Journal of Biomedical Physics and Engineering最新文献

Background: Multifunctional nanosystems, containing medical imaging components and cancer therapeutic drugs, can provide early cancer diagnosis and treatment.

Objective: The aim of this study was to investigate Magnetic Resonance Imaging (MRI), anticancer drug delivery, and fluorescence properties of curcumin-loaded PEGylated magnetite@graphene quantum dots nanocomposite.

Material and methods: In this experimental study, PEGylated magnetite@graphene quantum dots (Fe₃O₄@GQDs-PEG) nanocomposite was synthesized and loaded with curcumin (CUR-Fe₃O₄@GQDs-PEG). Then, the size, shape, magnetic property, MRI r2 relaxivity, drug loading and in vitro release, and fluorescence property of the nanocomposite were investigated. Evaluation of the cell toxicity against MCF-7 cells was performed for both unloaded and curcumin-loaded nanocomposites.

Results: The superparamagnetic nanocomposite showed high r2 relaxivity, drug release, and fluorescence property. The curcumin-loaded nanocomposite was significantly toxic to the breast cancer cell line at high concentrations.

Conclusion: CUR-Fe₃O₄@GQDs-PEG nanocomposite can be considered an anticancer drug carrier and an appropriate potential candidate for dual modal MRI and fluorescence imaging.

Background: Breast cancer, the most common cancer among women, necessitates early detection. Despite advances in Computer-Aided Diagnosis (CAD), lesion detection in mammograms remains challenging. Artificial Intelligence (AI) in radiology offers significant potential to enhance diagnostic accuracy in medical imaging.

Objective: This study compares object detection methods to identify the most effective model for smart diagnostic systems. This comprehensive study is the first to apply the advanced You Only Look Once version 12 (YOLO-v12) architecture for the automated detection and localization of lesions in mammographic images and to identify their malignancy or benignity status with high precision.

Material and methods: This comparative experimental study, utilizing retrospective data, also evaluated two state-of-the-art models, the Detection Transformer (DETR) and RetinaNet, for their performance. The models were trained and tested on the publicly available Categorized Digital Database for Low-Energy and Subtracted Contrast-Enhanced Spectral Mammography (CDD-CESM), which contains 1,982 mammograms with 3,720 annotated lesions of various types and sizes.

Results: YOLO-v12 demonstrated excellent diagnostic accuracy (mean Average Precision at an IOU threshold of 0.5 (mAP50)=0.98; Intersection Over Union (IOU)=0.95), significantly outperforming contemporary models and older YOLO versions.

Conclusion: The promising and robust results clearly underscore the remarkable potential of artificial intelligence technologies in effectively assisting radiologists with the early detection and diagnosis of breast cancer. These findings advocate for the implementation of YOLO-v12 in clinical mammography screening applications and suggest that future research should prioritize real-time diagnostic systems to further enhance breast cancer detection capabilities.

Background: Manual delineation of volumes for prostate radiotherapy treatment is a time-consuming task for radiation oncologists and is also prone to variability. Deep learning-based auto-segmentation methods showed promising results with accurate and high-fidelity contours.

Objective: The objective of this study was to evaluate the feasibility of a Computed Tomography (CT)-based deep learning auto-segmentation algorithm for multi-organ delineation in prostate radiotherapy.

Material and methods: In this single-institution retrospective study, a total of 118 patients with prostate cancer were included. We applied 3D nnU-net deep convolutional neural network architecture, a self-adapting ensemble method for simultaneous fast and reproducible multi-organ auto-contouring. The dataset was randomly divided into training and test sets from 95 and 23 patients, respectively. Intensity-modulated radiotherapy plans were generated for both manual and automatic delineations using identical optimization settings. Contours were assessed in terms of the Dice Similarity Coefficient (DSC), and average Hausdorff Distance (HD). Dose distributions were additionally evaluated using parameters derived from Dose-Volume Histograms (DVH).

Results: On the test set, 3D nnU-net achieved the best performance in the bladder (DSC:0.97, HD:4.13), right femur head (DSC:0.96, HD:3.58), left femur head (DSC:0.96, HD:3.95), rectum (DSC:0.9, HD:10.04), prostate (DSC:0.82, HD:3.68), lymph nodes (DSC:0.77, HD:15.5), and seminal vesicles (DSC:0.69, HD:10.95). DVH parameters of targets and Organ at Risks (OARs) were significantly different except for lymph nodes and femoral heads between treatment plans based on manual and automatic contours.

