Abstract:

Abstract:Flexible endoscopic robots, with their continuum structural characteristics, demonstrate unique advantages in minimally invasive surgery. However, the inherent nonlinear deformation features of continuum structures pose significant challenges to motion control precision. To address this technical bottleneck, this paper proposes an optimal teleoperation control method for flexible endoscopic robots based on neurodynamic optimization. First, a master-slave motion mapping mechanism in image space is established, coupled with a kinematic model of the flexible endoscope, to achieve accurate mapping between image feature velocities and driving velocities. Second, joint motion constraints are incorporated to formulate the robot control as a quadratic programming based optimal control problem, which is efficiently solved using a neurodynamic-based real-time solver. Experimental validation is conducted on a ureteroscopic robotic platform. Results demonstrate that the proposed method effectively suppresses manual operation errors and velocity oscillations, maintaining target tracking errors within 2.5% while significantly enhancing the accuracy and stability of instrument manipulation during lithotripsy procedures.

Abstract:Cervical cancer is one of the leading causes of cancer-related death among women globally, and radiotherapy is a common treatment method for cervical cancer. Among the various radiation therapies, brachytherapy, which involves placing the radiation source directly into an area close to the tumor, can deliver a high dose of radiation directly to the tumor, making it more applicable compared to other radiation methods. Accurate segmentation of organs-at-risk is crucial for accurately estimating radiotherapy doses and maximizing protection of normal tissues from radiation damage. However, automatic segmentation of tubular structures, such as the colon and rectum, remains challenging. Factors such as intestinal folds and motion artifacts can affect the segmentation performance, and the presence of the radiation source in brachytherapy can degrade CT image quality, further impacting the segmentation results. This paper proposes a method for the segmentation of tubular organ-at-risk in cervical cancer based on centerline and distance map information. By enhancing the network’s understanding of anatomical structures, the method improves the identification of the topological structure of tubular organs and their spatial relationships within the human body, thus improving segmentation accuracy and optimizing radiotherapy dose distribution. Through experimental evaluation on a cervical cancer brachytherapy dataset, performance analysis was conducted using metrics such as Dice similarity coefficient (DSC), intersection over union (IoU), Recall, and 95% Hausdorff distance (HD95). The experimental results show that the proposed method outperforms the baseline network ResUNet in most metrics, specifically with a DSC of 71.58%, an IoU of 52.12%, a Recall of 79.03%, which improve by 11.29%, 7.84% and 12.70%, respectively, compared to ResUNet. The HD95 is 10.06, which is a decrease of 1.76 compared to ResUNet. The results indicate that the proposed method effectively improves the segmentation accuracy of the colon and rectum in cervical cancer brachytherapy CT images, reducing the impact of complex organs and image quality on the segmentation results.

Abstract:Multiple-instance learning (MIL), as a weakly supervised learning method, has been widely applied in the field of medical image analysis in recent years. The paper reviews the progress of MIL applications in whole slide images, with a detailed analysis of its roles in tumor detection, subtype classification, and survival prediction. MIL holds unique advantages in weakly supervised learning , which can be optimized and extended through the introduction of new mechanisms to adapt to a broader range of application scenarios. The paper first reviews some widely used or uniquely advantageous MIL models, elaborating on their technical features and specific application contexts. Secondly, it introduces the application and technology advancements of MIL in multimodal medical image analysis. Finally, the current research progress of MIL is summarized, and its future development prospects are explored.

Abstract:Non-contact vital sign monitoring using millimeter-wave radar offers continuous and discreet identification. Cardiac motion is influenced by various complex factors, making it challenging to capture characteristic waveform information. To address this, the study employs millimeter-wave radar transmitting frequency modulated continuous waves to monitor and record cardiac data during sleep. Additionally, the paper proposes a deep convolutional neural network (CNN)-based identity recognition method using one-dimensional time-series radar signals of cardiac motion. The performance of this method is compared with three deep learning algorithms: long short-term memory Network, InceptionTime, and LSTformer. The final classification accuracies of all models exceed 85% on a dataset of heart signals collected in a resting state in the laboratory. Among the models, InceptionTime achieves the highest accuracy but requires the longest processing time. The long short-term memory and LSTformer models exhibit lower accuracy but faster processing. The CNN model proposed in this study demonstrates comparable accuracy to InceptionTime, while requiring less computational time, thus balancing accuracy and efficiency.

Abstract:Image denoising methods based on deep learning have effectively solved the problems of cumbersome parameter tuning and complex noise modeling in traditional denoising methods. However, the model training of supervised learning relies heavily on pairs of clean and noisy images, which limits the wide application of such models. Unsupervised learning denoising models only require single noisy images for training, but the existing unsupervised denoising methods still have the problem that it is difficult to balance network training efficiency and denoising performance. This paper proposes an efficient image denoising method, which improves the efficiency of denoising model training. Specifically, this method proposes a deep neighbor downsampler, which is used to obtain similar image pairs for training the noise model from the same noisy image. The research proposed sampler method not only meets the requirements that the pixels of the image pairs are adjacent and the appearances are similar, but also the deep neighbor downsampling discards some redundant information and avoids heavy dependence on assumptions about the noise distribution. Finally, the research verify the effectiveness of the research method through synthetic experiments with various noise distributions in the standard red green blue space and real image experiments. The experimental results confirm that the sampling strategy the research proposed effectively overcomes the balance problem between training efficiency and denoising performance.

