Low-dose CT (LDCT) is increasingly being adopted as a preferred method for lung cancer screening. However, the accompanying rise in image noise necessitates robust denoising strategies. Therefore, this study compared LDCT images with their denoised counterparts using objective image quality metrics and key nodule-related features.
The dataset utilized in this study was chest CT scans for lung cancer screening, sourced from the LDCT and Projection Data collection. Seven deep learning-based image denoising methods were used in this work. The denoising performance was evaluated using root-mean-square error (RMSE), peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), nodule size, CT density, and Lung-RADS classification.
For solid nodules, denoising improved SSIM from 51% to 60%–64%, reduced RMSE from 137.13 HU to 62.40–78.30 HU, and increased PSNR from 23.91 dB to 28.59–30.51 dB. It also reduced the percent difference in diameter (PDdia) from 2.05% to 1.44%–1.52%, in volume (PDvol) from 5.95% to 4.43%–4.70%, in mean HU value (PDHU) from 24.40% to 8.54%–15.33%. For subsolid nodules, denoising improved SSIM from 47% to 57%–61%, reduced RMSE from 110.87 HU to 54.62–63.96 HU, and increased PSNR from 25.78 dB to 30.53–31.61 dB. Before denoising, the PDdia, PDvol and PDHU were 15.41%, 40.16% and 10.69%, respectively, which were 7.54%–15.94%, 17.54%–29.29%, and 6.10%–8.25% after denoising. These improvements led to higher Lung-RADS categorization accuracy for solid nodules, while subsolid nodules remained more affected by noise and denoising-induced bias.
The integration of denoising techniques into LDCT workflows could potentially enhance early lung cancer detection without increasing radiation exposure. Nonetheless, validating their influence on diagnostic performance remains crucial for clinical adoption.
Deep learning has advanced breast tumor prediction research, but traditional single-modality models limit feature diversity and accuracy.
To develop and validate a multimodal deep learning approach that combines mammography and ultrasound imaging for improved breast tumor classification and enhanced clinical decision-making.
This retrospective study analyzed 663 female patients with breast lesions from 2018 to 2021, including 384 benign and 279 malignant cases. The two-stage prediction model employed improved modality-specific attention mechanisms: efficient channel attention (ECA-Net) for ultrasound and convolutional block attention module (CBAM) for mammography. The fused features were input into a stacking ensemble module with logistic regression (LR), support vector machine (SVM), random forest (RF), and Extra-Trees (ET) as base learners, and multilayer perceptron (MLP) neural network as meta-learner. Data was divided into training (464), validation (133), and test (66) sets with a 7:2:1 ratio.
The proposed multimodal prediction model—mammography ultrasound (MPM-MU) achieved superior performance with an area under the receiver operating characteristic (ROC) Curve (AUC) of 87.9 ± 0.21%, representing improvements of 13.4% and 15.6% over attention-enhanced mammography (74.5%) and ultrasound (72.3%) models, respectively. Ablation studies confirmed the effectiveness of both multimodal feature fusion and attention mechanisms in enhancing diagnostic performance.
The multimodal prediction model—mammography ultrasound (MPM-MU) with modality-specific attention mechanisms demonstrated superior performance in distinguishing between benign and malignant breast tumors compared to single-modality approaches. This approach assists radiologists in improving breast lesion classification accuracy and enhancing clinical decision-making, potentially reducing unnecessary biopsies and improving diagnostic consistency.
Head and neck cancer (HNC) patients undergoing radiotherapy (RT) may experience anatomical changes during treatment, which can compromise the validity of the initial treatment plan, necessitating replanning. However, ad hoc replanning disrupts clinical workflows and increases workload. Currently, no standardized method exists to quantify anatomical variation that necessitates replanning.
This project aimed to create geometrical metrics to describe anatomical changes in HNC patients during RT. The usefulness of these metrics was evaluated by a univariate analysis and through machine learning (ML) models to predict the need for replanning.
A cohort of 150 HNC patients treated at McGill University Health Centre was analyzed. Based on the shapes of the RT structures (body, PTV, mandible, neck, and submandibular contours), we developed 43 metrics and automatically calculated them through a Python pipeline that we called HNGeoNatomyX. Univariate analysis using linear regression was conducted to obtain the rate of change of each metric. We also obtained the relative variation of each metric between the pre-treatment and replanning-requested scans. Fraction-specific ML models (incorporating information available up to and including the specific fraction) for fractions 5, 10, and 15 were built using metrics, clinical data, and feature selection techniques. Model performance was estimated with repeated stratified 5-fold cross-validation resampling technique and the area under the curve (AUC) of the receiver operating characteristic (ROC) curve.
