Journals
2026 EN
Jung Myungjin · Kim Sejeong · Kim MinCheol
+3 more
This study presents a novel cable‐driven parallel robot (CDPR) assisted needle insertion method for X‐ray guided remote interventional pain procedures. The CDPR employs flexible cables to actuate a robotic end‐effector, and the proposed system ensures compatibility with X‐ray imaging while facilitating precise remote needle insertion by achieving a virtual remote center of motion. The proposed system addresses challenges associated with conventional rigid‐link type needle insertion robots in terms of a limited workspace and X‐ray interference. Design, workspace analysis, prototyping, control, and experimental results for feasibility validation are conducted to demonstrate the effectiveness in achieving of accurate needle guidance under C‐arm imaging. The gelatin phantom experiments confirmed the motion accuracy and the cadaver experiment underscored the system's feasibility for clinical applications. The proposed approach to robotic assistance in interventional pain procedures may enhance precision and reduce radiation exposure for both patients and clinicians.
Journals
2026 EN
Gu Zishan · Chen Jiayuan · Liu Fenglin
+2 more
Large vision language models (LVLMs) have achieved superior performance on natural image and text tasks, inspiring extensive fine‐tuning research. However, their robustness against hallucination in clinical contexts remains understudied. We propose the Medical Visual Hallucination Test (MedVH), a novel evaluation framework assessing hallucination tendencies in both medical‐specific and general‐purpose LVLMs. MedVH encompasses six tasks targeting medical hallucinations, including two traditional tasks and four novel tasks formatted as multi‐choice visual question answering and long response generation. Our extensive experiments with six evaluation metrics reveal that medical LVLMs, despite promising performance on standard medical tasks, are particularly susceptible to hallucinations—often more so than general models. This raises significant concerns about domain‐specific model reliability. For real‐world applications, medical LVLMs must accurately integrate medical knowledge while maintaining robust reasoning to prevent hallucination. We explore mitigation methods without model‐specific fine‐tuning, including prompt engineering and collaboration between general and domain‐specific models. Our work provides a foundation for future evaluation studies. The dataset is available at PhysioNet: https://physionet.org/content/medvh.
Journals
2026 EN
Jaberi Mina · Falk Tiago H.
The Metaverse is revolutionizing various fields, including healthcare, education, social interaction, and the workplace. Commercial multisensory devices (e.g., smell diffusion and haptic technologies) are available, and virtual and augmented reality (VR/AR) headsets are increasingly integrated with brain–computer interfaces (BCI). These integrations enable adaptive, personalized virtual immersive experiences that are more engaging, interactive, and effective. As these applications become mainstream, concerns arise regarding the security and privacy of personal information. Recent studies demonstrate that users can be identified with high accuracy using the data monitored from sensors available in VR/AR headsets. This literature survey investigates the types of personal user information that can be inferred from BCI‐instrumented headsets. In particular, it focuses on predicting age, gender, and ethnic/racial background from neurophysiological signals currently monitored by commercial devices. The survey highlights the predictive strength of electroencephalogram and electrocardiogram signal modalities, followed by eye tracking and iris scanning. It also considers future privacy risks posed by biometric and gesture‐based monitoring using non‐contact technologies such as computer vision and WiFi signal analysis. The survey concludes with recommendations for future research aimed at contributing to the development of robust frameworks that safeguard user privacy in the evolving Metaverse landscape.
Journals
2026 EN
Sinawang Prima Dewi · GarciaGradilla Victor · Soto Fernando
+4 more
Smart capsules are promising tools for minimally invasive sampling. However, existing designs rely on passive diffusion, which is ineffective for viscous samples such as mucus. To address this limitation, we present screw‐based pill for intelligent robotic extraction (S‐PIRE), designed for active sampling of viscous materials. S‐PIRE features a motorized 3D‐printed hydrodynamic screw, remotely activated via a Bluetooth microcontroller. Its magnetic actuation system ensures controlled positioning, while its integrated 3‐axis Hall Effect magnetic sensor provides spatial localization. S‐PIRE is magnetically guided for collecting viscous samples in vitro to demonstrate its sampling mechanism. Samples are securely stored in a detachable, disposable chamber for analysis. This proof of concept demonstrates S‐PIRE potential for in vivo applications, offering a precise, minimally invasive approach for future targeted mucus sampling from difficult‐to‐reach regions of the body like the gastrointestinal tract, which could aid in early disease detection and biomarker discovery.
