Cheng-Shie Wuu

Background: Radiotherapy-related complications can impair long-term outcomes in nasopharyngeal carcinoma (NPC), while high-volume literature screening remains a bottleneck in evidence synthesis. Methods: Four databases were searched through January 2026. After deduplication, 4916 records underwent two-stage manual screening, yielding 38 studies. Three Bidirectional Encoder Representations from Transformers (BERT) models, BERT-base, BioBERT, and PubMedBERT, were fine-tuned using stratified five-fold cross-validation, and screening performance was assessed using the area under the receiver operating characteristic curve (ROC-AUC), the area under the precision-recall curve (PR-AUC), and work saved over sampling (WSS) at the 100% and 85% recall thresholds (WSS@100% and WSS@85%).The clinical evidence was evaluated using the Prediction Model Risk of Bias Assessment Tool (PROBAST) and synthesized narratively according to endpoint, discrimination metric, time horizon, unit of analysis, validation level, and uncertainty reporting. Results: At complete recall, BioBERT and PubMedBERT achieved mean WSS values of 97.8% and 97.5%, respectively, compared with 89.7% for BERT-base. Separately, the clinical evidence review identified 39 discrimination estimates: 33 conventional binary ROC-AUCs, three Harrell’s C-indices, and three time-dependent AUCs. Only two studies contributed external-validation estimates and one contributed a temporal-validation estimate. Sixteen estimates lacked a usable measure of uncertainty, and PROBAST rated all studies as having a high overall risk of bias, driven by the analysis domain. Conclusions: Within the retrieved and manually annotated corpus, biomedical pretrained language models showed the potential to reduce the title-and-abstract screening burden while maintaining complete recall. Independently, the clinical evidence map identified promising discrimination but insufficiently robust and transportable evidence for routine clinical implementation.

Objective: This study aims to develop a preliminary explainable image-based auxiliary assessment framework for radiation pneumonitis (RP) by integrating a generative adversarial network (GAN), Gradient-weighted Class Activation Mapping (Grad-CAM), and ensemble learning to improve image-based classification performance, interpretability, and computational efficiency. Methods: Chest CT images from 46 lung cancer patients who underwent VMAT were retrospectively collected, yielding 542 RP and 1857 non-RP images. Images were preprocessed using Otsu-based segmentation and standardized before being input into the RP-GAN for feature extraction. Grad-CAM was applied to qualitatively visualize attention patterns across convolutional layers and guide feature-layer selection, while PCA reduced the extracted features from 12,288 to 730 dimensions. An ensemble stacking classifier combining RF, SVM, KNN, and XGBoost with logistic regression as a meta-learner was constructed. Model performance was evaluated using AUC, accuracy, PPV, NPV, specificity, recall, and F1-score. Results: Grad-CAM highlighted conv2d_4 and conv2d_5 as the most informative layers. PCA reduced training time from 49 min to 49 s with minimal performance loss. In the internal hold-out test set, the ensemble model achieved the highest point estimates for AUC and accuracy among the evaluated classifiers (AUC = 0.921; accuracy = 87.1%). The model identified RP-positive CT slices and provided qualitative visual cues for suspected RP-related regions. Conclusions: The proposed GAN–Grad-CAM ensemble framework showed promising internal classification performance for post-treatment CT-based RP image assessment with substantially improved computational efficiency. Its qualitative visual outputs may support clinical image review, although further external validation and quantitative localization assessment are required before clinical application as an auxiliary image-review tool.

Background/Objectives: This study developed and evaluated a BERT-assisted literature screening workflow to support meta-analyses of postradiotherapy complications in nasopharyngeal carcinoma patients. The aim was to automate key screening steps to improve downstream screening efficiency and consistency, while minimizing time and bias during manual reviews. Materials and Methods: A bidirectional encoder representations from transformers (BERT) model was integrated into a standard systematic review pipeline for studies on postradiotherapy complications in nasopharyngeal carcinoma. The workflow combined automated BERT-based classification with manual verification and followed PRISMA and PICOS guidelines for literature identification, screening, and eligibility assessment. Model training involved hyperparameter tuning and comparison of different optimizers to maximize screening performance against a manually curated reference set, with particular attention to discrimination (AUC) and processing time. Results: From an initial corpus of 6496 records, the combined automated and manual workflow identified 23 eligible studies for meta-analysis. The included studies showed substantial heterogeneity (I² = 86.85%), supporting the use of a random-effects model to pool outcomes. The BERT model optimized with an Adagrad optimizer achieved an AUC of 0.77 for relevant-study classification and reduced screening time to 1142 s. To demonstrate the workflow’s utility, a downstream meta-analysis was conducted using the identified studies. As a downstream application based on the identified studies, a quantitative synthesis was conducted, in which (meta-analysis of the 23 included studies), a random forest model—evaluated across those studies—achieved an AUC of 0.92 under a fixed-effect analysis for predicting postradiotherapy complications. Conclusions: Integrating BERT into the literature screening phase of meta-analysis for postradiotherapy nasopharyngeal carcinoma complications markedly improved screening efficiency while maintaining acceptable classification performance. This workflow demonstrates the feasibility of transformer-based assistance for systematic reviews and provides a foundation for developing disease-specific, AI-augmented evidence synthesis pipelines in oncology.