Conclusion: The 3D nnU-net architecture can be successfully used for multi-organ segmentation in the male pelvic area.

Background: Designing shields for gamma radiation sources is particularly important due to their extensive use in medical, industrial, and research studies.

Objective: This study aimed to explore the ability of an Artificial Neural Network (ANN) to identify the optimized shield for a typical gamma source. Despite the effectiveness of Monte Carlo simulations in determining optimal shielding materials and geometries, they are time-consuming and require numerous simulations for each configuration.

Material and methods: In this simulating study, the MCNPX Monte Carlo code was utilized to conduct simulations using a previously proposed shielding material. After validating the simulation accuracy, a large dataset was generated to serve as input and target data for the machine learning process. The method's precision was assessed by comparing the results of the ANN with those of Monte Carlo simulations. Dose calculations were performed using a water phantom.

Results: The deviation of less than 1% was computed between the simulation and the ANN. The network also exhibited satisfactory predictions for unknown data. Additionally, the dose was evaluated using a water phantom to assess further and optimize the selected shielding material.

Conclusion: The ANNs are widespread and significant in radiation shielding studies. The developed network can accurately predict unknown weight fraction combinations. The designed network can effectively predict unknown weight fraction combinations.

Background: Patient-Specific Quality Assurance (PSQA) is essential in radiotherapy to ensure accurate treatment delivery, particularly with advanced treatment planning techniques like RapidArc (RA).

Objective: The present study aimed to assess the use of Statistical Process Control (SPC) to evaluate Tolerance Limits (TL) and Action Limits (AL) in PSQA for various gamma criteria across different anatomical sites.

Material and methods: In this analytical study, RA treatment verification plans for brain (25), head and neck (50), thorax (25), and pelvis (50) were analyzed using an EPID to establish the Lower Control Limit (LCL). Gamma criteria (3%/3 mm, 3%/2 mm, 2%/3 mm, and 2%/2 mm) were evaluated, with the first ten samples used to calculate Individual Moving Range (I-MR) charts for TL and AL. Exponentially Weighted Moving Average (EWMA) and Cumulative Sum (CUSUM) charts were employed to detect control drifts.

Results: For the head and neck site, LCLs from I-MR charts for global gamma were from 96.82 (3%/3 mm) to 89.42 (2%/2 mm), and for local gamma, from 91.40 (3%/3 mm) to 83.06 (2%/2 mm). The brain site showed similar agreement, while the pelvis and thorax sites had LCLs of 94.84 and 94.73 for 3%/3 mm. EWMA and CUSUM charts revealed that most control charts stayed within TL, except for the stringent 2%/2 mm criterion. AL for 3%/3 mm were 96.35, 92.85, 95.77, and 92.34 for head and neck, pelvis, brain, and thorax, respectively.

Conclusion: I-MR, EWMA, and CUSUM charts are effective for establishing and monitoring TL and AL for RA-based PSQA, with site-specific limits required based on gamma criteria and measurement device.

In the wake of the COVID-19 pandemic, maintaining hygienic dental practice environments has become imperative due to the risk of SARS-CoV-2 exposure to dentists and patients. A novel infection control system has been developed, combining far-UVC light with an infrared-based hot air blower, aimed at significantly reducing pathogen spread in dental settings. This groundbreaking system leverages the germicidal properties of 222 nm far-UVC light known for inactivating viruses on surfaces and within aerosols, while circumventing the limitations posed by droplet size and composition through the strategic use of a hot air chamber. The invention includes an integrated built-in air suction device at the dental chair's headrest for removing contaminated air, a circular floor-level suction system for enhanced air circulation, and a state-of-the-art 222 nm far-UVC lamp within established safety parameters. The hot air chamber's primary function is to decrease droplet size via evaporation, thus augmenting 222 nm far-UVC light penetration to effectively neutralize pathogens. A supporting blower system evenly distributes the hot air for consistent droplet exposure to far-UVC light, while a HEPA-based air purifier re-circulates purified air back into the clinic. This integrated system not only aims to provide a safer environment by minimizing airborne transmission of viruses but also stands as a vital evolution in infection control within the dental industry. Its implementation in dental practices could revolutionize standards of care and patient safety in the ongoing global effort to mitigate infectious healthcare risks.

