Biologists have long pursued the ability to probe living systems with high spatial resolution and molecular specificity in order to elucidate fundamental biological structures and processes. An optimal imaging modality would also provide high temporal resolution, enabling real-time visualization of dynamic cellular activities and biochemical reactions. Conventional clinical imaging techniques such as X-ray imaging offer substantial tissue penetration, yet these are inherently limited in spatial resolution. In contrast, optical imaging techniques can achieve ultrafast acquisition rates and resolution at the single-molecule or even single-atom level; however, their application is restricted to superficial tissue layers owing to substantial light scattering and intrinsic autofluorescent effects. The inherent trade-off between penetration depth and spatial resolution continues to motivate the development of advanced optical imaging modalities capable of achieving deeper tissue access without sacrificing imaging precision.

Sudden cardiac death (SCD) remains one of the leading causes of mortality worldwide, with coronary artery disease (CAD) as its predominant underlying condition. However, noninvasive and accessible screening approaches for CAD are still limited. This study aims to develop and evaluate a photoplethysmography (PPG)-based method for CAD detection using a two-dimensional Gramian angular field (GAF) transformation combined with deep learning. We enrolled 89 patients with CAD and 70 healthy controls and converted their PPG signals into two GAF representations—Gramian angular summation field (GASF) and Gramian angular difference field. The GASF representation, which preserves both magnitude and phase relationships within the PPG waveform, was found to provide superior discriminative capability. Using GASF as input, the proposed SE-ResNet model achieved an accuracy of 92.43% (95% CI: 91.51–93.36), outperforming prior work that reported 83.8% accuracy (95% CI: 82.2–85.3). These results demonstrate that the GAF transformation enhances CAD detection by encoding the temporal–phase dynamics of PPG signals, which are often overlooked in conventional one-dimensional analyses. The proposed GASF-SE-ResNet framework therefore shows strong potential as a noninvasive low-cost tool for CAD screening and SCD risk reduction.

Machine learning (ML) and generative artificial intelligence (GAI) in recent years are rapidly revolutionizing the healthcare industry offering improved precision and efficiency of healthcare delivery. The use of these advanced technologies in healthcare such as medical imaging, drug discoveries, predictive analytics, and personalized medicine can diagnose diseases at the earliest, set smarter treatment plans, and improve patient outcomes. ML and generative AI aid in diagnostic accuracy and progress prediction particularly in fields like radiology and oncology, reducing error by 25% compared to traditional methods. Additionally, generative AI-based chatbots like ChatGPT have their role in healthcare through the immediate availability of medical information and assessing urgency and severity of when to seek medical guidance. Despite these advancements, ethical challenges persist such as data privacy, bias in AI algorithms, and lack of transparency. Therefore, techniques like data anonymization and the addition of controlled noise can be applied to remove identification from datasets. This narrative review highlights the implications of integrating ML and generative AI in healthcare affecting clinical practice, patient outcomes, and the broader healthcare system. ML and GAI enhance diagnostic precision and assist healthcare virtually, thereby allowing earlier diagnosis of disease, facilitating personalized treatment strategies, improving patient outcomes, and lowering costs.

Microgravity significantly challenges cognitive functions, especially spatial tasks, making effective countermeasures essential. Cognitive training improves cognitive performance, and high-definition transcranial direct current stimulation (HD-tDCS), a non-invasive neurostimulation, also improves cognition. However, their effects under microgravity remain unclear. This study investigates whether mental rotation (MR) training alone or combined with parietal HD-tDCS enhances MR and working memory (WM) under microgravity. Using a single-blind between-subjects design, participants were assigned to a no-treatment control group, a cognitive training group that received MR training, and a CT + tDCS group that received MR training combined with HD-tDCS. Participants performed MR and WM tasks before, during, and after 15-day head-down bed rest (HDBR) with behavioral and EEG recording. Microgravity exposure induced reversible behavioral declines and long-term negative neurophysiological effects on MR. Both MR training alone and combined with HD-tDCS effectively countered cognitive damage and exhibited lasting effects. HD-tDCS further enhanced the MR training benefits, improving neural efficiency. Both interventions enhanced WM and promoted re-adaptation after microgravity. Furthermore, the combined intervention increased WM resistance to interference. These results support that MR training, especially with HD-tDCS, effectively counter microgravity-induced cognitive damage, and further support non-invasive neurostimulation as a method to enhance cognitive functions and optimize training protocols under microgravity.