A Deep Reinforcement Learning Strategy for Surrounding Vehicles-Based Lane-Keeping Control.

As autonomous vehicles (AVs) are advancing to higher levels of autonomy and performance, the associated technologies are becoming increasingly diverse. Lane-keeping systems (LKS), corresponding to a key functionality of AVs, considerably enhance driver convenience. With drivers increasingly relying on autonomous driving technologies, the importance of safety features, such as fail-safe mechanisms in the event of sensor failures, has gained prominence. Therefore, this paper proposes a reinforcement learning (RL) control method for lane-keeping, which uses surrounding object information derived through LiDAR sensors instead of camera sensors for LKS. This approach uses surrounding vehicle and object information as observations for the RL framework to maintain the vehicle's current lane. The learning environment is established by integrating simulation tools, such as IPG CarMaker, which incorporates vehicle dynamics, and MATLAB Simulink for data analysis and RL model creation. To further validate the applicability of the LiDAR sensor data in real-world settings, Gaussian noise is introduced in the virtual simulation environment to mimic sensor noise in actual operational conditions.

TransNet: Transformer-Based Point Cloud Sampling Network.

As interest in point cloud processing has gradually increased in the industry, point cloud sampling techniques have been researched to improve deep learning networks. As many conventional models use point clouds directly, the consideration of computational complexity has become critical for practicality. One of the representative ways to decrease computations is downsampling, which also affects the performance in terms of precision. Existing classic sampling methods have adopted a standardized way regardless of the task-model property in learning. However, this limits the improvement of the point cloud sampling network's performance. That is, the performance of such task-agnostic methods is too low when the sampling ratio is high. Therefore, this paper proposes a novel downsampling model based on the transformer-based point cloud sampling network (TransNet) to efficiently perform downsampling tasks. The proposed TransNet utilizes self-attention and fully connected layers to extract meaningful features from input sequences and perform downsampling. By introducing attention techniques into downsampling, the proposed network can learn about the relationships between point clouds and generate a task-oriented sampling methodology. The proposed TransNet outperforms several state-of-the-art models in terms of accuracy. It has a particular advantage in generating points from sparse data when the sampling ratio is high. We expect that our approach can provide a promising solution for downsampling tasks in various point cloud applications.

Automated Laser-Fiber Coupling Module for Optical-Resolution Photoacoustic Microscopy.

Photoacoustic imaging has emerged as a promising biomedical imaging technique that enables visualization of the optical absorption characteristics of biological tissues in vivo. Among the different photoacoustic imaging system configurations, optical-resolution photoacoustic microscopy stands out by providing high spatial resolution using a tightly focused laser beam, which is typically transmitted through optical fibers. Achieving high-quality images depends significantly on optical fluence, which is directly proportional to the signal-to-noise ratio. Hence, optimizing the laser-fiber coupling is critical. Conventional coupling systems require manual adjustment of the optical path to direct the laser beam into the fiber, which is a repetitive and time-consuming process. In this study, we propose an automated laser-fiber coupling module that optimizes laser delivery and minimizes the need for manual intervention. By incorporating a motor-mounted mirror holder and proportional derivative control, we successfully achieved efficient and robust laser delivery. The performance of the proposed system was evaluated using a leaf-skeleton phantom in vitro and a human finger in vivo, resulting in high-quality photoacoustic images. This innovation has the potential to significantly enhance the quality and efficiency of optical-resolution photoacoustic microscopy.

Whole-Body Photoacoustic Imaging Techniques for Preclinical Small Animal Studies.

Photoacoustic imaging is a hybrid imaging technique that has received considerable attention in biomedical studies. In contrast to pure optical imaging techniques, photoacoustic imaging enables the visualization of optical absorption properties at deeper imaging depths. In preclinical small animal studies, photoacoustic imaging is widely used to visualize biodistribution at the molecular level. Monitoring the whole-body distribution of chromophores in small animals is a key method used in preclinical research, including drug-delivery monitoring, treatment assessment, contrast-enhanced tumor imaging, and gastrointestinal tracking. In this review, photoacoustic systems for the whole-body imaging of small animals are explored and summarized. The configurations of the systems vary with the scanning methods and geometries of the ultrasound transducers. The future direction of research is also discussed with regard to achieving a deeper imaging depth and faster imaging speed, which are the main factors that an imaging system should realize to broaden its application in biomedical studies.

An Enhanced Affine Projection Algorithm Based on the Adjustment of Input-Vector Number.

