Cerebral aneurysms and brain tumors are leading life-threatening diseases worldwide. By deliberately occluding the target lesion to reduce the blood supply, embolization has been widely used clinically to treat cerebral aneurysms and brain tumors. Conventional embolization is usually performed by threading a catheter through blood vessels to the target lesion, which is often limited by the poor steerability of the catheter in complex neurovascular networks, especially in submillimeter regions. Here, we propose magnetic soft microfiberbots with high steerability, reliable maneuverability, and multimodal shape reconfigurability to perform robotic embolization in submillimeter regions via a remote, untethered, and magnetically controllable manner. Magnetic soft microfiberbots were fabricated by thermal drawing magnetic soft composite into microfibers, followed by magnetizing and molding procedures to endow a helical magnetic polarity. By controlling magnetic fields, magnetic soft microfiberbots exhibit reversible elongated/aggregated shape morphing and helical propulsion in flow conditions, allowing for controllable navigation through complex vasculature and robotic embolization in submillimeter regions. We performed in vitro embolization of aneurysm and tumor in neurovascular phantoms and in vivo embolization of a rabbit femoral artery model under real-time fluoroscopy. These studies demonstrate the potential clinical value of our work, paving the way for a robotic embolization scheme in robotic settings.

DNA computing and DNA data storage are emerging fields that are unlocking new possibilities in information technology and diagnostics. These approaches use DNA molecules as a computing substrate or a storage medium, offering nanoscale compactness and operation in unconventional media (including aqueous solutions, water-in-oil microemulsions and self-assembled membranized compartments) for applications beyond traditional silicon-based computing systems. To build a functional DNA computer that can process and store molecular information necessitates the continued development of strategies for computing and data storage, as well as bridging the gap between these fields. In this Review, we explore how DNA can be leveraged in the context of DNA computing with a focus on neural networks and compartmentalized DNA circuits. We also discuss emerging approaches to the storage of data in DNA and associated topics such as the writing, reading, retrieval and post-synthesis editing of DNA-encoded data. Finally, we provide insights into how DNA computing can be integrated with DNA data storage and explore the use of DNA for near-memory computing for future information technology and health analysis applications.

Egocentric vision has gained increasing popularity recently, opening new avenues for human-centric applications. However, the use of the egocentric fisheye cameras allows wide angle coverage but image distortion is introduced along with strong human body self-occlusion imposing significant challenges in data processing and model reconstruction. Unlike previous work only leveraging synthetic data for model training, this paper presents a new real-world EgoCentric Human Pose (ECHP) dataset. To tackle the difficulty of collecting 3D ground truth using motion capture systems, we simultaneously collect images from a head-mounted egocentric fisheye camera as well as from two third-person-view cameras, circumventing the environmental restrictions. By using self-supervised learning under multi-view constraints, we propose a simple yet effective framework, namely EgoFish3D, for egocentric 3D pose estimation from a single image in different real-world scenarios. The proposed EgoFish3D incorporates three main modules. 1) The third-person-view module takes two exocentric images as input and estimates the 3D pose represented in the third-person camera frame; 2) the egocentric module predicts the 3D pose in the egocentric camera frame; and 3) the interactive module estimates the rotation matrix between the third-person and the egocentric views. Experimental results on our ECHP dataset and existing benchmark datasets demonstrate the effectiveness of the proposed EgoFish3D, which can achieve superior performance to existing methods.