Journals
2026 EN
Sakemi Yusuke · Okamoto Yuji · Morie Takashi
+3 more
Large‐scale deep learning models are increasingly constrained by their immense energy consumption, which limits their scalability and applicability for edge intelligence. In‐memory computing (IMC) offers a promising solution by addressing the von Neumann bottleneck inherent in traditional deep learning accelerators, significantly reducing energy consumption. However, the analog nature of IMC introduces hardware nonidealities that degrade model performance and reliability. This article presents a novel approach to directly train physical models of IMC, formulated as ordinary differential equation (ODE)‐based physical neural networks (PNNs). To enable the training of large‐scale networks, a technique called differentiable spike‐time discretization is proposed, which reduces the computational cost of ODE‐based PNNs by up to 20 times in speed and 100 times in memory. Such large‐scale networks enhance learning performance by exploiting hardware nonidealities on the CIFAR‐10 dataset. The proposed bottom‐up methodology is validated through post‐layout SPICE simulations on the IMC circuit with nonideal characteristics using the sky130 process. The proposed PNN approach reduces the discrepancy between model behavior and circuit dynamics by at least an order of magnitude. This work paves the way for leveraging nonideal physical devices, such as nonvolatile resistive memories, for energy‐efficient deep learning applications.
Journals
2026 EN
Su Zhaowen · Fang Lijun · Kim Hoyeon
+1 more
Automatic navigation of microrobot swarms driven by magnetic fields has attracted considerable attention due to potential applications in biomedical fields. However, achieving minimal loss and maintaining swarm cohesion while traversing heterogeneous landscapes over long distances remains a challenge. This article introduces a control strategy based on autonomous field‐of‐view (FOV) planning for navigating microrobot swarms across large workspaces that span multiple FOVs. High‐resolution global images of the workspace are obtained using an image stitching method that combines phase correlation and template matching. Global path planning is accomplished with the A* algorithm, followed by local path planning utilizes the optimized informed rapidly‐exploring random tree star (OI‐RRT*) algorithm in each FOV to ensure swarm adaptation. The strategy also incorporates an FOV planning algorithm to optimize FOV positioning, along with a displacement platform to ensure smooth transitions between FOVs. A real‐time visual feedback control system monitors both channel width and swarm position. This strategy improves swarm navigation efficiency and stability, as demonstrated through experimental validation, and holds significant potential for targeted drug delivery and other biomedical applications.
Journals
2026 EN
Zhao Song · Zhang Liwen · Zhang Shengbin
+6 more
Electrical stimulation and microneedle‐mediated drug delivery emerge as promising therapies in gastrointestinal (GI) motility disorders and inflammatory conditions. However, on‐demand intervention therapy in enclosed narrow GI remains a challenge. Herein, a magnetic‐driven soft membrane robot is presented that synergistically combines microneedle‐mediated electrical stimulation and drug delivery. The membrane robot's bipolar magnetization enables switching between two surfaces by external magnetic fields, where N‐pole drives treatment surface with microneedle to penetrate GI wall and S‐pole initiates smooth surface for low resistance locomotion. The membrane robot utilizes magnetically coupled resonant wireless transmission to enable regulated electrical stimulation with 86.7% efficiency at 6 cm distance, while providing tunable voltage (0–20 V) and programmable pulse waveforms (0.4–50 ms width) for adaptive bioelectrical modulation. The drug‐loaded microneedle array serves dual roles as both a penetrating electrode and a therapeutic interface, delivering electrical stimulation while simultaneously releasing encapsulated agents upon tissue penetration. In vitro experiments of the multimode motion and multifunctional treatment are validated in a fresh pig gut. This integrated membrane magnetic robot offers great potential in GI diagnostics, personalized neuromodulation, and on‐demand drug release applications.