Abstract:As a vital part of traditional Chinese medicine, acupuncture has extensive application value all over the world. However, the reliance on practitioners’ experience for acupoint localization in traditional acupuncture methods leads to a lack of standardization, restricting its reproducibility and broader adoption. Acupuncture robots, as a kind of intelligent medical devices, offer new opportunities for standardizing and promoting acupuncture techniques. This paper introduces an improved YOLOv8-Pose model, YOLO-PointMap, designed to address challenges in dense acupoint distribution and weak feature recognition. By incorporating dynamic convolution to optimize the C2f module and introducing a channel-attention-based feature fusion module, the model achieves significant advancements in multi-scale feature extraction and integration. Experimental results show that the end point error (EPE), percentage of correct keypoints (PCK) and mAP50-95(Pose) indexes of YOLO-PointMap on the test set are superior to the existing models, with the values reaching 3.27, 1.00 and 84.90% respectively, especially in dense key point identification and weak feature region localization. It provides strong support for the development of acupuncture robot technology, and shows the potential application value in the fields of virtual reality and intelligent interaction.

Abstract:For cone beam computed tomography (CBCT), there has long been a desire to modulate the intensity and distribution of the X-rays to accommodate the patient’s anatomy as the gantry rotates from one projection to another. This would reduce both image artifacts and radiation dose. However, the current beam modulation setups, such as dynamic bowtie filters, may be too complex for practical use in clinical applications. This study aimed to investigate a simplified dynamic beam filtration strategy for CBCT imaging to reduce image artifacts and radiation dose. In this study, the beam filtration was designed to vary dynamically as the CBCT gantry rotates around the object. Specifically, two distinct components were integrated: the sheet filter part and the bowtie filter part. The dynamic beam filtration setup has two working schemes, one is a combination of dynamic sheet filter and dynamic bowtie filter, denoted as dynamic filter-dynamic bowtie (DFDB); the other is a combination of dynamic sheet filter and static bowtie filter, denoted as dynamic filter-static bowtie (DFSB). Numerical imaging experiments were performed for three human body parts: the shoulder, chest, and knee. In addition, the Monte Carlo simulation platform MC-GPU was used to generate the dose distribution maps. Results showed that the proposed DFDB and DFSB beam filtration schemes can significantly reduce the image artifacts and thus improve the CBCT image quality. Depending on the scanned object, the total radiation dose could be reduced by 30%. The proposed simple dynamic beam filtration strategy, especially the DFSB approach, could be beneficial in the future to improve the CBCT image quality with reduced image artifacts and radiation dose.

Abstract:The ability to replicate the microenvironment of the human body through the fabrication of scaffolds is a significant achievement in the biomedical field. However, the search for the ideal scaffold is still in its infancy and there are significant challenges to overcome. In the modern era, the scientific community is increasingly turned to natural substances due to their superior biological ability, lower cost, biodegradability, and lower toxicity than synthetic lab-made products. Chitosan is a well-known polysaccharide that has recently garnered a high amount of attention for its biological activities, especially in 3D bone tissue engineering. Chitosan closely matches the native tissues and thus stands out as a popular candidate for bioprinting. This review focuses on the potential of chitosan-based scaffolds for advancements and the drawbacks in bone treatment. Chitosan-based nanocomposites have exhibited strong mechanical strength, water-trapping ability, cellular interaction, and biodegradability. Chitosan derivatives have also encouraged and provided different routes for treatment and enhanced biological activities. 3D tailored bioprinting has opened new doors for designing and manufacturing scaffolds with biological, mechanical, and topographical properties.

Abstract:Ketamine, an N-methyl-D-aspartate receptor (NMDAR) antagonist, is clinically utilized for sedation, anesthesia, and the treatment of refractory depression. However, its addictive properties restrict its broader clinical application. A dose of 0.5 mg/kg is commonly used as an antidepressant in clinical settings, while 15 mg/kg represents the dose typically associated with abuse. The effects of varying doses of ketamine on brain network activation remain unclear. In this experiment, two representative doses of ketamine, 0.5 mg/kg and 15 mg/kg, were administered via intraperitoneal injection for 7 consecutive days. Brain network activation was assessed by examining the expression of the immediate early gene protein (cFos). The results indicated that, compared to the saline control group, 0.5 mg/kg ketamine significantly increased the number of cFos-positive cells in the medial prefrontal cortex, intermediate lateral septal nucleus, and periaqueductal gray matter. Conversely, 15 mg/kg ketamine significantly increased cFos expression in the nucleus accumbens, lateral habenula, hippocampal CA3 region, amygdala, and ventral tegmental area. These findings suggest that ketamine’s activation of brain networks is dose-dependent, with different doses activating distinct brain regions. This study provides a foundation for investigating the neuropharmacological effects of different ketamine doses and investigating brain regions associated with its antidepressant and addictive properties.