no missed epilepsy lesions worldwide.’” Of the 33 universities and hospitals that worked on the study, only four were from the U.S.
While the Meld Graph is not yet clinically available, the research team has released the AI tool as an open-source software. They are running workshops to train clinicians and researchers worldwide on how to use it, including at Ohio’s Cleveland Clinic. I am relieved that a U.S.-based Epilepsy Center has access to the Meld Graph, despite our federal administration’s isolationist tendencies. But where the federal government has put a freeze on new grant proposals to the National Institutes of Health (NIH) and instructed the NIH to disregard negotiated rates for “indirect costs” for their grants, I fear that going forward, the U.S. epilepsy-community may not have access to the new technology. Worse still, the NIH-funded neuroscience research will likely be thwarted by funded cuts, also resulting in decreased access to new innovative epilepsy treatment options.
Given the current political landscape, for now, I’m hoping that eventually enough Congresspeople will realize the importance of scientific health research and restore the NIH funds. Otherwise, America might well get left out of future collaborations like the Meld Graph.
Leave a Reply