Univariate analysis showed that body- and neck-related metrics were most predictive of replanning need. Our best specific multivariate models for fractions 5, 10, and 15 yielded testing scores of 0.82, 0.70, and 0.79, respectively. Our models early predicted replanning for 76% of the true positives.
The created metrics have the potential to characterize and distinguish which patients will necessitate RT replanning. They show promise in guiding clinicians to evaluate RT replanning for HNC patients and streamline workflows.
This work presents the development and pilot implementation of a comprehensive remote (postal) dosimetry audit for Ir-192 High Dose Rate interstitial brachytherapy, integrating independent experimental and computational dosimetry procedures into a unified workflow.
A compact, water-equivalent phantom was designed to accommodate two plastic catheters, ten Optically Stimulated Luminescent Dosimeters (OSLDs), and two radiochromic films, allowing for independent point-approximating and 2D dose measurements. In the pilot study, a user-selected treatment plan (36 source dwell positions) was generated using a clinical Treatment Planning System (TPS), after considering the optimal dose range of the dosimeters. By analyzing the DICOM-RT files, a computational dosimetry audit test was also performed using Monte Carlo (MC) simulations, enabling independent 3D dose calculations for the same plan and phantom geometry. All dosimetry results were compared to TPS calculations (TG43 and a Model-Based Dose Calculation Algorithm, MBDCA) using the 3D Gamma Index (GI) test, dose difference maps, and dose-volume histogram comparisons, wherever applicable. All procedures were designed for a minimum clinical workload burden.
The pilot study was completed within 10 days of phantom delivery to the clinical site. If necessary, measurements were corrected by applying appropriate correction factors determined by conducting side studies. GI passing criteria were adapted to the uncertainty of each dosimetry system. Excellent agreement between MBDCA dose predictions and experimental or MC results was observed. Within the volume of interest, a systematic overestimation by TG43 relative to MC results (median difference: +2.16%) was attributed to missing scatter conditions and phantom material.
Despite the labor-intensive workflow for the auditing institution, the developed protocol is suitable for remote Ir-192 audits with acceptable uncertainties. Combining experimental and computational methods strengthens the reliability of audit outcomes. Overall results of this work highlight the advantages of an integrated dosimetry protocol for comprehensive and rigorous auditing programs.
Conventional helical computed tomography (CT) is limited by constraints related to centripetal acceleration, with current rotation speeds nearing the boundaries of engineering feasibility. This study addresses these challenges by proposing a novel CT system design
This study introduces an innovative multi-source static CT architecture that employs an array-based, fully integrated x-ray source paired with a photon stream detector. This system utilizes a complete circular configuration and implements a sequential exposure strategy for each source, thereby eliminating the need for mechanical rotation typical of traditional helical CT. Leveraging the compact integration of the x-ray source and the fine pixel structure of the detector, we developed a compressed-sensing iterative reconstruction algorithm based on the bilateral extended Feldkamp-Davis-Kress (bixFDK), referred to as “iVision” reconstruction. In addition, a software-based scatter correction algorithm was implemented. These enhancements collectively improve system performance by significantly boosting spatial resolution.
The multi-source static CT system met all key regulatory standards for performance indicators, including image noise, uniformity, measurement accuracy, spatial resolution, and density resolution. Notably, the system achieved a spatial resolution of up to 25 LP/cm@MTF = 10%, positioning it at the forefront of ultra-high-resolution CT imaging. In volunteer clinical scans, the system consistently delivered sharper images of anatomical regions such as the head, chest, abdomen, and musculoskeletal structures, outperforming conventional helical CT, particularly in detailed visualization of bones and joints, pulmonary tissues, and internal organs.
Based on the above results and analyses, this multi-source static CT system achieves diagnostic-quality imaging, aligning with clinical practice standards. It excels in capturing fine anatomical detail and detecting critical pathological features. Its high-resolution capability also supports precise diagnoses of hip fractures and facilitates detailed trabecular bone assessment, thus improving overall diagnostic accuracy.
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：