Xiaoyue Ni

Abstract Urinary stone disease is a common urological condition with increasing incidence, particularly in developed countries. Laser lithotripsy (LL) has become a preferred minimally invasive treatment due to its high precision and low tissue damage. Recent studies suggest that cavitation plays a critical role in stone damage during LL, and three-dimensional passive cavitation mapping (3D-PCM) has emerged as a promising tool for detecting these events. However, clinical translation of 3D-PCM remains challenging due to limitations in imaging depth, field of view (FOV), and procedural compatibility. Here, we present a large-FOV dual-modality imaging system (3D-PCM and B-mode ultrasound) based on a large-aperture planar ultrasound array. Through array optimization and model-based reconstruction, our system achieves an expanded FOV of ∼40 × 40 mm 2 at a clinically relevant imaging depth of ∼110 mm, while maintaining high spatial resolution of ∼0.6 mm laterally and ∼0.4 mm axially. In vivo experiments in a porcine model demonstrate that the reconstructed cavitation distribution correlates well with stone damage. Our technology has the potential to provide real-time treatment feedback during LL without disrupting the standard workflow.

Reliable real-time 3D shape sensing is essential for robust control and interpretation of deformable systems during motion. Existing vision-based approaches require line-of-sight and complex instrumentation, limiting operation in occluded and space-constrained settings. Here, we introduce a scalable, skin-like sensor that reconstructs its continuous 3D deformation in real time from distributed strain measurements. The device embeds a 2D array of mirror-stacked, printed oxidized eutectic gallium-indium (o-EGaIn) strain gauges within an elastomeric film to measure off-neutral-axis strains. Combined with a mechanics-informed observation model and a fast optimization routine, the system estimates local curvature, elongation, offset, and orientation under concurrent stretching, bending, and indentation, enabling reconstruction of complex surfaces. A 5-by-5 array with a 12 mm pitch achieves a mean surface reconstruction error of 0.62 mm with 0.1s latency across all tested scenarios. When conforming to complex surfaces, the sensor provides fast 3D shape mapping of the underlying geometry. Demonstrations involving palm gesturing, finger indentation, and contact-induced balloon deformation highlight utility for epidermal motion tracking, haptic interaction, and intraoperative monitoring.

Reliable real-time 3D shape sensing is essential for robust control and interpretation of deformable systems during motion. Existing vision-based approaches require line-of-sight and complex instrumentation, limiting operation in occluded and space-constrained settings. Here, we introduce a scalable, skin-like sensor that reconstructs its continuous 3D deformation in real time from distributed strain measurements. The device embeds a 2D array of mirror-stacked, printed oxidized eutectic gallium-indium (o-EGaIn) strain gauges within an elastomeric film to measure off-neutral-axis strains. Combined with a mechanics-informed observation model and a fast optimization routine, the system estimates local curvature, elongation, offset, and orientation under concurrent stretching, bending, and indentation, enabling reconstruction of complex surfaces. A 5-by-5 array with a 12 mm pitch achieves a mean surface reconstruction error of 0.62 mm with 0.1s latency across all tested scenarios. When conforming to complex surfaces, the sensor provides fast 3D shape mapping of the underlying geometry. Demonstrations involving palm gesturing, finger indentation, and contact-induced balloon deformation highlight utility for epidermal motion tracking, haptic interaction, and intraoperative monitoring.

Wearable Human Activity Recognition (HAR) still lacks a representation that is both explicit and adaptable. Handcrafted time-series features (TSFs) capture meaningful motion statistics and remain competitive on standard benchmarks, but they are usually used as fixed preprocessing outputs. Deep models learn adaptable representations directly from raw signals, but those representations are typically latent and difficult to inspect. We address this gap by treating handcrafted TSFs as feature anchors: explicit intermediate representations that remain inside the model and are adjusted by neural context instead of being discarded. We propose the Temporal Conditioning Network for Feature Anchors (TCNet), which extracts handcrafted anchors, encodes complementary time-domain and frequency-domain context from raw IMU windows, and predicts context-conditioned scale, bias, and gating parameters to modulate anchor groups directly in feature space. This design keeps anchor semantics visible while allowing the representation to adapt to the classification objective. Across five HAR benchmarks, TCNet achieves 70.2% mF1 on USC-HAD, 85.1% mF1 on Daphnet, 93.9% mF1 on MHealth, and 94.5% mF1 on PAMAP2. Relative to rTsfNet, it improves by 4.5 points on USC-HAD, 14.6 points on Daphnet, and 6.5 points on MHealth. Ablations show that the gains come primarily from anchor guidance rather than simple branch fusion, and feature-space analyses indicate that several discriminative TSF families are not reliably accessible in standard latent representations. These results suggest that, for HAR, handcrafted TSFs are most useful when they remain explicit and adaptable within the model. The code is available at: https://github.com/ni-x-lab/TCNet-har

