Heart rate measurement using a continuous wave Doppler radar sensor (CW-DRS) has been applied to cases where non-contact detection is required, such as the monitoring of vital signs in home healthcare. However, as a CW-DRS measures the speed of movement of the chest surface, which comprises cardiac and respiratory signals by body motion, extracting cardiac information from the superimposed signal is difficult. Therefore, it is challenging to extract cardiac information from superimposed signals. Herein, we propose a novel method based on a matched filter to solve this problem. The method comprises two processes: adaptive generation of a template via singular value decomposition of a trajectory matrix formed from the measurement signals, and reconstruction by convolution of the generated template and measurement signals. The method is validated using a dataset obtained in two different experiments, i.e., experiments involving supine and seated subject postures. Absolute errors in heart rate and standard deviation of heartbeat interval with references were calculated as $1.93\pm 1.76\mathrm{bpm}$ and $57.0\pm 28.1\mathrm{s}$ for the lying posture, and $9.72\pm 7.86\mathrm{bpm}$ and $81.3\pm 24.3\mathrm{s}$ for the sitting posture.

This research proposes a noncontact heart rate measurement method using medical radar and artificial intelligence techniques. This technique has a significant role in the design and development of a wireless system that monitors the body’s vital signs. Firstly, based on a signal model describing chest surface movement, we propose a method to create a dataset for the training process using the long-short-term memory model. Secondly, a novel method to extract chest motion from the received radar signal is proposed. Finally, the heart rate will be estimated by using the obtained model and the received motion signal. The performance of the proposed method is evaluated through the root mean square error parameter as well as compared with other methods. Experimental results evaluated according to Bland–Altman achieved an accuracy of 96.67%.

This paper proposes a system to effectively identify brain tumors on MRI images using artificial intelligence algorithms and ADAS optimization function. This system is developed with the aim of assisting doctors in diagnosing one of the most dangerous diseases for humans. The data used in the study is patient image data collected from Bach Mai Hospital, Vietnam. The proposed approach includes two main steps. First, we propose the normalization method for brain MRI images to remove unnecessary components without affecting their information content. In the next step, Deep Convolutional Neural Networks are used and then we propose to apply ADAS optimization function to build predictive models based on that normalized dataset. From there, the results will be compared to choose the most optimal method. Those results of the evaluated algorithms through the coefficient F1-score are greater than 94 and the highest value is 97.65.

Objectives: This research aims to design and develop a hierarchical embedded system that utilizes respiratory sound features for diagnosing COPD and other respiratory disorders. The system is engineered to achieve high accuracy and efficiency while minimizing energy consumption, making it suitable for deployment in mobile devices or embedded systems. Methods: Lung sounds are segmented into individual breathing cycles. Thirty-nine main respiratory features in the time and frequency domains are extracted and organized into four layers. The Random Forest algorithm and Artificial Neural Network are fine-tuned with the dataset and applied to disease classification. The first layer contains recorded information, while the second and third layers contain features extracted from fixed-length sound segments classified via Random Forest. The fourth layer utilizes Wavelet Transform to convert breathing patterns into Spectrogram images, which the Artificial Neural Network processes for disease diagnosis. The system is implemented on the Xilinx PYNQ-Ultra96-V2 FPGA development board. Results: The system achieves the highest accuracy of 98.81% for five disease classes: COPD, Healthy, URTI, Bronchitis, and Pneumonia, and saves 52.5% of energy consumption compared to CPU-GPU-based traditional methods. Conclusion: This study demonstrates the effectiveness of the proposed method in diagnosing COPD and other respiratory disorders. The hierarchical embedded system is designed with high accuracy and energy efficiency, with potential real-world applications to support clinical diagnosis.

Objective. To localize and distinguish between benign and malignant tumors on MRI. Method. This work proposes a high-performance method for brain tumor feature extraction using a combination of complex network and U-Net architecture. And then, the common machine-learning algorithms are used to discriminate between benign and malignant tumors. Experiments and Results. The dataset of brain MRI of a total of 230 brain tumor patients in which 77 high-grade glioma patients and 153 low-grade glioma patients were processed. The results of classifying benign and malignant tumors achieved an accuracy of 99.84. Conclusion. The high accuracy of experiment results demonstrates that the use of the complex network and U-Net architecture can significantly improve the accuracy of brain tumor classification. This method could potentially be useful for clinicians in aiding diagnosis and treatment planning for brain tumor patients.

Accurate distance estimation using a monocular camera remains a significant challenge in computer vision, especially in achieving a balance between computational efficiency and precision. Traditional sensor-based methods often lack robustness under varying conditions and tend to be more expensive compared to other approaches. On the other hand, modern deep learning methods can be computationally demanding and may not be suitable for real-time applications on resource-constrained devices. To address these gaps, we propose a lightweight and efficient distance estimation model that integrates the pinhole camera geometry with YOLOv8, a state-of-the-art object detection model. Our approach leverages the geometric simplicity of the pinhole model while benefiting from YOLOv8’s superior object detection capabilities. Experimental results demonstrate the effectiveness of our method, achieving a mean absolute error of 0.085 m and a mean absolute percentage error of 1.94% over distances ranging from 1.1 m to 7.5 m. Furthermore, the method achieves 98.61% accuracy within its effective range (1.1–5.5 m). These findings highlight the practical impact of our model, offering a cost-effective and computationally efficient solution for distance estimation tasks in robotics, autonomous systems, and similar real-world applications.