ABSTRACT
Purpose: Fast and automated plan generation is desirable in radiation therapy (RT), in particular, for MR-guided online adaptive RT (MRgOART) or real-time (intrafractional) adaptive RT (MRgRART), to reduce replanning time. The purpose of this study is to investigate the feasibility of using deep learning to quickly predict deliverable adaptive plans based on a target dose distribution for MRgOART/MRgRART. Methods: A conditional generative adversarial network (cGAN) was trained to predict the MLC leaf sequence corresponding to a target dose distribution based on reference plan created prior to MRgOART using a 1.5T MR-Linac. The training dataset included 50 ground truth dose distributions and corresponding beam parameters (aperture shapes and weights) created during MRgOART for 10 pancreatic cancer patients (each with five fractions). The model input was the dose distribution from each individual beam and the output was the predicted corresponding field segments with specific shape and weight. Patient-based leave-one-out-cross-validation was employed and for each model trained, four (44 training beams) out of five fractionated plans of the left-out patient were set aside for testing purposes. We deliberately kept a single fractionated plan in the training dataset so that the model could learn to replan the patient based on a prior plan. The model performance was evaluated by calculating the gamma passing rate of the ground truth dose vs. the dose from the predicted adaptive plan and calculating max and mean dose metrics. Results: The average gamma passing rate (95%, 3mm/3%) among 10 test cases was 88%. In general, we observed 95% of the prescription dose to PTV achieved with an average 7.6% increase of max and mean dose, respectively, to OARs for predicted replans. Complete adaptive plans were predicted in ≤20 s using a GTX 1660TI GPU. Conclusion: We have proposed and demonstrated a deep learning method to generate adaptive plans automatically and rapidly for MRgOART. With further developments using large datasets and the inclusion of patient contours, the method may be implemented to accelerate MRgOART process or even to facilitate MRgRART.
ABSTRACT
BACKGROUND AND AIMS: The prevalence of non-alcoholic fatty liver disease (NAFLD) is increasing in Saudi Arabia (SA), but descriptions of the clinical and metabolic characteristics of these patients are limited. The present study aims to fill this gap. METHODS: Demographic, clinical, and laboratory data of all NAFLD patients from 2009 to 2019 were retrieved from the Systematic Observatory Liver Disease Registry (SOLID) [n=832 (337 males; 495 females); mean (± standard deviation, SD) age was 42.6±13.6 years; mean body mass index (BMI) was 35.0±9.3kg/m2]. Non-invasive surrogate scores of fibrosis (eg AST to Platelet Ratio Index (APRI), Fibrosis-4 (FIB-4), and NAFLD fibrosis (NFS) scores) were calculated and analyzed. In addition, data from NAFLD patients with normal and high alanine aminotransferase (ALT) were compared using two different methods: the standard laboratory reference range which defines normal as ALT<61 IU/L, and the range proposed by a recent national study which sets upper limits of normal ALT at 33 IU/l for men and 22 IU/l for women. RESULTS: Hyperlipidemia was the most common comorbidity (41.7%), followed by type 2 diabetes mellitus (T2DM) (35.3%) and hypertension (28.4%). Prevalence of advanced fibrosis varied widely across definitions [FIB-4, N=19 (2.5%); APRI, N=21 (2.8%); NFS, N=62 (8.6%)] and exhibited sexual dimorphism with males having worse metabolic characteristics. NAFLD patients with normal ALT were more likely to be older, female, have a lower BMI, and have a higher prevalence of cirrhosis, DM, hypertension, hyperlipidemia, and renal dysfunction. CONCLUSION: Patients with NAFLD have metabolic characteristics associated with several comorbidities, including NAFLD patients with normal ALT. Mechanistic studies are needed to examine and analyze complex, interactive effects between sex, age, and other factors that may accelerate NAFLD disease progression.
ABSTRACT
BACKGROUND AND OBJECTIVES: Hepatitis C virus (HCV) genotype (G) knowledge is essential for determining type, duration and rate of response to antiviral therapy, possible route of HCV transmission, and future vaccine development. Our aim was to study HCV genotypes and to provide precise data on genotype distribution in both genders and different age groups amongst Saudi patients. DESIGN AND SETTING: Genotype data from molecular laboratories at four different tertiary care hospitals in Riyadh from January 2006 until December 2010 were collected and analyzed. PATIENTS AND METHODS: Consecutive data on genotype, sex and age was collected from 1013 Saudi patients. Genotyping was done by selective hybridization of amplicons to HCV genotype-specific oligonucleotides. RESULTS: We found G1 in 262 patients (25.9%), G2 in 44 (4.4 %), G3 in 29 (2.9 %), G4 in 608 (60%), and 3 patients (0.3%) each of G5 and G6. In addition, 64 (6.3%) patients had mixed genotypes, mostly G4 and G1. On subtyping in 191 G1 patients, 67 (35.1%) were G1a, and 124 (64.9 %) G1b. Age distribution showed that 18 (1.7%) were 0-20 years, 173 (17.1 %) 21-40 years, 521 (51.4%) 41-60 years and 301(29.7%) > 60 years. There was no significant difference in frequency of G1, G3 and G4 among the two genders. CONCLUSION: G1 and G4 are the predominant genotypes in Saudi patients infected with HCV (85.9%), with a similar distribution among the two sexes and no significant changes in genotype distribution over the past decade.