Wearable Human Activity Recognition (HAR) still lacks a representation that is both explicit and adaptable. Handcrafted time-series features (TSFs) capture meaningful motion statistics and remain competitive on standard benchmarks, but they are usually used as fixed preprocessing outputs. Deep models learn adaptable representations directly from raw signals, but those representations are typically latent and difficult to inspect. We address this gap by treating handcrafted TSFs as feature anchors: explicit intermediate representations that remain inside the model and are adjusted by neural context instead of being discarded. We propose the Temporal Conditioning Network for Feature Anchors (TCNet), which extracts handcrafted anchors, encodes complementary time-domain and frequency-domain context from raw IMU windows, and predicts context-conditioned scale, bias, and gating parameters to modulate anchor groups directly in feature space. This design keeps anchor semantics visible while allowing the representation to adapt to the classification objective. Across five HAR benchmarks, TCNet achieves 70.2% mF1 on USC-HAD, 85.1% mF1 on Daphnet, 93.9% mF1 on MHealth, and 94.5% mF1 on PAMAP2. Relative to rTsfNet, it improves by 4.5 points on USC-HAD, 14.6 points on Daphnet, and 6.5 points on MHealth. Ablations show that the gains come primarily from anchor guidance rather than simple branch fusion, and feature-space analyses indicate that several discriminative TSF families are not reliably accessible in standard latent representations. These results suggest that, for HAR, handcrafted TSFs are most useful when they remain explicit and adaptable within the model. The code is available at: https://github.com/ni-x-lab/TCNet-har

Spatial patterning of material phases underpins the functional diversity of natural and engineered composites. However, phase architectures are typically fixed once formed, limiting adaptability. Here, we introduce a digital composite with reprogrammable solid-liquid phase architectures at voxel resolution. Each elastomeric voxel contains a liquid metal composite capable of electrically switching between nonvolatile solid and liquid states within seconds, analogous to rewriting data on a hard disk. High-throughput experiments and coupled modeling demonstrate precise tuning of viscoelastic and plastic properties, as well as programmable constitutive behaviors and strain distributions. A modular assembly strategy allows scalable 3D construction of reprogrammable composites into free-form, bulk geometries. By encoding phase states as digital inputs, the composite unlocks unprecedented access to real-time, voxel-level tuning of material properties.

The study proposes a machine learning framework that optimizes the approval processes of Neo Banks loan applications and overcomes the limitations of traditional credit scoring models. Using feature engineering, ensemble methods, and hyperparameter optimization on customer demographics, financials, and transaction details to determine loan acceptance. The results show that the improved models have the accuracy of 97.87% and 88.59% recall and are far better compared to traditional methods (logistic regression: 63.76% recall). Significant predictors are education level (importance: 0.35), income (0.30), and family size, with the probability of approval among high-income, high-education customers being 45% higher. The framework reduces the false negative rate, allowing the Neo Banks to focus on the cream of the crop applicants but avoiding risks. Examples of practical strategies are customized marketing and dynamic pricing. This work talks about limitations such as data imbalance, and future research suggests integrating real-time behavioral data and fairness-aware modeling. It fills a gap between tech innovation and operational requirements, providing a scalable method for updating credit risk evaluation in digital banking.

Soft Electronics A Lorentz-force-driven, liquid metal–embedded surface delivers rapid, reversible 3D shape morphing for precise low-velocity flow control. With minimal power, it modulates near-wall flows in real time, offering versatile, programmable actuation for small UAVs, bio-inspired aerodynamics, and environmental sensing—bridging soft electronics with advanced fluid dynamics. More details can be found in the Research Article by Donghyun You, Leonardo P. Chamorro, Xinchen Ni, John A. Rogers, and co-workers (DOI: 10.1002/aisy.202500457).

Assessing the mechanical properties of soft tissues holds broad clinical relevance. Advances in flexible electronics offer possibilities for wearable monitoring of tissue stiffness. However, existing technologies often rely on tethered setups or require frequent calibration, restricting their use in ambulatory environments. This study introduces a mechano-acoustic wave sensing technology for automated, wireless elastography. The patch-form sensor maintains conformal contact with the skin, regardless of body motion or deformation. It provides continuous, depth-sensitive estimation of subcutaneous tissue stiffness through real-time surface wave dispersion analysis. Theoretical and experimental investigations on phantom materials and tissues spanning a wide range of Young's modulus (in kilopascals to megapascals) demonstrate the capability of the device to rapidly and robustly evaluate the stiffness at depths up to several centimeters. The device shows compatibility with various tissue models, with results consistent with in-parallel ultrasound elastography measurements. Deployment of the device during exercises confirms its viability for ambulatory monitoring, enabling continuous assessment of variation in tissue stiffness.

