Background: Accurate determination of individual radiation doses is critical in dosimetry, and international standards emphasize the importance of reporting measurement uncertainties. Neglecting measurement uncertainty undermines the reliability of reported dosimetric values, as emphasized by international standards.Objective: This study aimed to evaluate the measurement of uncertainties in two Thermoluminescent Dosimeters (TLDs)-100 and GR-200, in the dose range of 30–150 cGy.Material and Methods: In this experimental study, we evaluated the measurement of uncertainties in two thermoluminescent dosimeters, TLD-100 and GR-200, in the dose range of 30–150 cGy. The study investigated both random (Type A) and systematic (Type B) uncertainties. Random uncertainties include the non-uniformity of dosimeter sensitivity Element Correction Coefficient (ECC), variability in dosimeter response at zero dose, and repeatability. Systematic uncertainties include the dependence of dosimeter response on radiation energy, radiation angle, nonlinearity of dosimeter response, fading, ambient light effects, and reference calibration errors.Results: The total uncertainty for TLD-100 was calculated as 10.99% at a 68% confidence level (21.99% at 95% confidence), while for GR-200, it was 13.63% at a 68% confidence level (27.26% at 95% confidence). These values are well below the 42% threshold recommended by the International Atomic Energy Agency (IAEA) for personal dosimetry services. Conclusion: The GR-200 exhibits higher sensitivity for low-dose radiation detection; however, its greater uncertainty makes TLD-100 more suitable for clinical and personal dosimetry applications, where precision and reliability are paramount.
Background: Infant brain tissue segmentation is critical for understanding early brain development and detecting neurodevelopmental abnormalities. However, accurate segmentation remains challenging due to partial-volume effects, imaging noise, and low contrast between tissues in isointense regions.Objective: This study aimed to automatically segment infant brain Magnetic Resonance Imaging (MRI) images into white matter, gray matter, and cerebrospinal fluid.Material and Methods: In this analytical study, we addressed the challenges of infant brain tissue segmentation. We proposed a deep, lightweight dense backbone architecture incorporating transition blocks. The backbone was enhanced with a spatial–channel attention mechanism to capture global dependencies and emphasize task-relevant features, as well as multi-resolution deep supervision to mitigate vanishing gradients and reduce overfitting. Moreover, we used class weighting strategy to deal with class imbalance and low contrast between brain tissues.Results: Our method attains Dice similarity scores of 91.80%, 92.07% and 94.58% for white matter, gray matter, and cerebrospinal fluid, respectively. Quantitative and qualitative evaluations show that the proposed approach accurately segments infant brain MR images and reduces errors in isointense regions. Conclusion: The results demonstrate that the proposed method precisely delineates infant brain tissues and effectively overcomes common segmentation challenges.
Beyond Physical Shielding: The Case for Biological Protection in Deep Space Beyond Physical Shielding: The Case for Biological Protection in Deep Space Beyond Physical Shielding: The Case for Biological Protection in Deep Space Beyond Physical Shielding: The Case for Biological Protection in Deep Space Beyond Physical Shielding: The Case for Biological Protection in Deep Space Beyond Physical Shielding: The Case for Biological Protection in Deep Space Beyond Physical Shielding: The Case for Biological Protection in Deep Space Beyond Physical Shielding: The Case for Biological Protection in Deep Space Beyond Physical Shielding: The Case for Biological Protection in Deep Space Beyond Physical Shielding: The Case for Biological Protection in Deep Space
Background: Breast Cancer (BC) remains a leading cause of cancer-related deaths among women. Diagnosis from histopathology images can be inconsistent in pathologist assessments, highlighting the need for reliable computer-aided diagnostic systems. Existing approaches overlook multi-layer features or employ extraction methods that lack generalizability.Objective: This study proposes a layer-wise feature integration framework to enhance BC histopathology classification.Material and Methods: In this analytical and computational study, using the BreakHis and ICIAR/BACH datasets, features are extracted from intermediate and high-level layers of pre-trained Convolutional Neural Networks (CNNs). Mutual Information (MI) between each layer’s features and true labels is computed to identify the most informative layers, and those with the highest MI are selected. To address overfitting, Principal Component Analysis (PCA) with optimal Cumulative Explained Variance (CEV) is applied. The transformed features are then concatenated and classified using a Support Vector Machine (SVM).Results: The method is evaluated using VGG19, DenseNet, and ConvNextTiny across both datasets and demonstrates significant improvements in image-wise and patient-wise classification. ConvNextTiny, enhanced with Synthetic Minority Oversampling Technique (SMOTE) for class-imbalance mitigation, achieves the highest accuracy of 99.14% at 40× in BreakHis dataset. Conclusion: Although the achieved accuracies of 97.32% and 99.14% (without and with imbalance handling, respectively), are slightly below the best-reported results, the method offers notable advantages in simplicity and computational efficiency. It provides a lightweight alternative to ensemble-based or fine-tuned deep models. Results highlight the critical role of intermediate layers in capturing discriminative features and challenge reliance on final-layer outputs for histopathology image classification.
Background: Neurodegenerative Diseases (NDDs) refer to disorders, leading to the gradual degeneration of neurons in the human nervous system and profoundly affecting individuals’ motor and cognitive functions. Early diagnosis and monitoring of the progression of NDDs are of critical importance.Objective: The primary aim of this study is to present an automatic diagnostic method using time-domain analysis of short-term gait signals of two consecutive steps, for the classification of Parkinson’s Disease (PD), Huntington’s Disease (HD), Amyotrophic Lateral Sclerosis (ALS), and a Healthy Control Group (HC).Material and Methods: This analytical study used datasets from PhysioNet, including 64 five-minute recordings from 13 individuals with ALS, 15 with Parkinson’s, 20 with Huntington’s, and 16 healthy individuals. To reduce noise and enhance signal quality, a median filter was applied. The preprocessed gait signals were then systematically segmented into consecutive two-step windows, which were used to construct the feature vectors. Finally, the constructed feature vectors were used as inputs for both linear and nonlinear classifiers. The proposed method was evaluated by calculating the accuracy index.Results: The quaternary classification of ALS vs. PD vs. HD vs. HC using the bag-tree classifier achieved an accuracy of 82.8%. Furthermore, for the binary classifications of HC vs. HD, HC vs. ALS, HC vs. PD and NDD vs. HC, the accuracies were 88.4%, 97.6%, 90% and 89.1%, respectively. Conclusion: Gait dynamics analysis has a high potential for the rapid diagnosis of neurodegenerative diseases. Short term analysis in two consecutive stages enables the use of the Kinect sensor for gait analysis.
