The New Scientist Book Club has various issues with Masud Husain's prize-winning popular science book about neurology The New Scientist Book Club stepped away from science fiction for our October read, turning to the winner of the Royal Society Trivedi Science Book Prize instead, serendipitously announced just in time for us to start on our next literary adventure. Six books had been up for the award, from Daniel Levitin's to Sadiah Qureshi's . Judges picked Masud Husain's and they praised it effusively, calling it "a beautiful exploration of how problems in the brain can cause people to lose their sense of self", and citing how these medical histories are "skilfully interwoven with Husain's personal story of moving to the UK as an immigrant in the 1960s, where he found himself grappling with his own sense of belonging". Sandra Knapp, chair of the judging panel for the 2025 Royal Society Trivedi Science Book Prize, explains why neurologist Masud Husain's collection of case studies is such an enlightening, compassionate book The first thing to say is: our book club members are much tougher judges than those on the panel for the Royal Society prize! While I think we were excited to get to grips with this book, and to venture into the world of non-fiction for a change, there were many issues that were raised and picked over by our readers. Let's tackle the positives first.

It's an increasingly common experience: You wander into the kitchen, quietly muttering under your breath, when you hear a disembodied feminine voice say, "I'm sorry, I didn't quite catch that." We can all agree that Alexa's tendency to eavesdrop is, at times, a little creepy. But is it possible to harness that ability to improve our health? That's the question that researcher David Simon and his coauthors sought to answer in a recent paper published in Cell Press. Simon, a legal ethicist at Harvard University, and his team imagined a hypothetical near-future scenario in which Alexa came equipped with the power to diagnose cognitive conditions like Alzheimer's and dementia simply by analyzing an elder person's speech patterns.

The announcement by American actor Bruce Willis that he will be "stepping away" from the big screen has drawn attention to aphasia, a little-known condition with many causes. Willis's family said on Wednesday he had recently been diagnosed with the condition, which they said is "impacting his cognitive abilities". "As a result of this and with much consideration, Bruce is stepping away from the career that has meant so much to him," the family said in a statement. The cause of the 67-year-old's condition was not revealed. Aphasia can affect the ability to produce and comprehend both written and spoken communication.

MIT researchers have developed a system that gleans far more labeled training data from unlabeled data, which could help machine-learning models better detect structural patterns in brain scans associated with neurological diseases. The system learns structural and appearance variations in unlabeled scans, and uses that information to shape and mold one labeled scan into thousands of new, distinct labeled scans. System helps machine-learning models glean training information for diagnosing and treating brain conditions. MIT researchers have devised a novel method to glean more information from images used to train machine-learning models, including those that can analyse medical scans to help diagnose and treat brain conditions. An active new area in medicine involves training deep-learning models to detect structural patterns in brain scans associated with neurological diseases and disorders, such as Alzheimer's disease and multiple sclerosis.

An active new area in medicine involves training deep-learning models to detect structural patterns in brain scans associated with neurological diseases and disorders, such as Alzheimer's disease and multiple sclerosis. But collecting the training data is laborious: All anatomical structures in each scan must be separately outlined or hand-labeled by neurological experts. And, in some cases, such as for rare brain conditions in children, only a few scans may be available in the first place. In a paper presented at the recent Conference on Computer Vision and Pattern Recognition, the MIT researchers describe a system that uses a single labeled scan, along with unlabeled scans, to automatically synthesize a massive dataset of distinct training examples. The dataset can be used to better train machine-learning models to find anatomical structures in new scans -- the more training data, the better those predictions.

Brain imaging data are important in brain sciences yet expensive to obtain, with big volume (i.e., large p) but small sample size (i.e., small n). To tackle this problem, transfer learning is a promising direction that leverages source data to improve performance on related, target data. Most transfer learning methods focus on minimizing data distribution mismatch. However, a big challenge in brain imaging is the large domain discrepancies in cognitive experiment designs and subject-specific structures and functions. A recent transfer learning approach minimizes domain dependence to learn common features across domains, via the Hilbert-Schmidt Independence Criterion (HSIC). Inspired by this method, we propose a new Domain Independent Support Vector Machine (DI-SVM) for transfer learning in brain condition decoding. Specifically, DI-SVM simultaneously minimizes the SVM empirical risk and the dependence on domain information via a simplified HSIC. We use public data to construct 13 transfer learning tasks in brain decoding, including three interesting multi-source transfer tasks. Experiments show that DI-SVM's superior performance over eight competing methods on these tasks, particularly an improvement of more than 24% on multi-source transfer tasks.