We are currently in an era of escalating technological complexity and profound societal transformations, where artificial intelligence (AI) technologies exemplified by large language models (LLMs) have reignited discussions on the 'Technological Singularity'. 'Technological Singularity' is a philosophical concept referring to an irreversible and profound transformation that occurs when AI capabilities surpass those of humans comprehensively. However, quantitative modeling and analysis of the historical evolution and future trends of AI technologies remain scarce, failing to substantiate the singularity hypothesis adequately. This paper hypothesizes that the development of AI technologies could be characterized by the superposition of multiple logistic growth processes. To explore this hypothesis, we propose a multi-logistic growth process model and validate it using two real-world datasets: AI Historical Statistics and Arxiv AI Papers. Our analysis of the AI Historical Statistics dataset assesses the effectiveness of the multi-logistic model and evaluates the current and future trends in AI technology development. Additionally, cross-validation experiments on the Arxiv AI Paper, GPU Transistor and Internet User dataset enhance the robustness of our conclusions derived from the AI Historical Statistics dataset. The experimental results reveal that around 2024 marks the fastest point of the current AI wave, and the deep learning-based AI technologies are projected to decline around 2035-2040 if no fundamental technological innovation emerges. Consequently, the technological singularity appears unlikely to arrive in the foreseeable future.

See DATA.md in CODE.zip (https://github.com/ni-x-lab/PM) for a detailed description of the data. <br>

With the development of robotic welding automation, there is a strong interest in welding seam identification and localization methods with high accuracy, real-time performance, and robustness. This paper proposed a 3D workpiece weld identification and localization method based on DBSCAN (density-based spatial clustering of applications with noise) to realize stable feature extraction for multiple joint types. Firstly, this method employs combinatorial filtering to effectively eliminate non-target point clouds, including outliers and installation platform point clouds, which can minimize the computational load. Secondly, DBSCAN is used to classify workpiece point clouds into different clusters, which can be used for point cloud segmentation of flat workpieces and curved workpieces. Thirdly, the edge detection and feature extraction methods are used to obtain joint gap and weld feature points while combining the information of point clouds for different types of welds. Finally, based on the identification and localization of the welds, welding path planning and attitude planning are implemented. Experimentation results indicated that the proposed method exhibits robustness across various types of welded joints, including butt joints with straight seams, butt joints with curved seams, butt joints with curved workpieces, and lap joints. Meanwhile, the average error of joint gap detection was 0.11 mm and the processing time of a 90 mm straight-seam butt joint is 701.12 ms.

Shape-morphing soft materials are ubiquitous in living systems. They are of increasing relevance to emerging technologies in soft machines, flexible electronics, and smart medicine. The past decade has witnessed phenomenal investment in developing active materials that can shift their shapes, and henceforth their performing functions. However, creating schemes to swiftly reprogram target shapes after fabrication remains challenging. Complexities associated with the operating physics and disturbances from the environment can stop the use of deterministic theoretical models to guide inverse design and control strategies. In this work, we describe a dynamically reprogrammable metasurface with embedded actuation, sensing, and feedback control. The voltage-controlled electromagnetic force drives a flexible, conductive 2D mesh into a diverse set of complex 3D surfaces in the presence of a static magnetic field. The unusual construction of the metasurface from a matrix of filamentary metal traces enables the system to adopt an approximately linear model for inverse design. Implementing an in-situ stereo-imaging feedback strategy with a digitally controlled actuation scheme guided by an optimization algorithm yields surfaces that can morph into target shapes without any presuming models. The closed-loop self-evolving inverse design approach opens opportunities for physical simulations for non-linear or non-ideal systems.

Recent progress in wearable devices provides new opportunities for precise, non-invasive, long-term recording of body mechanics. The soft device incorporating a single MEMS accelerometer captures subtle vibration of the skin with a resolution of 1×10-4 g/√Hz in the frequency range from 0 to 800 Hz. In this study, we introduce an on-body mechano-acoustic sensing technology based on a skin-mounted accelerometer array to assess the mechanical profiles of subdermal tissues in-vivo, similar to seismology. A system-level wearable device construction, optimized for a comfortable skin interface and high precision, incorporates a broadband dual-accelerometer sensor, an audio actuator, and a Bluetooth System-on-Chip and enables a wireless, automated operation. An automated algorithm, leveraging the spectral analysis of surface waves (SASW) methods, computes the depth-sensitive, elastic modulus information of the propagation media from the mechanical dispersion relationship. Comprehensive theoretical and experimental investigation on bi-layer phantom materials and biological tissues, with a storage moduli range of 19–1439 kPa, demonstrates the capability of rapid and robust evaluation of modulus in a depth range of 2–46 mm. The device identifies the softening of porcine tissues with increasing injected water content and the changes of modulus of muscle under different levels of tension. The results are in agreement with the in-parallel Ultrasound Elastography measurements. A quantitative assessment of the stiffness profile of the bicep brachii and rectus femoris muscle during gym exercises demonstrate the device operation in an ambulant environment for a non-invasive and continuous assessment of deep tissue stiffness with a temporal resolution of 0.5 s.