Parallel imaging is a commonly used technique to accelerate magnetic resonance imaging (MRI) data acquisition. Mathematically, parallel MRI reconstruction can be formulated as an inverse problem relating the sparsely sampled k-space measurements to the desired MRI image. Despite the success of many existing reconstruction algorithms, it remains a challenge to reliably reconstruct a high-quality image from highly reduced k-space measurements. Recently, implicit neural representation has emerged as a powerful paradigm to exploit the internal information and the physics of partially acquired data to generate the desired object. In this study, we introduced IMJENSE, a scan-specific implicit neural representationbased method for improving parallel MRI reconstruction. Specifically, the underlying MRI image and coil sensitivities were modeled as continuous functions of spatial coordinates, parameterized by neural networks and polynomials, respectively. The weights in the networks and coefficients in the polynomials were simultaneously learned directly from sparsely acquired k-space measurements, without fully sampled ground truth data for training. Benefiting from the powerful continuous representation and joint estimation of the MRI image and coil sensitivities, IMJENSE outperforms conventional image or k-space domain reconstruction algorithms. With extremely limited calibration data, IMJENSE is more stable than supervised calibrationless and calibrationbased deep-learning methods. Results show that IMJENSE robustly reconstructs the images acquired at 5× and 6× accelerations with only 4 or 8 calibration lines in 2D Cartesian acquisitions, corresponding to 22.0% and 19.5% undersampling rates. The high-quality results and scanning specificity make the proposed method hold the potential for further accelerating the data acquisition of parallel MRI.

Accurate navigation and targeting are critical for neurological interventions including biopsy and deep brain stimulation. Real-time image guidance further improves surgical planning and MRI is ideally suited for both pre- and intra-operative imaging. However, balancing spatial and temporal resolution is a major challenge for real-time interventional MRI (i-MRI). Here, we proposed a deep unrolled neural network, dubbed as LSFP-Net, for real-time i-MRI reconstruction. By integrating LSFP-Net and a custom-designed, MR-compatible interventional device into a 3 T MRI scanner, a real-time MRI-guided brain intervention system is proposed. The performance of the system was evaluated using phantom and cadaver studies. 2D/3D real-time i-MRI was achieved with temporal resolutions of 80/732.8 ms, latencies of 0.4/3.66 s including data communication, processing and reconstruction time, and in-plane spatial resolution of 1 × 1 mm2. The results demonstrated that the proposed method enables real-time monitoring of the remote-controlled brain intervention, and showed the potential to be readily integrated into diagnostic scanners for image-guided neurosurgery.

Human-infecting pathogens that transmit through the air pose a significant threat to public health. As a prominent instance, the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) that caused the COVID-19 pandemic has affected the world in an unprecedented manner over the past few years. Despite the dissipating pandemic gloom, the lessons we have learned in dealing with pathogen-laden aerosols should be thoroughly reviewed because the airborne transmission risk may have been grossly underestimated. From a bioanalytical chemistry perspective, on-site airborne pathogen detection can be an effective non-pharmaceutic intervention (NPI) strategy, with on-site airborne pathogen detection and early-stage infection risk evaluation reducing the spread of disease and enabling life-saving decisions to be made. In light of this, we summarize the recent advances in highly efficient pathogen-laden aerosol sampling approaches, bioanalytical sensing technologies, and the prospects for airborne pathogen exposure measurement and evidence-based transmission interventions. We also discuss open challenges facing general bioaerosols detection, such as handling complex aerosol samples, improving sensitivity for airborne pathogen quantification, and establishing a risk assessment system with high spatiotemporal resolution for mitigating airborne transmission risks. This review provides a multidisciplinary outlook for future opportunities to improve the on-site airborne pathogen detection techniques, thereby enhancing the preparedness for more on-site bioaerosols measurement scenarios, such as monitoring high-risk pathogens on airplanes, weaponized pathogen aerosols, influenza variants at the workplace, and pollutant correlated with sick building syndromes.