An enhanced affine projection algorithm (APA) is proposed to improve the filter performance in aspects of convergence rate and steady-state estimation error, since the adjustment of the input-vector number can be an effective way to increase the convergence rate and to decrease the steady-state estimation error at the same time. In this proposed algorithm, the input-vector number of APA is adjusted reasonably at every iteration by comparing the averages of the accumulated squared errors. Although the conventional APA has the constraint that the input-vector number should be integer, the proposed APA relaxes that integer-constraint through a pseudo-fractional method. Since the input-vector number can be updated at every iteration more precisely based on the pseudo-fractional method, the filter performance of the proposed APA can be improved. According to our simulation results, it is demonstrated that the proposed APA has a smaller steady-state estimation error compared to the existing APA-type filters in various scenarios.

A Variable-Sampling Time Model Predictive Control Algorithm for Improving Path-Tracking Performance of a Vehicle.

This paper proposes a novel model predictive control (MPC) algorithm that increases the path tracking performance according to the control input. The proposed algorithm reduces the path tracking errors of MPC by updating the sampling time of the next step according to the control inputs (i.e., the lateral velocity and front steering angle) calculated in each step of the MPC algorithm. The scenarios of a mixture of straight and curved driving paths were constructed, and the optimal control input was calculated in each step. In the experiment, a scenario was created with the Automated Driving Toolbox of MATLAB, and the path-following performance characteristics and computation times of the existing and proposed MPC algorithms were verified and compared with simulations. The results prove that the proposed MPC algorithm has improved path-following performance compared to those of the existing MPC algorithm.

Multifunctional Nanodroplets Encapsulating Naphthalocyanine and Perfluorohexane for Bimodal Image-Guided Therapy.

Although nanocarriers containing perfluorocarbon (PFC) have been widely investigated as an ultrasound (US) imaging agent and a high intensity focused ultrasound (HIFU) agent, these carriers have suffered from low stability and biocompatibility limiting their further biomedical applications. Here, we developed surface cross-linked polymer nanodroplets as a HIFU therapeutic agent guided by bimodal photoacoustic (PA) and US imaging. Pluronic F127 was reacted with 4-nitrophenyl chloroformate (NPC) and mixed with naphthalocyanine (Nc) in dichloromethane, which was added into the aqueous solution of amine-functionalized six-arm-branched poly(ethylene glycol) (PEG) to form an oil-in-water emulsion for the cross-linking reaction between the terminal NPC of Pluronic F127 and the primary amine of six-arm PEG. The resulting solution was sonicated with liquid perfluorohexane (PFH) to prepare PEG cross-linked Pluronic F127 nanoparticles encapsulating Nc and PFH (Nc/PFH@PCPN). Nc/PFH@PCPN appeared to be stable without any coalescence or vaporization in the physiological condition. Upon the application of HIFU, Nc/PFH@PCPN was vaporized and showed increased US intensity for 180 min. The Nc dye in the nanodroplets enabled the stable encapsulation of PFH and the bimodal US/PA imaging. In vivo PA/US image-guided HIFU ablation therapy confirmed that the nanodroplets increased the cavitation effect, induced necrosis and apoptosis of tumor cells, and reduced tumor growth significantly for 12 days. Taken together, the multifunctional Nc/PFH@PCPN was successfully developed as a new platform for PA/US image-guided HIFU therapy.

Smart Microbubble Eluting Theranostic Stent for Noninvasive Ultrasound Imaging and Prevention of Restenosis.

A pH-responsive microbubble-eluting theranostic stent is developed for real-time ultrasound imaging of stent implanted blood vessels and dissolution of fat-rich plaques to prevent the blocking of blood vessels in rats. This smart theranostic stent can be effectively applied to facilitate noninvasive monitoring and prevent restenosis after stent implantation.

A Multifunctional Subphthalocyanine Nanosphere for Targeting, Labeling, and Killing of Antibiotic-Resistant Bacteria.

Developing a material that can combat antibiotic-resistant bacteria, a major global health threat, is an urgent requirement. To tackle this challenge, we synthesized a multifunctional subphthalocyanine (SubPc) polymer nanosphere that has the ability to target, label, and photoinactivate antibiotic-resistant bacteria in a single treatment with more than 99 % efficiency, even with a dose as low as 4.2 J cm(-2) and a loading concentration of 10 nM. The positively charged nanosphere shell composed of covalently linked SubPc units can increase the local concentration of photosensitizers at therapeutic sites. The nanosphere shows superior performance compared to corresponding monomers presumably because of their enhanced water dispersibility, higher efficiency of singlet-oxygen generation, and phototoxicity. In addition, this material is useful in fluorescence labeling of living cells and shows promise in photoacoustic imaging of bacteria in vivo.

In vivo photoacoustic and fluorescence cystography using clinically relevant dual modal indocyanine green.

