This paper investigates the adaptability of four state-of-the-art Artificial Intelligence (AI) models to the Australian mammographic context through transfer learning, explores the impact of image enhancement on model performance and analyses the relationship between AI outputs and histopathological features for clinical relevance and accuracy assessment. A total of 1712 screening mammograms (n=856 cancer cases and n=856 matched normal cases) were used in this study. The 856 cases with cancer lesions were annotated by two expert radiologists and the level of concordance between their annotations was used to establish image subsets. The area under the receiver operating characteristic curve (AUC) was used to evaluate the performance of Globally aware Multiple Instance Classifier (GMIC), Global-Local Activation Maps (GLAM), I&H and End2End AI models, both in the pre-trained and transfer learning modes, with and without applying the Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm. The four AI models with and without transfer learning in the high-concordance subset outperformed those in the entire dataset. Applying the CLAHE algorithm to mammograms improved the performance of the AI models. In the high-concordance subset with transfer learning and CLAHE algorithm applied, the AUC of the GMIC model was highest (0.912), followed by GLAM model (0.909), I&H (0.893) and End2End (0.875). There were significant differences (P<0.05) in the performances of the four AI models between high-concordance subset and entire dataset. The AI models demonstrated significant differences in malignancy probability concerning different tumour size categories in mammograms. The performance of AI models was affected by several factors such as concordance classification, image enhancement and transfer learning. Mammograms with strong concordance of radiologists’ annotations, applying image enhancement and transfer learning could enhance the accuracy of AI models.

We suggest progress regarding the following six physics opportunities. List all elementary particles. Describe dark matter. Explain ratios of dark matter to ordinary matter. Explain eras in the history of the universe. Link properties of objects. Interrelate physics models. We use models based on Diophantine equations.

A Twi-English parallel corpus is certainly an important resource for Machine Translation of Twi (ISO 639-3), a Low-Resource African Language (LRAL) which is mainly spoken in Ghana and Ivory Coast. Currently large-scale multi-domain Twi-English parallel corpus is still unavailable partly due to the difficulties and the arduous efforts required in its design. In this paper, we present TWIENG: a large-scale multi-domain Twi-English parallel corpus. We crawled the sentences from the web using web crawlers, translated, aligned, tokenized and compiled to create the corpus. We crawled English sentences from Ghanaian indigenous electronic news portals, Ghanaian Parliamentary Hansards, Twi Bible and crowdsourcing via google forms. The sentences were translated by professional translators and linguists, they were then aligned, tokenized and compiled. The corpus was curated using the sketch engine, a corpus manager and analysis software developed by Lexical Computing Limited. The corpus was manually evaluated by Twi professional linguists. The Corpus has 5,419 parallel sentences.

Abstract: Several publications have raised the issue that the development of diabetes is preceded by alteration of the microbiome (dysbiosis) and hence, the role of environmental factors, triggering dysbiosis, should be considered. Antibiotics are powerful agents inducing dysbiosis and the authors wanted to explore the possible relationship between the consumption of different major classes of antibiotics and the prevalence of diabetes (type-1, /T1D/, type-2 /T2D/) in thirty European countries. According to our hypothesis, if such association exists, the dominant use of certain major antibiotic classes might be reflected in the prevalence of T1D and T2D in different countries. Comparisons were performed between the prevalence of diabetes (T1D and T2D) estimated for 2019 and featured in the Diabetes Atlas with the average yearly consumption of major antibiotic classes of the previous 10 years (2010-19) extracted from the ECDC yearly reports on antibiotic consumption in Europe. Pearson correlation and variance analysis were used to estimate the possible relationship. Strong, positive (enhancer) associations were found between the prevalence of T1D and the consumption of tetracycline (J01A /p: 0.001/) and the narrow spectrum penicillin (J01CE /p: 0,006/, CF /p: 0.018/). Strong negative (inhibitor) association was observed with broad-spectrum, beta-lactamase resistant penicillin (J01CR /p: 0.003/), macrolide (J01F /p: 0.008/) and quinolone (J01M /p: 0.001/). T2D showed significant positive associations with cephalosporin (J01D /p: 0.048/) and quinolone (J01M /p: 0.025/), and a non-significant negative association was detected with broad-spectrum, beta-lactamase-sensitive penicillin (J01CA /p: 0.67/). Countries with the highest prevalence of diabetes (first 10 positions) showed concordance with the higher consumption of “enhancer” and the lower consumption of “inhibitor” antibiotics (first 10 positions) as indicated by variance analysis. Countries with high prevalence of T1D showed high consumption of tetracycline (p: 0.015), and narrow spectrum, beta-lactamase sensitive penicillin (p: 0.008), and low consumption of “inhibitor” antibiotics (broad-spectrum, beta-lactamase resistant, combination penicillin (p: 0.005), cephalosporin (p: 0.036), and quinolone (p: 0.003). Countries with a high prevalence of T2D consumed more cephalosporin (p: 0.084), quinolone (p: 0.54), and less broad-spectrum, beta-lactamase sensitive penicillin (p: 0.012) than other countries. Conclusion/Interpretation: The development of diabetes-related dysbiosis might be attached to higher consumption of specific classes of antibiotics, showing positive (enhancer) associations with the prevalence of diabetes, and the low consumption of other classes of antibiotics shoving negative (inhibitory) associations. Those groups of antibiotics are different in T1D and T2D

This study evaluated the effectiveness of an electronic system for managing drug-sensitive pulmonary tuberculosis patients in the Kyrgyz Republic. This was a cohort study using programmatic data. The study included patients registered on the paper-based system in 2019 and the electronic system between June 2021 and May 2022. Data was taken from 302 individuals from the electronic system for completeness and concordance in comparison with the paper-based system. The study showed that for most variables the completeness and concordance was 85.3%-93.0% and was lowest for non-mandatory fields such as medication side effects (26.8% vs 13.6%). No significant difference was observed in time taken from symptom onset to diagnosis and treatment initiation between the two systems. However, the electronic system had a significantly higher percentage of patients who initiated treatment on the same day of diagnosis (80.3% vs. 57.1%). The proportion with successful outcomes was similar in both groups, but the electronic system had a significantly lower proportion of patients with outcomes that were not evaluated or recorded (4.8% vs. 14.3%, p < 0.001). The study highlights the potential advantages of implementing an electronic TB register system for improving records. We advocate for similar studies in other regions in Kyrgyz Republic.