Browse Articles

Predicting visual field global and local parameters from OCT measurements using explainable machine learning

Abstract Glaucoma is characterised by progressive vision loss due to retinal ganglion cell deterioration, leading to gradual visual field (VF) impairment. The standard VF test may be impractical in some cases, where optical coherence tomography (OCT) can offer predictive insights into VF for multimodal diagnoses. However, predicting VF measures from OCT data remains challenging. To address this, five regression models were developed to predict VF measures from OCT, Shapley Additive exPlanations (SHAP) analysis was performed for interpretability, and a clinical software tool called OCT to VF Predictor was developed. To evaluate the models, a total of 268 glaucomatous eyes (86 early, 72 moderate, 110 advanced) and 226 normal eyes were included. The machine learning models outperformed recent OCT-based VF prediction deep learning studies, with correlation coefficients of 0.76, 0.80 and 0.76 for mean deviation, visual field index and pattern standard deviation, respectively. Introducing the pointwise normalisation and step-size concept, a mean absolute error of 2.51 dB was obtained in pointwise sensitivity prediction, and the grayscale prediction model yielded a mean structural similarity index of 77%. The SHAP-based analysis provided critical insights into the most relevant features for glaucoma diagnosis, showing promise in assisting eye care practitioners through an explainable AI tool.

Fracture properties of dolomite and prediction of fracture toughness based on BP-ANN

Unraveling the causal association between inflammatory bowel diseases and uveitis through mendelian randomization analysis

Abstract To investigate the causal relationship between inflammatory bowel disease (IBD) and uveitis, we conducted a two-sample bidirectional Mendelian randomization (MR) analysis utilizing summary data from genome-wide association studies (GWAS). The primary statistical analysis was performed using the inverse-variance weighted (IVW) method. False discovery rate (FDR) correction was used to control for false positives in multiple testing. In addition, sensitivity analyses were carried out using the MR Egger intercept test and Cochran’s Q test. The MR analysis revealed that genetically determined IBD (OR = 1.141, 95% CI 1.080–1.205, P = 2.21 × 10−6, PFDR = 6.90 × 10−6), ulcerative colitis (UC) (OR = 1.113, 95% CI 1.032–1.201, P = 0.006, PFDR = 0.009), and Crohn’s diseases (CD) (OR = 1.073, 95% CI 1.017–1.133, P = 0.010, PFDR = 0.011) had a causal effect on uveitis. Conversely, the reverse MR analysis did not reveal significant causal link of uveitis on IBD, including its two subtypes. Furthermore, the results of the MR-Egger and weighted median methods were consistent with the IVW method. No evidence of heterogeneity or pleiotropy was detected by sensitivity analysis. Our findings confirm that IBD and its main subtypes had a causal connection with uveitis. Further research is needed to elucidate the underlying pathophysiological mechanisms driving this association.

Machine learning-based identification of co-expressed genes in prostate cancer and CRPC and construction of prognostic models

Apolipoprotein E-ε4 allele is associated with perihematomal brain edema and poor outcomes in patients with intracerebral hemorrhage

Cross-sectional study on the association between serum uric acid levels and non-alcoholic fatty liver disease in an elderly population

Predicting preterm birth using machine learning methods

Identification of IGF2BPs-related mRNA signature for predicting the overall survival of lung adenocarcinoma

Effect of Dzyaloshinskii Moriya interaction on magnetic tunnel junction based molecular spintronics devices (MTJMSD)

Impact of root distribution patterns on the elastic deformation resistance capacity and pore water development in root reinforced soil

The effectiveness of a theory‑based health education program on self-efficacy and breastfeeding behaviors continuity of working mothers in Iran

Comprehensive analysis of the global, regional, and national burden of benign prostatic hyperplasia from 1990 to 2021

Exploring the microbiomes of camel ticks to infer vector competence: insights from tissue-level symbiont-pathogen relationships

Distinct clinicopathological features and treatment differences in breast cancer patients of young age

Abstract The incidence of breast cancer in young women (aged under 40) is on the rise and is associated with more aggressive tumor characteristics and lower survival rates. Breast cancer is most frequently diagnosed in the sixth decade, and most research presents results based on data from older patients. By using large-scale clinico-pathologic and transcriptomic data from the Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) (n = 1932), we aimed to explore age-related differences in treatment, tumor characteristics, and gene expression signatures. Young patients presented more aggressive clinico-pathologic features such as higher histological grade, more frequent lymph node metastasis involvement, and estrogen receptor negativity. Accordingly, age below 40 years was associated with lower mRNA expression of the estrogen- and progesterone receptors, encoded by ESR1 and PGR, a higher proportion of the basal-like subtype, and increased transcription patterns reflecting stemness. Young breast cancer patients showed reduced survival, also within the basal-like subtype. We observed age-related differences in treatment, with more patients receiving chemotherapy among the young. Our results confirm a more challenging disease in young patients with breast cancer despite the more abundant use of chemotherapy. This argues for increased attention to young patients in current management and future research in breast cancer.

Integrated network pharmacology, molecular docking and experimental validation to investigate the mechanism of tannic acid in nasopharyngeal cancer

The effectiveness evaluation of industry education integration model for applied universities under back propagation neural network

Mutation of the gidB gene causes intrinsic streptomycin resistance in Bacillus velezensis

Magnetic field detection with single mode spin wave interference in asymmetric structure

The biodegradable polymer poly(butylene succinate-co-adipate) modulates the community structures of actively growing bacteria in rotifer culture water

Effect of neoadjuvant chemotherapy on CD14 + CD16 + monocytes and soluble CD163 in Egyptian breast cancer patients

Abstract Neoadjuvant chemotherapy (NACT) influences the anticancer response by favourably altering the immune microenvironment. However, the effects of NACT on peripheral monocytes and their prognostic contribution to the NACT response have not yet been clarified. We aimed to evaluate the potential therapeutic responses and possible predictive value of double-positive (CD14 + CD16 +) monocytes and soluble CD163 (sCD163) in Egyptian breast cancer patients. Blood samples were obtained before and after neoadjuvant therapy from 30 patients with invasive breast cancer, and the expression of CD14 and CD16 was assessed via flow cytometry. The patients’ sCD163 levels were also determined in both the serum and culture supernatant using enzyme-linked immunosorbent assay (ELISA). The results revealed that NACT was associated with a significant decrease in double-positive monocytes and sCD163 levels. In addition, both double-positive monocytes and serum sCD163 were significantly associated with a partial clinical response. Double-positive monocytes and serum sCD163 levels may be related to therapeutic response, suggesting their possible predictive value in breast cancer patients receiving NACT.