Background: Driver fatigue detection is crucial for traffic safety. Electroencephalography (EEG) signals, which directly reflect the human mental state, provide a reliable approach for identifying fatigue.

Objective: This study aimed to investigate the effectiveness of EEG microstate analysis in detecting driver fatigue by analyzing variations in microstate features between normal and fatigued states.

Material and methods: This analytical study aimed to develop a supervised machine learning approach for driver fatigue detection using EEG microstate features. EEG data were collected from 10 individuals in both normal and fatigued states. Microstate analysis was performed to extract key features, including duration, occurrence, coverage, and Microstate Mean Power (MMP), from four types of microstates labeled A, B, C, and D. These features were then used as inputs to train and test a Support Vector Machine (SVM) for classifying each EEG segment into either normal state or fatigue state.

Results: The classification achieved high accuracy, particularly when combining MMP and occurrence features. The highest accuracy recorded was 98.77%.

Conclusion: EEG microstate analysis, in combination with SVM, proves to be an effective method for detecting driver fatigue. This approach can be utilized for real-time driver monitoring and fatigue alert systems, enhancing road safety.

Background: Colorectal Cancer (CRC) remains a major health challenge, with perturbation in the cell cycle playing a crucial role in its progression. Hsa-miR-16-5p (miR-16-5p) is a crucial tumor suppressor, but its precise role in modulating cell cycle genes, particularly in the presence of ionizing radiation, is not fully understood.

Objective: This study investigated the role of miR-16-5p in modulating the expression of cell cycle genes in colorectal cancer cells exposed to ionizing radiation.

Material and methods: In this experimental study, HT-29 cells were transfected with miR-16-5p using the polyfectamine transfection reagent. Expression levels of miR-16-5p, Cyclin-Dependent Kinase 4 (CDK4), Cyclin E1 (CCNE1), and Cyclin D1 (CCND1) were quantified by real-time Polymerase Chain Reaction (PCR). To assess the changes after irradiation, cells were exposed to 4 Gy.

Results: Ionizing radiation significantly downregulated miR-16-5p compared to controls, while transfection of miR-16-5p significantly increased its expression level. However, irradiation of 4 Gy did not significantly alter CCND1 or CCNE1, but decreased CDK4 expression. The miR-16-5p transfection significantly suppressed CCND1, CCNE1 and CDK4 compared to controls. The expression of CCND1, CCNE1, and CDK4 significantly decreased when miR-16-5p transfection was performed before 4 Gy irradiation compared to both 4 Gy irradiation alone and the control group.

Conclusion: Our results highlight the role of miR-16-5p in modulating key cell cycle genes in CRC. Increasing miR-16-5p expression could improve radiosensitivity and represent a therapeutic strategy for the treatment of CRC.

Background: The use of Hematoxylin-and-Eosin (H&E) staining is widely accepted as the most reliable method for diagnosing pathological tissues. However, the conventional H&E staining process for tissue sections is time-consuming and requires significant labor. In contrast, Confocal Microscopy (CM) enables quick and high-resolution imaging with minimal tissue preparation by fluorescence detection. However, it seems harder to interpret images from CM than H&E-stained images.

Objective: This study aimed to modify an unsupervised deep-learning model to generate H&E-like images from CM images.

Material and methods: This analytical study evaluated the efficacy of CM and virtual H&E staining for skin tumor sections related to Basal Cell Carcinoma (BCC). The acridine orange staining, combined with virtual staining techniques, was used to simulate H&E dyes; accordingly, an unsupervised CycleGAN framework, trained to virtually stain CM images was implemented. The training process incorporated adversarial and cycle consistency losses to ensure a precise mapping between CM and H&E images without compromising image content. The quality of the generated images was assessed by comparing them to the original images.

Results: The CM images, specifically focusing on subtyping BCC and evaluating skin tissue characteristics, were qualitatively assessed. The H&E-like images generated from CM using the CycleGAN model exhibited both qualitative and quantitative similarities to real H&E images.

Conclusion: The integration of CM with deep learning-based virtual staining provides advantages for diagnostic applications by streamlining laboratory staining procedures.