Conventional X-ray-based cystography uses radio-opaque materials, but this method uses harmful ionizing radiation and is not sensitive. In this study, we demonstrate nonionizing and noninvasive photoacoustic (PA) and fluorescence (FL) cystography using clinically relevant indocyanine green (ICG) in vivo. After transurethral injection of ICG into rats through a catheter, their bladders were photoacoustically and fluorescently visualized. A deeply positioned bladder below the skin surface (i.e., ~1.5-5 mm) was clearly visible in the PA and FL image using a laser pulse energy of less than 2 mJ/cm2 (1/15 of the safety limit). Then, the in vivo imaging results were validated through in situ studies. Our results suggest that dual modal cystography can provide a nonionizing and noninvasive imaging tool for bladder mapping.

Smart laser Sintering: Deep Learning-Powered powder bed fusion 3D printing in precision medicine.

Medicines remain ineffective for over 50% of patients due to conventional mass production methods with fixed drug dosages. Three-dimensional (3D) printing, specifically selective laser sintering (SLS), offers a potential solution to this challenge, allowing the manufacturing of small, personalized batches of medication. Despite its simplicity and suitability for upscaling to large-scale production, SLS was not designed for pharmaceutical manufacturing and necessitates a time-consuming, trial-and-error adaptation process. In response, this study introduces a deep learning model trained on a variety of features to identify the best feature set to represent drugs and polymeric materials for the prediction of the printability of drug-loaded formulations using SLS. The proposed model demonstrates success by achieving 90% accuracy in predicting printability. Furthermore, explainability analysis unveils materials that facilitate SLS printability, offering invaluable insights for scientists to optimize SLS formulations, which can be expanded to other disciplines. This represents the first study in the field to develop an interpretable, uncertainty-optimized deep learning model for predicting the printability of drug-loaded formulations. This paves the way for accelerating formulation development, propelling us into a future of personalized medicine with unprecedented manufacturing precision.

Functional photoacoustic imaging: from nano- and micro- to macro-scale.

Functional photoacoustic imaging is a promising biological imaging technique that offers such unique benefits as scalable resolution and imaging depth, as well as the ability to provide functional information. At nanoscale, photoacoustic imaging has provided super-resolution images of the surface light absorption characteristics of materials and of single organelles in cells. At the microscopic and macroscopic scales. photoacoustic imaging techniques have precisely measured and quantified various physiological parameters, such as oxygen saturation, vessel morphology, blood flow, and the metabolic rate of oxygen, in both human and animal subjects. This comprehensive review provides an overview of functional photoacoustic imaging across multiple scales, from nano to macro, and highlights recent advances in technology developments and applications. Finally, the review surveys the future prospects of functional photoacoustic imaging in the biomedical field.

Recent Advances in Photoacoustic Agents for Theranostic Applications.

Photoacoustic agents are widely used in various theranostic applications. By evaluating the biodistribution obtained from photoacoustic images, the effectiveness of theranostic agents in terms of their delivery efficiency and treatment responses can be analyzed. Through this study, we evaluate and summarize the recent advances in photoacoustic-guided phototherapy, particularly in photothermal and photodynamic therapy. This overview can guide the future directions for theranostic development. Because of the recent applications of photoacoustic imaging in clinical trials, theranostic agents with photoacoustic monitoring have the potential to be translated into the clinical world.

Review on ultrasound-guided photoacoustic imaging for complementary analyses of biological systems in vivo.

Photoacoustic imaging has been developed as a new biomedical molecular imaging modality. Due to its similarity to conventional ultrasound imaging in terms of signal detection and image generation, dual-modal photoacoustic and ultrasound imaging has been applied to visualize physiological and morphological information in biological systems in vivo. By complementing each other, dual-modal photoacoustic and ultrasound imaging showed synergistic advances in photoacoustic imaging with the guidance of ultrasound images. In this review, we introduce our recent progresses in dual-modal photoacoustic and ultrasound imaging systems at various scales of study, from preclinical small animals to clinical humans. A summary of the works reveals various strategies for combining the structural information of ultrasound images with the molecular information of photoacoustic images.

Deep brain stimulation by blood-brain-barrier-crossing piezoelectric nanoparticles generating current and nitric oxide under focused ultrasound.

Deep brain stimulation via implanted electrodes can alleviate neuronal disorders. However, its applicability is constrained by side effects resulting from the insertion of electrodes into the brain. Here, we show that systemically administered piezoelectric nanoparticles producing nitric oxide and generating direct current under high-intensity focused ultrasound can be used to stimulate deep tissue in the brain. The release of nitric oxide temporarily disrupted tight junctions in the blood-brain barrier, allowing for the accumulation of the nanoparticles into brain parenchyma, and the piezoelectrically induced output current stimulated the release of dopamine by dopaminergic neuron-like cells. In a mouse model of Parkinson's disease, the ultrasound-responsive nanoparticles alleviated the symptoms of the disease without causing overt toxicity. The strategy may inspire the development of other minimally invasive therapies for neurodegenerative diseases.

