Abstract Background: Vocal impairment constitutes a prominent prodromal biomarker for Parkinson’s disease (PD). However, existing deep learning solutions often rely on computationally intensive architectures or biased validation protocols. This study proposes the lightweight spectrotemporal hybrid network (LST-Net), a resource-efficient framework for robust PD screening. Methods: The model employs a dual-stream architecture that concurrently processes raw one-dimensional waveforms and two-dimensional Mel-spectrograms to capture complementary cues, specifically transient temporal jitters and spectral energy patterns. The proposed architecture was evaluated on the Italian Parkinson’s Voice and Speech dataset using a strict subject-independent 5-fold cross-validation protocol to prevent data leakage. Results: The proposed LST-Net achieved a robust segment-level accuracy of 96.69%. By implementing a subject-level majority voting strategy, mimicking clinical diagnosis where patients are assessed over time, the model successfully classified all subjects in the held-out test sets (100% subject-level accuracy). Quantitative Gradient-weighted Class Activation Mapping analysis confirmed the model’s focus on clinically relevant low-frequency tremors (mean energy: 61.56% ±27.90%). Conclusions: This framework demonstrates that explicit spectrotemporal fusion yields superior diagnostic performance without heavy computational burdens, offering a verifiable and scalable solution for real-time deployment on edge devices.
Abstract Background: Photoplethysmography (PPG) is an optical, noninvasive method widely applied to detect changes in blood volume within the skin. Accurate modeling of light-tissue interaction is essential for improving PPG sensor design and signal interpretation. Methods: In this study, the three main layers of human skin were simulated using the nonsequential mode of Zemax optical software. Each layer was assigned specific optical properties to accurately represent physiological conditions. A light source with a wavelength of 600 nm was directed onto the modeled skin, and the propagation of light through the tissue to the photodetector was simulated. The direct current (DC) component of the PPG signal was first modeled, representing the baseline optical response. Then, the alternating current (AC) component (pulsatile component linked to cardiac blood-volume changes) was added by simulating periodic changes in blood volume within the dermal vessels during cardiac cycles. Results: The light transport from the source through the skin tissue was successfully simulated, modeling the interaction of photons within the three skin layers. Subsequently, the resulting signal reaching the photodetector was simulated, capturing both DC and AC components of the PPG waveform. Two of the model’s output PPG signals were compared with experimental data, showing good agreement in waveforms. Conclusion: Using a three-layer skin tissue model combined with the Henyey–Greenstein scattering model to calculate light’s angular distribution, along with accurate optical properties of skin layers and blood volume changes during each cardiac cycle, the PPG signal was successfully simulated. This simulation marks a significant step toward developing PPG sensors for the physiological measurement of blood components on the skin.
Abstract Background: The aim of this study is to compare tumor control probability (TCP) for the planning target volume (PTV) using Poisson, Niemierko, and Zaider-Minerbo models, and normal tissue complication probability (NTCP) for organs at risk (OARs) using Niemierko, Lyman-Kutcher-Burman (LKB) and relative seriality (RS) models. Methods: Radiotherapy dosimetry data of sixty breast cancer patients were utilized to calculate TCP and NTCP with various models. Patients underwent planning using a 16-slice CT scanner, delivering 50 Gy in 25 fractions using 6 MV photon beam energy and the Monaco treatment planning system (TPS). Results: Significant differences in TCP were found between Poisson and Zaider-Minerbo models, and between Zaider-Minerbo and equivalent uniform dose (EUD) based models, with no difference between Poisson and EUD-based models. For NTCP, differences were observed between RS model and LKB for heart, spinal cord, and thyroid, as well as between RS model and EUD-based model for the heart, spinal cord and thyroid. LKB and EUD-based models showed differences when considering the lungs as an additional OAR. Conclusion: Differences among models for calculation TCP and NTCP emphasize the importance of considering specific treatment objectives and patient characteristics in breast cancer radiotherapy planning, ensuring optimized therapeutic outcomes while minimizing adverse effects on healthy tissues.
Background: During the coronavirus disease 2019 (COVID-19) pandemic, specialists faced therapeutic challenges with various medications administered in different doses for patients. This study aimed to evaluate the cytotoxicity of some of these treatments on epithelial cells. Methods: In this experimental study, the 2,5-diphenyl-2H-tetrazolium bromide assay was utilized to assess the cytotoxicity of chloroquine (CQ) and ribavirin on HeLa cells, which were obtained from the Pasteur Institute Cell Bank, at various concentrations. The cells were then exposed to trypan blue, loaded into a hemocytometer, and examined under low-power microscopy to count the number of viable cells (unstained). The concentration of cells per milliliter was then calculated. Untreated cells were the control, and each treatment was conducted in triplicate. Results: The results indicated that ribavirin had a cytotoxic effect on the cultured cells at commonly accessible concentrations, up to a 1:32 dilution, while CQ was highly toxic, even at a dilution of 1/1024. The experiment demonstrated that, throughout the exposure period, CQ reduced the number of cells across all concentrations (0.001–1 mg/ml). Ribavirin exhibited lower toxicity than CQ. Both the medications used for COVID-19 treatment were toxic, with CQ showing a higher level of toxicity than ribavirin in epithelial cells. Conclusion: Based on the results of this study, ribavirin exhibited cytotoxic effects on the cultured cells, and CQ was found to be highly toxic. Therefore, the cytotoxicity may be observed in the oral cavity due to medication use.
Background: Microwave imaging using ultra-wideband (UWB) radar systems has emerged as a promising noninvasive technique for breast tumor detection. However, skin artifacts significantly reduce imaging accuracy and tumor localization performance. Methods: An optimized artifact removal algorithm based on autocorrelation analysis was developed for industrial wireless sensor networks (IWSNs) in microwave imaging. The transmitted signal utilized linear frequency modulation with a pulse width of 2 ns and a bandwidth of 20 MHz. Skin artifacts were identified through autocorrelation and subsequently removed from received signals while preserving tumor reflections. The proposed method was evaluated under different receiver antenna configurations and compared with Wiener filtering and singular value decomposition (SVD) methods. Results: Simulation results demonstrated that the proposed algorithm achieved lower root mean square error (RMSE), higher damping speed, improved image quality, and reduced computational complexity compared with conventional methods. The proposed approach also showed superior detection accuracy and robustness under different signal-to-noise ratios and antenna arrangements. Conclusion: The proposed autocorrelation-based artifact removal algorithm effectively suppresses skin artifacts while preserving diagnostically relevant tumor signals. The method improves microwave imaging performance and has strong potential for reliable and accurate breast cancer detection in clinical applications.
