When Rice University doctoral candidate Barclay Jumet first launched a high school business designing, making and selling bow ties — learning to sew on his mother’s college sewing machine — he never imagined that same skill set would one day help him reinvent how people communicate. Today alongside adviser Daniel J. Preston, Jumet is co-founder of Actile Technologies, a startup transforming everyday fabrics into smart, touch-based communication devices.

Wearable ultrasound devices represent a transformative advancement in therapeutic applications, offering noninvasive, continuous, and targeted treatment for deep tissues. These systems leverage flexible materials (e.g., piezoelectric composites, biodegradable polymers) and conformable designs to enable stable integration with dynamic anatomical surfaces. Key innovations include ultrasound-enhanced drug delivery through cavitation-mediated transdermal penetration, accelerated tissue regeneration via mechanical and electrical stimulation, and precise neuromodulation using focused acoustic waves. Recent developments demonstrate wireless operation, real-time monitoring, and closed-loop therapy, facilitated by energy-efficient transducers and AI-driven adaptive control. Despite progress, challenges persist in material durability, clinical validation, and scalable manufacturing. Future directions highlight the integration of nanomaterials, 3D-printed architectures, and multimodal sensing for personalized medicine. This technology holds significant potential to redefine chronic disease management, postoperative recovery, and neurorehabilitation, bridging the gap between clinical and home-based care.

Recurrent spontaneous abortion (RSA) is a complex, multifactorial condition that presents significant diagnostic challenges. Current clinical guidelines are often inadequate for idiopathic cases or emerging biomarkers, and artificial intelligence (AI) models struggle to integrate multimodal data.

To address these issues, this research developed RSA-KG, a graph-based, RAG-enhanced AI knowledge graph. The system synthesizes multimodal clinical data by integrating 5 international RSA guidelines, utilizing natural language processing (NLP) and multimodal models for data processing.

Evaluations demonstrated that Large Language Models (LLMs) enhanced by RSA-KG significantly outperformed both naive retrieval-augmented generation (RAG) and raw models in diagnostic accuracy. Furthermore, reproductive specialists rated the outputs from the RSA-KG system more favorably than those from raw models or other medical LLMs. RSA-KG represents a novel approach to RSA management, overcoming the limitations of traditional AI by modeling systemic interactions and integrating real-time evidence.

Nowadays, compute-intensive programs, like those for training artificial intelligence and machine learning models, are used extensively. Modern compilers use vectorization techniques to exploit parallel processing capabilities to improve the performance of such programs. A group of scientists from the University of Southern California, Cisco AI Research, and Intel Labs designed a data-driven, graph-based learning framework for automatic vectorization called autograph, which utilizes deep reinforcement learning to have an intelligent agent learn an optimal policy. Autograph greatly outperformed other approaches across different datasets. The work was published on June 2, 2025, in an article titled “A Graph-Based Learning Framework for Compiler Loop Auto-Vectorization” in Intelligent Computing, a Science Partner Journal.

Critical concerns regarding the security and privacy of information transmitted within Internet of Medical Things systems have increased greatly, since these systems manage and generate substantial amounts of sensitive private data. Current traditional security methods have not yet adapted to evolving cyber threats, making the need for data security in medical settings crucial. Recently, a security framework based on blockchain technology and distributed reinforcement learning has been presented to address these challenges. The new framework ensures that data are stored securely and transmitted reliably while minimizing resource usage, and it also enables security measures to adapt to changing threat patterns and enhance system resilience against attacks. In summary, the new methods demonstrated improved memory consumption and transaction latency compared to existing approaches, while maintaining high data throughput. This work was published in Intelligent Computing, a Science Partner Journal, under the title “Privacy-Preserving Strategies in the Internet of Medical Things Using Reinforcement Learning and Blockchain” by Dounia Doha and Ping Guo.

University of Texas at Dallas researchers have taken an important step forward in building a neuromorphic computer by creating a small-scale prototype that learns patterns and makes predictions using fewer training computations than conventional AI systems. Their next challenge is to scale up the proof-of-concept to larger sizes.