Untreated neonatal jaundice can have severe consequences. Effective screening for neonatal jaundice can prevent long-term complications in infants. Non-invasive approaches may be beneficial in settings with limited resources. This feasibility study explores a texture-based machine learning approach for early detection of neonatal jaundice. Clinical data and skin images of 200 infants were captured from four body locations using the Neonatal Jaundice Screening and Assessment Plate. Data were split into training/validating (n = 160) and blind testing (n = 40) datasets. Ninety-two features (three clinical, 89 texture-based) were extracted after image processing. Eight machine learning models were compared for bilirubin level prediction. The best performing model, Support Vector Machine (SVM), was implemented in a web-based application (AmberSNAP) and tested using blind testing dataset. SVM paired with RRelief-F feature selection achieved optimal performance for head and sternum measurements, while SVM with Univariate Regression performed best for abdomen and lower leg measurements. Blind testing demonstrated good performance in bilirubin level prediction (mean absolute error: 1.675 mg/dL; root mean square error: 2.192 mg/dL), with moderate correlation between predicted and measured values (r = 0.644, p < 0.001). These findings suggest that texture-based machine learning is a feasible approach for neonatal jaundice screening in low-resource settings.

The primary treatment for head and neck cancer is radiotherapy, which can cause complications and effects, such as the ability to speak, taste, produce saliva, and swallow. An oral stent is an immobilization tool for minimizing the dose in the tongue (or hard palate) by locking the tongue position during radiation delivery. It can improve the treatment accuracy due to less uncertainty caused by tongue position uncertainty between treatment fractions. However, commercial oral stents are not widely adopted in developing countries due to their unaffordable price. This study aimed to design the universal oral stent (UOS) to achieve high efficiency, ease to use, and low-cost fabrication based on 3D printing technology. There were five experiments to evaluate the UOS design and fabrication versus the modified cut syringe, including finite element analysis (FEA), the usability test, the micro Vickers hardness test, single beam dose analysis, and dose calculation on treatment plan simulations. The proposed UOS design and fabrication presented a high capability to apply for clinical use.

Background: Volumetric modulated arc therapy (VMAT) planning is a time-consuming process of radiation therapy. With a deep learning approach, 3D dose distribution can be predicted without the need for an actual dose calculation. This approach can accelerate the process by guiding and confirming the achievable dose distribution in order to reduce the replanning iterations while maintaining the plan quality. Methods: In this study, three dose distribution predictive models of VMAT for prostate cancer were developed, evaluated, and compared. Each model was designed with a different input data structure to train and test the model: (1) patient CT alone (PCT alone), (2) patient CT and generalized organ structure (PCTGOS), and (3) patient CT and specific organ structure (PCTSOS). The generative adversarial network (GAN) model was used as a core learning algorithm. The models were trained slice-by-slice using 46 VMAT plans for prostate cancer, and then used to predict and evaluate the dose distribution from 8 independent plans. Results: VMAT dose distribution was generated with a mean prediction time of approximately 3.5 s per patient, whereas the PCTSOS model was excluded due to a mean prediction time of approximately 17.5 s per patient. The highest average 3D gamma passing rate was 80.51 ± 5.94, while the lowest overall percentage difference of dose-volume histogram (DVH) parameters was 6.01 ± 5.44% for the prescription dose from the PCTGOS model. However, the PCTSOS model was the most reliable for the evaluation of multiple parameters. Conclusions: This dose prediction model could accelerate the iterative optimization process for the planning of VMAT treatment by guiding the planner with the desired dose distribution.

The clinical adoption of deep learning (DL) for histopathological image analysis is hindered by performance degradation caused by color variations arising from disparate staining protocols and scanning technologies. As morphological features may effectively provide the diagnostic information in hematoxylin and eosin slides, this study investigated grayscale conversion as an approach to standardize input for DL. We evaluated six grayscale algorithms against RGB across: (1) a single-center baseline, (2) a mixed multicenter training, (3) a cross-scanner generalization test, and (4) a cross-center generalization test. Furthermore, a novel attention-based grayscale conversion method (ACSRM) was investigated. It utilizes the transformer's attention mechanism to preserve critical color information through long-range pixel dependencies. In homogeneous settings, the best-performing grayscale methods achieved performance comparable to RGB (All-class F1 differences: −0.01 to 0.04 and no intersection over union differences). In mixed-center training, at least one of the grayscale algorithms outperformed RGB in every model, with 23 of 30 model combinations exhibiting statistically distinct decision behaviors (Wilcoxon signed-rank test: p < 0.05). Under the distribution-shift scenario, grayscale methods demonstrated better generalization: ACSRM (with Swin-Transformer-base (Swin-B)) outperformed the RGB baseline by 0.31 (0.14 and 0.45) on a specific class in cross-scanner tests, while demonstrating comparable performance on the remaining classes. Similarly, Luster with Swin-B improved F1-scores from 0.50 to 0.78 in cross-center evaluation. Statistical analysis confirmed significant differences in predictive behavior for these combinations on (3) and (4) (McNemar's Test: p < 0.05). Overall, ACSRM and Luster emerged as the most effective strategies for enhancing DL generalization, facilitating reliable clinical deployment.