Purpose: This study develops a predictive model for radiation pneumonitis (RP) risk in lung cancer patients after volume-modulated arc therapy (VMAT) that leverages high-dimensional dosiomics and dose–volume histogram (DVH) features within IDLSS (incremental-dose interval-based lung subregion) lung subregions. Methods: We retrospectively analyzed data from 136 lung cancer patients treated with VMAT between 2015 and 2022, including 39 patients who developed RP greater than Grade 2. Using the IDLSS method, seven regions of interest (ROIs), including the Planning Target Volume (PTV), normal lung, and five subdivided lung areas, were delineated on pretreatment Computed Tomography (CT) images. DVH, radiomics, and dosiomics features were extracted from these ROIs and organized into nine distinct feature sets. A comprehensive pipeline was applied, integrating IDLSS-defined lung subregions, high-dimensional dosiomics features, LASSO-based feature selection, and SMOTE oversampling to address class imbalance in the training data. Logistic regression, random forest, and feedforward neural networks were constructed and optimized via tenfold cross-validation. Model performance across different feature sets was evaluated via the average AUC, F1 score, and other performance metrics. Results: LASSO regression revealed that BMI and volume within the 5–10 Gy and 10–20 Gy lung subregions were significant predictors of RP. The performance evaluation demonstrated that the dosiomics features consistently outperformed the DVH features across the models. Combining radiomics and dosiomics achieved the highest predictive accuracy (AUC = 0.91, ACC = 0.89, NPV = 0.95, PPV = 0.78, F1 score = 0.82, sensitivity = 0.88, specificity = 0.90). Applying SMOTE during training significantly improved sensitivity without compromising specificity, confirming the value of balancing strategies in enhancing model performance. Incorporating all the features together did not provide additional performance gains. Conclusions: Integrating radiomics and dosiomics features extracted from IDLSS-defined lung subregions significantly enhances the ability to predict RP after VMAT, surpassing traditional DVH metrics. The substantial contribution of dosiomics features highlights the importance of spatial dose heterogeneity in RP risk assessment.

Preclinical in vivo studies have drastically improved over the past decade with the development of cone beam computed tomography (CBCT) image-guided small animal irradiation systems. Such systems produce 220 or 225 kV x-rays with square and circular field sizes ranging from 0.5 to 10 mm. The dosimetry of such equipment involving kilovoltage small-field dosimetry has not received as much attention as the megavoltage small-field dosimetry. The dosimetry of megavoltage small fields can be challenging due to lateral charged particle disequilibrium, detector volume averaging effect, and high dose gradients. Clinically there has been a rapid increase in the use of small fields in modern radiotherapy techniques such as stereotactic radiosurgery (SRS) and stereotactic radiotherapy (SBRT). This study presents dosimetric properties of image-guided small animal irradiation systems. Both EBT Gafchromic films and 3-D PRESAGE measured beam data were presented and compared with the calculated dose distribution from a commissioned planning system. For megavoltage small-field dosimetry, EBT3 films and PRESAGE dosimeters were used to measure the dose distributions for MLC-delimited 6 x 6 to 20 x 20 mm² square fields, and for selected IMRT and VMAT plans with small field sizes or segments of 1-4 cm. A single-beam optical CT scanner was used as the readout mechanism of the radiation-induced 3-D information in the PRESAGE phantoms. Measured data sets were compared with calculated results from Eclipse Acuros XB. The results for kilovoltage small animal irradiator showed that PDD data measured from EBT films and PRESAGE dosimeters are in agreement within 3%; profiles and 2-D dose distributions measured from PRESAGE present a larger penumbra compared with those from EBT films. Discrepancies between measured beam data and treatment planning data were identified. For megavoltage small fields, the measured data percent depth dose, beam profiles, and dose distributions were found to agree within experimental uncertainties for EBT films and PRESAGE dosimeters. Beam profiles for MLC-delimited field sizes less than 10 mm reveal some discrepancies in the penumbra region between measured data and calculated results. Dose distributions from EBT film and PRESAGE measurements demonstrate that 3-D dosimetry measurement for small-field IMRT and VMAT QA may be necessary to ensure a complete verification of dose delivery and accuracy.