Review on Multispectral Photoacoustic Analysis of Cancer: Thyroid and Breast.

In recent decades, photoacoustic imaging has been used widely in biomedical research, providing molecular and functional information from biological tissues in vivo. In addition to being used for research in small animals, photoacoustic imaging has also been utilized for in vivo human studies, achieving a multispectral photoacoustic response in deep tissue. There have been several clinical trials for screening cancer patients by analyzing multispectral responses, which in turn provide metabolomic information about the underlying biological tissues. This review summarizes the methods and results of clinical photoacoustic trials available in the literature to date to classify cancerous tissues, specifically of the thyroid and breast. From the review, we can conclude that a great potential exists for photoacoustic imaging to be used as a complementary modality to improve diagnostic accuracy for suspicious tumors, thus significantly benefitting patients' healthcare.

Listening to drug delivery and responses via photoacoustic imaging.

Administrating pharmaceutic agents efficiently to achieve the therapeutic effect is the aim of all drug delivery techniques. Recent drug delivery systems aim to deliver high doses of drugs to disease sites accurately while maximizing therapeutic effects and minimizing potential side effects. Key approaches apply image guidance techniques for the quantification of drug biodistribution and pharmacokinetic parameters during drug delivery. This review highlights recent research on image-guided drug delivery systems based on photoacoustic imaging, which has been attracting attention for its non-invasiveness, non-ionizing radiation, and real-time imaging functions. Photoacoustic imaging based on the photothermal conversion efficiency of agents can be easily combined with various phototherapeutics, making them highly suitable for drug delivery therapy platforms. Here, we summarize and compare the characteristics of various types of photoacoustic imaging systems, focus on contrast-enhanced photoacoustic imaging and controlled release of therapeutics in drug delivery systems for synergistic therapies.

Photoacoustic imaging systems based on clinical ultrasound platform.

Photoacoustic imaging has drawn a significant amount of attention due to its unique capacity for functional, metabolic, and molecular imaging, which is achieved by the combination of optical excitation and acoustic detection. With both strengths of light and ultrasound, photoacoustic images can provide strong optical contrast at high ultrasound resolution in deep tissue. As photoacoustic imaging can be used to visualize complementary information to ultrasound imaging using the same data acquisition process, several studies have been conducted on combining photoacoustic imaging with existing clinical ultrasound systems. This review highlights our development of a photoacoustic/ultrasound dual-modal imaging system, various features and functionalities implemented for clinical translation, and preclinical/clinical studies performed by using the systems.

State of the Art in Carbon Nanomaterials for Photoacoustic Imaging.

Photoacoustic imaging using energy conversion from light to ultrasound waves has been developed as a powerful tool to investigate in vivo phenomena due to their complex characteristics. In photoacoustic imaging, endogenous chromophores such as oxygenated hemoglobin, deoxygenated hemoglobin, melanin, and lipid provide useful biomedical information at the molecular level. However, these intrinsic absorbers show strong absorbance only in visible or infrared optical windows and have limited light transmission, making them difficult to apply for clinical translation. Therefore, the development of novel exogenous contrast agents capable of increasing imaging depth while ensuring strong light absorption is required. We report here the application of carbon nanomaterials that exhibit unique physical, mechanical, and electrochemical properties as imaging probes in photoacoustic imaging. Classified into specific structures, carbon nanomaterials are synthesized with different substances according to the imaging purposes to modulate the absorption spectra and highly enhance photoacoustic signals. In addition, functional drugs can be loaded into the carbon nanomaterials composite, and effective in vivo monitoring and photothermal therapy can be performed with cell-specific targeting. Diverse applied cases suggest the high potential of carbon nanomaterial-based photoacoustic imaging in in vivo monitoring for clinical research.

Review on Optical Imaging Techniques for Multispectral Analysis of Nanomaterials.

Biomedical imaging is an essential tool for investigating biological responses in vivo. Among the several imaging techniques, optical imaging systems with multispectral analysis of nanoparticles have been widely investigated due to their ability to distinguish the substances in biological tissues in vivo. This review article focus on multispectral optical imaging techniques that can provide molecular functional information. We summarize the basic principle of the spectral unmixing technique that enables the delineation of optical chromophores. Then, we explore the principle, typical system configuration, and biomedical applications of the representative optical imaging techniques, which are fluorescence imaging, two-photon microscopy, and photoacoustic imaging. The results in the recent studies show the great potential of the multispectral analysis techniques for monitoring responses of biological systems in vivo.