Journals
2026 EN
Wang Shuai · Xiao Zhuoran · Wang Meng
+4 more
Prolonged invasive mechanical ventilation can lead to secretion retention in airways, which increases the risk of airway obstruction and lung infections. Clearing secretions in time is essential to guarantee ventilation effectiveness and patients’ safety. However, manual and scheduled auscultations, widely adopted in clinical practices, can hardly continuously recognize the extent of secretion retentions to determine an optimal timing of secretion suction, which also place a burden on clinicians. Herein, an autonomous recognition model of secretion retention is proposed based on a ResNet‐34 convolutional neural network to perform triple classifications, which represent three levels of secretion retention, namely no secretions, secretions appear, and secretions retained. Respiratory sounds in central airways are continuously recorded and fed into the model in the form of Mel spectrograms. Experimental results demonstrate that the model can continuously monitor the breathing sounds and autonomously recognize different extent of secretion retentions in real time with overall accuracy of 89.08%. An edge computing module to integrate the recognition model into ventilators with limited computing power is also developed and approved, which indicates its potential clinical applications. The proposed autonomous recognition model facilitates clinicians to optimize secretion‐suction schedules to guarantee the ventilation effectiveness and patients’ safety.
Journals
2026 EN
Yun Yumin · Park Junhyeong · Kong Minsik
+4 more
Neuromorphic computing for spiking neural networks (SNNs) has attracted tremendous attention for next‐generation computing in various applications. Amorphous indium–gallium–zinc–oxide (IGZO) is a promising candidate for optoelectronic synaptic transistors and neuron circuits due to its photoconductivity, low‐temperature fabrication process, and extremely low leakage current. Herein, an IGZO‐based optoelectronic neuromorphic system integrated with light guides is reported. The IGZO‐based optoelectronic synaptic transistor demonstrates high retention through a negative gate bias by the IGZO‐based neuron circuit and successfully emulates synaptic plasticity such as long‐term potentiation and long‐term depression. The proposed IGZO‐based neuron circuit with a double‐gate thin‐film transistor operates a dual role as a pulse generator in the programming stage and an integrate‐and‐fire neuron in the inference stage. The operation of the proposed neuron circuit through experiments and simulations is verified. The designed parameters are optimized by considering the interaction between synapses and neurons through SNN simulation with the MNIST dataset. Synapse–neuron codesign with the same fabrication in this study facilitates advancements in future optoelectronic neuromorphic systems.
Journals
2026 EN
Di Weicheng · Dang Shihao · Gong Zheng
+5 more
Under restricted global positioning access, navigating micro aerial vehicles (MAVs) is particularly challenging. Therefore, the ability to autonomously estimate velocity and position based on onboard sensors becomes critical. While vision or radar‐based approaches face limitations for MAVs due to payload constraints, poor lighting, or featureless environments, bio‐inspired airflow sensing offers a promising alternative. Airflow interaction with MAVs provides continuous motion cues during flight, enabling airflow odometry—a feasible yet accuracy‐limited solution. This article presents a novel skin‐like airflow odometry system using distributed flexible thermal anemometers embedded in wingtips, allowing real‐time motion estimation where conventional methods fail. Computational fluid dynamics is first employed to analyze the feasibility and sensitivity without changing their aerodynamic profile. A transformer‐enhanced gated recurrent unit network then extracts high‐precision airflow velocity, while model‐based multisensor fusion reduces dead‐reckoning drift and is validated through experiments. To the authors’ knowledge, this demonstrates the first airflow odometry system with positional accuracy surpassing previous research for MAV applications.