When developing context-aware systems, automatic surgical phase recognition and tool presence detection are two essential tasks. There exist previous attempts to develop methods for both tasks but majority of the existing methods utilize a frame-level loss function (e.g., cross-entropy) which does not fully leverage the underlying semantic structure of a surgery, leading to sub-optimal results. In this paper, we propose multi-task learning-based, LAtent Space-constrained Transformers, referred as LAST, for automatic surgical phase recognition and tool presence detection. Our design features a two-branch transformer architecture with a novel and generic way to leverage video-level semantic information during network training. This is done by learning a non-linear compact presentation of the underlying semantic structure information of surgical videos through a transformer variational autoencoder (VAE) and by encouraging models to follow the learned statistical distributions. In other words, LAST is of structure-aware and favors predictions that lie on the extracted low dimensional data manifold. Validated on two public datasets of the cholecystectomy surgery, i.e., the Cholec80 dataset and the M2cai16 dataset, our method achieves better results than other state-of-the-art methods. Specifically, on the Cholec80 dataset, our method achieves an average accuracy of 93.12±4.71%, an average precision of 89.25±5.49%, an average recall of 90.10±5.45% and an average Jaccard of 81.11 ±7.62% for phase recognition, and an average mAP of 95.15±3.87% for tool presence detection. Similar superior performance is also observed when LAST is applied to the M2cai16 dataset.

Magnetic fiber robots have revealed great potential for future minimally invasive robotic surgery. However, with the miniaturization of fiber robots, the integration of functional microtools at the distal end becomes extremely challenging, since it requires assembling multifunctional structures on the curved surfaces of fiber ends, which typically have very small curvature radii. In this study, a submillimeter fiber robot with integrated micro-manipulation tool at the distal end is reported using a “Swiss roll” assembly strategy. This approach enables the one-step assembly of multiscale structures (mm-µm) from a 2D film to a 3D tool on the curved surface of the fiber robot. The multiscale structure consists of a millimeter-scale (mm) magnetic thin film with integrated micrometer-scale (µm) feature structures, which is inspired from the cat tongue covered by numerous little papillae. The fiber robot can perform multiple functions including endovascular clot grabbing, liquid delivery, and sampling under the manipulation of the magnetic field. The strategy provides a universal protocol for integrating and assembling functional components at the distal end of fiber robots, contributing significantly to the functionalization and miniaturization of interventional medical robots.

Surface-enhanced Raman spectroscopy (SERS), well acknowledged as afingerprinting and sensitive analytical technique, has exerted highapplicational value in a broad range of fields including biomedicine,environmental protection, food safety among the others. In the endlesspursuit of ever-sensitive, robust, and comprehensive sensing and imaging,advancements keep emerging in the whole pipeline of SERS, from the designof SERS substrates and reporter molecules, synthetic route planning,instrument refinement, to data preprocessing and analysis methods. Artificialintelligence (AI), which is created to imitate and eventually exceed humanbehaviors, has exhibited its power in learning high-level representations andrecognizing complicated patterns with exceptional automaticity. Therefore,facing up with the intertwining influential factors and explosive data size, AIhas been increasingly leveraged in all the above-mentioned aspects in SERS,presenting elite efficiency in accelerating systematic optimization anddeepening understanding about the fundamental physics and spectral data,which far transcends human labors and conventional computations. In thisreview, the recent progresses in SERS are summarized through the integrationof AI, and new insights of the challenges and perspectives are provided in aimto better gear SERS toward the fast track.

The operating skills of vascular interventionists have an important impact on the effect of surgery. However, current research on behavior recognition and skills learning of interventionists' operating skills is limited. In this study, an innovative deep learning-based multimodal information fusion architecture is proposed for recognizing and analyzing eight common operating behaviors of interventionists. An experimental platform integrating four modal sensors is used to collect multimodal data from interventionists. The ANOVA and Manner-Whitney tests is used for relevance analysis of the data. The analysis results demonstrate that there is almost no significant difference (p <0.001) between the actions related to the unimodal data, which cannot be used for accurate behavior recognition. Therefore, a study of the fusion architecture based on the existing machine learning classifier and the proposed deep learning fusion architecture is carried out. The research findings indicate that the proposed deep learning-based fusion architecture achieves an impressive overall accuracy of 98.5%, surpassing both the machine learning classifier (93.51%) and the unimodal data (90.05%). The deep learning-based multimodal information fusion architecture proves the feasibility of behavior recognition and skills learning of interventionist's operating skills. Furthermore, the application of deep learning-based multimodal fusion technology of surgeon's operating skills will help to improve the autonomy and intelligence of surgical robotic systems.