Carbon nanomaterials have gained the attention of researchers in the biomedical field due to their versatile characteristics, such as unique structural, physical, and chemical properties. Their tunable structures with respect to the synthesis process have led to their applications in various fields of biomedicine. We critically analysed the sustainable synthesis approaches, including green chemistry routes, thermal carbonisation, microwave-assisted methods, hydrothermal processes, pyrolytic techniques, and emerging computationally guided methods. This review encompasses diverse carbon sources, including biomass waste, synthetic polymers, coal derivatives, graphite, and nanocellulose. We demonstrate that dimensional control achieved through strategic precursor synthesis directly influences biomedical performance in cancer therapeutics, tissue-engineering scaffolds, multimodal bioimaging platforms, and cardiovascular applications. Unlike previous application-centric or waste-valorisation reviews, this work integrates sustainability principles with rational materials design, providing researchers with a predictive framework for developing tailored carbon nanomaterials for specific biomedical applications.
Large Language Models face substantial challenges in machine translation for low-resource languages, particularly in specialised domains like legal translation. This study addresses two key obstacles: (i) severe over-tokenisation that fragments semantic units, and (ii) scarcity of parallel training data. We expand the Gemma-3-1B-Instruct ( https://huggingface.co/google/gemma-3-1b-it ) model with language-specific tokens, initialising new embeddings using novel cross-lingual word embedding initialisation, followed by two-stage training: (i) continual pre-training on parallel legal data, and (ii) LoRA [9] fine-tuning with language-specific adapters. Evaluation on four typologically diverse low-resource languages—Irish (GA), Kannada (KN), Traditional Chinese (ZH_TW), and Maithili (MAI)—demonstrates that our cross-lingual initialisation outperforms FOCUS initialisation [4] across all languages and all metrics. Parallel data pre-training substantially surpasses monolingual approaches, and our models outperform NLLB-600M ( https://huggingface.co/facebook/nllb-200-distilled-600M ) baselines across all languages by considerable margins. However, vocabulary expansion models still lag behind non-expanded baselines, indicating more extensive pre-training is needed to fully realise tokenisation improvements.
Academic performance among health professional students is influenced by social determinants. This study assessed factors associated with academic performance among undergraduate health professional students at Mbarara University of Science and Technology (MUST), Uganda. We conducted a cross-sectional survey among 335 undergraduate students using a self-administered questionnaire. Academic performance was categorized as poor or good. Bivariable and multivariable logistic regression analyses were performed to identify factors independently associated with poor academic performance. Of the 335 students (mean age 23.86 ± 6.4 years), 70.8
Alcohol use is a major modifiable risk factor for dementia. However, there is a paucity of literature on alcohol use and the development of dementia in sub-Saharan Africa, especially in Uganda. This study investigated alcohol use and the development of dementia among older persons in rural Uganda. A community-based case-control study among 588 older persons aged 60 years and above in Northern and Southwestern Uganda. Using the IDEA cognitive screening tool (294 Cases and 294 controls) older persons were enrolled. Interview data were collected using a questionnaire. Data were entered in EpiData version 3.0, cleaned, and transferred to SPSS version 26.0 for statistical analysis. Logistic regression was used to determine the risk of development of dementia at a 95
Dementia is a significant public health issue affecting over 50 million people worldwide. Alcohol consumption has been identified as a modifiable risk factor for dementia and its progression. This study aimed to explore the sociocultural factors influencing sex differences in alcohol consumption and community perceptions of its potential cognitive effects, including dementia risk in rural Uganda, where alcohol use is prevalent. We employed a qualitative study design in the Ibanda and Lira districts, guided by the Gender Analysis Framework (GAF). We conducted in-depth interviews with 20 older adults (aged 60 years or older) who consume alcohol and 10 key informant interviews with community leaders. We used a deductive thematic analysis approach to generate themes and sub-themes using GAF. Four main themes are presented in this study: Access to resources, Practices and participation, Beliefs and perceptions, and Institutions, Laws and Policies. We found that men had greater access to alcohol due to their economic independence and social norms, often drinking in social settings. This was facilitated by peer relationships and the acceptance of drinking as a social norm for men. Women’s access to alcohol is limited by household responsibilities and cultural expectations, with consumption mainly at home and in smaller quantities. Community norms discourage alcohol consumption by women and are associated with shame and family disruption. Most participants believed that alcohol could cause memory problems, although some felt that other factors, such as age, could contribute to cognitive decline. Weak enforcement of alcohol regulations and religious influences were noted, with women reported to be more receptive to religious guidance on quitting alcohol consumption. Gender disparities exist in alcohol access, consumption patterns, and social norms in rural Uganda. Participants commonly associated alcohol use with forgetfulness and short-term memory problems, perceptions that are consistent with early cognitive effects of alcohol but not with later stages of dementia. There is a need for gender-specific interventions to address alcohol consumption and to raise awareness about its potential impact on cognitive health in this population.