Both polymer gels and PRESAGE radiochromic solid dosimeter, in conjunction with optical CT scanning system, have been employed to measure 3-D dose distribution. The 3-D dose maps obtained from these systems can provide a useful tool for dose verification on complex treatments such as IMRT, radiosurgery, and RapidArc. These complex treatments present high dose gradient regions in the boundaries between the target and the surrounding critical organs. Dose accuracy in these areas can be critical, and may affect the treatment. There is a pressing need for a dosimeter that allows for accurate determination of 3-D dose distribution with high spatial resolution. In this paper, performance of polymer gels and PRESAGE dosimeter with optical CT scanning is reviewed and evaluated in terms of their sensitivity calibration, irradiation, optimization of scanning procedures, precision, and accuracy. Clinical applications of optical-CT dosimetry are presented.

Both ¹²⁵I and ¹⁰³Pd sources have been widely used in the permanent prostate implant. An important consideration for the choice of brachytherapy sources is the relative biological effectiveness (RBE) for the source/seed used in the implantation. As RBE is closely related to the microdosimetric parameter, it is desirable to calculate the dose mean lineal energies for both ¹²⁵I and ¹⁰³Pd at various radial distances to the seed surface. Monte Carlo simulation was performed for photons emitted from ¹²⁵I and ¹⁰³Pd. Energy depositions from photons and all their secondary electrons were tracked. Dose distributions of lineal energy, d(y), were calculated for spheres of 1 µm in diameter and at various radial distances to the seed surface. From the dose distribution of lineal energy, the dose mean lineal energy, y_D, was derived. The results showed that the radiation qualities are constant in the distance range from 0.5 to 5 cm. In this distance range, the quality factor, relative to gamma rays from ⁶⁰Co, is 2.2 for ¹²⁵I and 2.5 for ¹⁰³Pd.

Dose distributions generated from intensity‐modulated‐radiation‐therapy (IMRT) treatment planning present high dose gradient regions in the boundaries between the target and the surrounding critical organs. Dose accuracy in these areas can be critical, and may affect the treatment. With the increasing use of IMRT in radiotherapy, there is an increased need for a dosimeter that allows for accurate determination of three‐dimensional (3D) dose distributions with high spatial resolution. In this study, polymer gel dosimetry and an optical CT scanner have been employed to implement 3D dose verification for IMRT. A plastic cylinder of diameter and height, filled with BANG®3 polymer gels (MGS Research, Inc., Madison, CT) and modified to optimal dose‐response characteristics, was used for IMRT dose verification. The cylindrical gel phantom was immersed in a water tank for an IMRT irradiation. The irradiated gel sample was then scanned with an optical CT scanner (MGS Research Inc., Madison, CT) utilizing a single laser beam and a single photodiode detector. Similar to the x‐ray CT process, filtered back‐projection was used to reconstruct the 3D dose distribution. The dose distributions measured from the gel were compared with those from the IMRT treatment planning system. For comparative dosimetry, a solid water phantom of , having the same geometry as the water tank for the gel phantom, was used for EDR2 film and ion chamber measurements. Root mean square (rms) deviations for both dose difference and distance‐to‐agreement (DTA) were used in three‐dimensional analysis of the dose distribution comparison between treatment planning calculations and the gel measurement. Comparison of planar dose distributions among gel dosimeter, film, and the treatment planning system showed that the isodose lines were in good agreement on selected planes in axial, coronal, and sagittal orientations. Absolute point‐dose verification was performed with ion chamber measurements at four different points, varying from 48% to 110% of the prescribed dose. The measured and calculated doses were found to agree to within 4.2% at all measurement points. For the comparison between the gel measurement and treatment planning calculations, rms deviations were 2%–6% for dose difference and for DTA, at 60%–110% doses levels. The results from this study show that optical CT based polymer gel dosimetry has the potential to provide a high resolution, accurate, three‐dimensional tool for IMRT dose distribution verification.