Medical Bulletin 16/ November/ 2024

Researchers have developed an AI powered model that -- in 10 seconds -- can determine during surgery if any part of a cancerous brain tumor that could be removed remains, a study published in Nature suggests. The technology, called FastGlioma, outperformed conventional methods for identifying what remains of a tumor by a wide margin, according to the research team led by University of Michigan and University of California San Francisco.

When a neurosurgeon removes a life threatening tumor from a patient's brain, they are rarely able to remove the entire mass. Neurosurgical teams employ different methods to locate that residual tumor during a procedure. They may get MRI imaging, which requires intraoperative machinery that is not available everywhere. The surgeon might also use a fluorescent imaging agent to identify tumor tissue, which is not applicable for all tumor types.

In this international study of the AI driven technology, neurosurgical teams analyzed fresh, unprocessed specimens sampled from 220 patients who had operations for low- or high-grade diffuse glioma. FastGlioma detected and calculated how much tumor remained with an average accuracy of approximately 92%.

In a comparison of surgeries guided by FastGlioma predictions or image- and fluorescent-guided methods, the AI technology missed high-risk, residual tumor just 3.8% of the time -- compared to a nearly 25% miss rate for conventional methods. To assess what remains of a brain tumor, FastGlioma combines microscopic optical imaging with a type of artificial intelligence called foundation models.

These are AI models, such as GPT-4 and DALL·E 3, trained on massive, diverse datasets that can be adapted to a wide range of tasks. After large scale training, foundation models can classify images, act as chatbots, reply to emails and generate images from text descriptions. To build FastGlioma, investigators pre-trained the visual foundation model using over 11,000 surgical specimens and 4 million unique microscopic fields of view.

Reference: Hollon, T., Kondepudi, A., Pekmezci, M., Hou, X., Scotford, K., Jiang, C., ... & Hervey-Jumper, S. (2024). Visual foundation models for fast, label-free detection of diffuse glioma infiltration.

Gabapentinoid use was associated with an increased risk of hip fractures, especially in patients who were frail or had chronic kidney disease, according to a study published in JAMA Network. The researchers tracked patients hospitalised for hip fractures in Victoria, Australia from March 2013 through June 2018, who had used gabapentinoids before the injury.

Study co-author and Director of the Centre for Medicine Use and Safety (CMUS) at the Monash Institute of Pharmaceutical Sciences, Professor Simon Bell, said gabapentinoids can be effective for neuropathic pain but can also result in adverse events including dizziness, gait disturbance and balance disorder. The researchers analysed the data for 28,293 patients who experienced hip fractures across a five-year period.

“Our results showed patients had 30 per cent increased odds of suffering a hip fracture within two months of being dispensed a gabapentinoid medication,” Professor Bell said.

“The link between gabapentinoids and hip fractures existed across different age groups but the odds of hip fracture was higher among patients who were frailer or had chronic kidney disease, so these should be important considerations when deciding when to prescribe gabapentinoids.”

Lead author of the study and CMUS PhD candidate Miriam Leung, said the study highlights that caution is needed before prescribing gabapentinoids, especially for people who are prone to falls and fractures.

“Our findings highlight the importance of assessing each patient’s risk before prescribing gabapentinoids,” Ms Leung said.

Reference: Leung MTY, Turner JP, Marquina C, et al. Gabapentinoids and Risk of Hip Fracture. JAMA Netw Open. 2024;7(11):e2444488. doi:10.1001/jamanetworkopen.2024.44488

Young people’s mental health may depend on how they use social media, rather than how much time they spend using it, according to a new study by University of B.C. researchers.

The research published in Journal of Experimental Psychology: General, examined the effects of quitting social media versus using it more intentionally.

The results showed that users who thoughtfully managed their online interactions, as well as those who abstained from social media entirely, saw mental health benefits particularly in reducing symptoms of anxiety, depression and loneliness.

With social media use nearly universal among young adults, especially those ages 17-29, concerns over its impact on mental health have grown.

In the six-week study, 393 Canadian young adults with some mental health symptoms and concerns about social media’s impact on their mental health were split into three groups:

• a control group that continued their usual routines

• an abstinence group asked to stop using social media entirely

• a “tutorial” group that was coached in intentional usage

The tutorials guided participants on fostering meaningful online connections, limiting interactions that encouraged self-comparison, and carefully selecting who they followed.

Both the abstinence and tutorial groups reduced their social media use and experienced fewer social comparisons—a common trigger for anxiety and low self-esteem. While the tutorial group didn't cut back on social media as much as those who tried to abstain completely, they reported notable improvements in loneliness and fear of missing out.

Researchers used an approach with the tutorial group that emphasized quality over quantity in social media interactions. By muting or unfollowing accounts that triggered envy or negative self-comparisons and prioritizing close friendships, tutorial participants built a healthier online environment. Rather than passively scrolling, they were encouraged to actively engage with friends by commenting or sending direct messages—a behaviour that tends to deepen meaningful connections while helping users feel more socially supported.

Reference: Mikami, A. Y., Khalis, A., & Karasavva, V. (2024). Logging out or leaning in? Social media strategies for enhancing well-being.Journal of Experimental Psychology: General. Advance online publication. https://doi.org/10.1037/xge0001668

Without intervention, global plastic waste could double by 2050, a new machine learning study predicts. However, according to simulations by the study’s authors, a mix of policy interventions could cut plastic waste by more than 90% and it could cut plastics-related emissions by a third. With UN treaty negotiations underway, these findings provide a crucial blueprint for tackling the plastic crisis. Plastic production has increased relentlessly for decades, leading to surging plastic waste generation and environmental mismanagement. As plastic degrades, it fragments into micro- and nano-plastics, which harm ecosystems globally – from the Arctic to deep ocean habitats – and pose significant health risks, including heightened cancer risks, cardiovascular disease, and reproductive issues. The plastic lifecycle also intensifies climate change through emissions from oil and gas extraction, production, and waste processing. The disproportionate plastic waste burden on the Global South and the frequency of situating plastic facilities near marginalized communities have sparked pressing environmental justice concerns. Recently, there has been global momentum to address these issues, culminating in a 2022 United Nations resolution to negotiate a legally binding treaty to curb plastic pollution.

To help in this effort, Samuel Pottinger and colleagues developed a novel machine learning model to forecast trends in global plastic production, trade, and waste management to 2050. They also simulated the effects of eight plausible policy interventions to mitigate waste and emissions. Pottinger et al. found that, without interventions, annual mismanaged plastic waste is anticipated to almost double by 2050, reaching 121 million metric tons. Concurrently, annual greenhouse gas emissions from the global plastic system are projected to rise by 37% over the same period. However, the authors also show that a combined policy intervention approach involving a production cap, recycling mandate, packaging tax, and infrastructure investment could reduce mismanaged plastic waste by up to 91% and decrease 2050 plastics-related emissions by about a third. “Collectively, these observations provide timely insight into how to maximize the impact of the UN plastic pollution treaty both as it is being drafted and over the longer time horizon of its implementation,” write Pottinger et al. “It is clear from these results that, with sufficient political will, there is enough technical potential to dramatically reduce mismanaged plastic waste and meaningfully address some of the more insidious associated issues.”

Reference: A. Samuel Pottinger et al., Pathways to reduce global plastic waste mismanagement and greenhouse gas emissions by 2050.Science0,eadr3837DOI:10.1126/science.adr3837