The Public Says'No Thanks' The maker of the eCoffee Energyband says it electrically stimulates your nerves to keep you alert. Researchers are skeptical, and critics see it as a way for China's bosses to keep workers productive. Forget coffee, you can now stay alert by strapping on a wristband that lightly zaps you awake. That's what eCoffee Energyband, a Chinese gadget that sells for just over $100, is claiming to do. First released in late 2023, the product is a lightweight wearable with two electrode pads that sit against the inner wrist.

In a small clinical study, people with above-the-knee amputations said it helped them navigate more easily and felt more like part of their body. A subject with the osseointegrated mechanoneural prosthesis overcomes an obstacle placed in their walking path by volitionally flexing and extending their phantom knee joint. Control signals from their residual knee muscles are used to produce movement of the powered prosthetic knee that mirrors the phantom knee. MIT researchers have developed a new bionic knee that is integrated directly with the user's muscle and bone tissue. It can help people with above-the-knee amputations walk faster, climb stairs, and avoid obstacles more easily than they could with a traditional prosthesis, which is attached to the residual limb by means of a socket and can be uncomfortable. For several years, Hugh Herr, SM '93, co-director of the K. Lisa Yang Center for Bionics, has been working with his colleagues on techniques that can extract neural information from muscles left behind after an amputation and use that information to help guide a prosthetic limb.

Evidence-based medicine promises to improve the quality of healthcare by empowering medical decisions and practices with the best available evidence. The rapid growth of medical evidence, which can be obtained from various sources, poses a challenge in collecting, appraising, and synthesizing the evidential information. Recent advancements in generative AI, exemplified by large language models, hold promise in facilitating the arduous task. However, developing accountable, fair, and inclusive models remains a complicated undertaking. In this perspective, we discuss the trustworthiness of generative AI in the context of automated summarization of medical evidence.

The NLP community typically relies on performance of a model on a held-out test set to assess generalization. Performance drops observed in datasets outside of official test sets are generally attributed to "out-of-distribution'' effects. Here, we explore the foundations of generalizability and study the various factors that affect it, articulating generalizability lessons from clinical studies. In clinical research generalizability depends on (a) internal validity of experiments to ensure controlled measurement of cause and effect, and (b) external validity or transportability of the results to the wider population. We present the need to ensure internal validity when building machine learning models in natural language processing, especially where results may be impacted by spurious correlations in the data. We demonstrate how spurious factors, such as the distance between entities in relation extraction tasks, can affect model internal validity and in turn adversely impact generalization. We also offer guidance on how to analyze generalization failures.

Machine learning (ML) has shown great promise for revolutionizing a number of areas, including healthcare. However, it is also facing a reproducibility crisis, especially in medicine. ML models that are carefully constructed from and evaluated on a training set might not generalize well on data from different patient populations or acquisition instrument settings and protocols. We tackle this problem in the context of neuroimaging of Alzheimer's disease (AD), schizophrenia (SZ) and brain aging. We develop a weighted empirical risk minimization approach that optimally combines data from a source group, e.g., subjects are stratified by attributes such as sex, age group, race and clinical cohort to make predictions on a target group, e.g., other sex, age group, etc. using a small fraction (10%) of data from the target group. We apply this method to multi-source data of 15,363 individuals from 20 neuroimaging studies to build ML models for diagnosis of AD and SZ, and estimation of brain age. We found that this approach achieves substantially better accuracy than existing domain adaptation techniques: it obtains area under curve greater than 0.95 for AD classification, area under curve greater than 0.7 for SZ classification and mean absolute error less than 5 years for brain age prediction on all target groups, achieving robustness to variations of scanners, protocols, and demographic or clinical characteristics. In some cases, it is even better than training on all data from the target group, because it leverages the diversity and size of a larger training set. We also demonstrate the utility of our models for prognostic tasks such as predicting disease progression in individuals with mild cognitive impairment. Critically, our brain age prediction models lead to new clinical insights regarding correlations with neurophysiological tests.

Brain-computer interface company Neuralink announced on 25 May that it has received approval from the US Food and Drug Administration (FDA) for a clinical study in humans. Neuralink made the announcement on Twitter: "We are excited to share that we have received the FDA's approval to launch our first-in-human clinical study." The tweet said that the approval "represents an important first step that will one day allow our technology to help many people". The firm also said that the recruitment is not yet open for the trial, and it has yet to give any further details about what the trial will entail. Neuralink was formed in 2016 by Elon Musk and a group of scientists and engineers with the ultimate aim of making devices that interface with the human brain – both reading information from neurons as well as feeding information directly back into the brain.

United States regulators have given approval for Elon Musk's start-up Neuralink to test its brain implants on people. Neuralink said on Thursday that it received clearance from the US Food and Drug Administration (FDA) for the first human clinical study of implants which are intended to let the brain interface directly with computers. "We are excited to share that we have received the FDA's approval to launch our first-in-human clinical study," Neuralink said in a post on Twitter – which is owned by Musk. Neuralink prototypes, which are the size of a coin, have so far been implanted in the skulls of monkeys, demonstrations by the startup showed. With the help of a surgical robot, a piece of the skull is replaced with a Neuralink disk, and its wispy wires are strategically inserted into the brain, an early demonstration showed.

Despite the great promise that machine learning has offered in many fields of medicine, it has also raised concerns about potential biases and poor generalization across genders, age distributions, races and ethnicities, hospitals, and data acquisition equipment and protocols. In the current study, and in the context of three brain diseases, we provide evidence which suggests that when properly trained, machine learning models can generalize well across diverse conditions and do not necessarily suffer from bias. Specifically, by using multi-study magnetic resonance imaging consortia for diagnosing Alzheimer's disease, schizophrenia, and autism spectrum disorder, we find that well-trained models have a high area-under-the-curve (AUC) on subjects across different subgroups pertaining to attributes such as gender, age, racial groups, and different clinical studies and are unbiased under multiple fairness metrics such as demographic parity difference, equalized odds difference, equal opportunity difference etc. We find that models that incorporate multi-source data from demographic, clinical, genetic factors and cognitive scores are also unbiased. These models have better predictive AUC across subgroups than those trained only with imaging features but there are also situations when these additional features do not help.

A colorful combination of companies--aerospace, retail, gaming, security, and telecommunications--have in recent years discovered the power of pairing artificial intelligence (AI) and virtual reality (VR) to help alleviate the complexities within their industries. A few years ago, both these innovations seemed almost foreign, even to the most advanced industries. Today, it's almost a critical mission for any business to maximize the capabilities of both AI and VR if they are looking to survive in the hyper-active and competitive marketplace. The build-up of the Internet-of-Things (IoT), mobile communication, and 5G tools have also now largely contributed to the rapid expansion of these technologies across multiple fields. While for a greater part many experts claimed that these developments would mainly form part of advanced businesses, especially in fields such as military, security, engineering, architecture and aviation, the need for state-of-the-art tools have quickly manifested itself within the healthcare and alternative medicine market in recent years.

Andrew Watson is Vice President of AI and R & D at Healx. Prior to joining Healx he worked at the technology giant Dyson, where he was the founding member of the Machine Learning Research Department, leading the research and implementation of machine learning and artificial intelligence across a variety of global product categories. In his time as Director of Machine Learning at Dyson, Andrew also established a new research group, focused on the intersection between machine learning and cutting-edge biomedical research. Healx is an AI-powered, patient-inspired technology company, dedicated to helping rare disease patients around the world access life-improving therapies. There are 7,000 known rare diseases that affect 400 million people across the globe but only 5% of those conditions have approved treatments.