back to top
  • AI & Machine Learning, Biology, Computing, Data & AI, Life Sciences
    Abstract Medical trainees depend on diverse visual reference material to develop diagnostic pattern-recognition skills, yet such material is often limited in diversity and accessibility. This paper presents a classifier-in-the-loop generative framework for synthesising realistic histopathology images: a Stable Diffusion v1.5 model is fine-tuned using Low-Rank Adaptation (LoRA) on the PathMNIST dataset, consisting of nine colorectal […]
  • Biology, Chemistry, Life Sciences, Physical Sciences
    Abstract The number of diabetes patients is increasing worldwide, but many cases stay undiagnosed in low-resource regions. The A1C blood test is invasive and costly, therefore it is often not available in these regions. In this research, DiaBreath is proposed as a low-cost noninvasive screening system based on exhaled breath analysis. Volatile organic compounds (VOCs) […]
  • Computing, Data & AI, Data Science, Economics, Social Sciences, Humanities & Policy
    Abstract Digital commerce platforms compete with local vendors under frictions of time, price, and visibility. We test whether an interpretable choice model with reinforcement and social influence can yield substantial changes in aggregate market share as control parameters vary. We simulate repeated customer choice among vendors using a utility function that combines vendor attributes (price, […]
  • AI & Machine Learning, Computing, Data & AI
    Abstract Local businesses are important to neighborhood service access, but access to these services is limited in some areas. This study uses GIS and machine learning to identify possible limited service zones for restaurants and supermarkets across Dublin, Pleasanton, and Livermore in Alameda County, California. The model dataset included 1,134 Alameda County census block groups, […]
  • Bioinformatics, Life Sciences
    Abstract GRCh38, the previous human reference genome assembled primarily with short read sequencing, contained substantial unresolved regions, with roughly eight percent of the genome represented by gaps within highly repetitive telomeric and centromeric sequences, whereas the T2T-CHM13 assembly provides a gapless, telomere-to-telomere human reference genome. This progress was enabled by the development of long-read sequencing […]
  • AI & Machine Learning, Computing, Data & AI
    Abstract Prior mental health NLP work used social-media data, sport language models, sentence embeddings, and tweet-sized transformer output for the purposes of classification and representation. This initial comparative experiment aims to determine if various models may be able to cluster sport focused athlete mental health responses into four specific research labels: anxiety, ADHD, depression and […]
  • AI & Machine Learning, Computing, Data & AI
    Abstract Large language models (LLMs) are increasingly used to assist students and teachers, yet the benchmarks used to select them rarely test the school-specific everyday knowledge that student-facing applications draw on. Most benchmarks also score only a single best answer, even though many commonsense questions admit several plausible responses. We construct a High School Commonsense […]
  • AI & Machine Learning, Computing, Data & AI
    Abstract Large-scale deep learning models have become the dominant approach in audio NLP, yet their computational demands make deployment impractical in resource-constrained environments. This study evaluates five lightweight architectures (SVM, MLP, BiLSTM, MobileNet, CRNN) against a transformer baseline on an audio gender classification task. Each model was trained and fine-tuned on a 14,461-sample dataset of […]
  • Earth & Environment, Environmental Science
    Abstract Per- and polyfluoroalkyl substances (PFAS) are man-made fluorinated chemicals that are persistent in the environment with very slow degradation. Consequently, their release to the environment has led to widespread contamination of the environment and bioaccumulation in human blood. PFAS are endocrine disruptors and act by interfering with endocrine and immune system communication to disrupt […]
  • Biology, Earth & Environment, Environmental Science, Life Sciences
    Abstract Background/Objective: International wildlife trade monitoring relies on accurate reporting from both importing and exporting nations. However, regulations built on this data often prove inadequate, as gaps in records allow disease risks to slip through the cracks. This study investigated discrepancies in United States live mammal import tracking data in the CITES Trade Database between […]
  • AI & Machine Learning, Computing, Data & AI
    Abstract Political misinformation spreads rapidly online, motivating automated tools to support verification. This study compared four traditional classifiers—Multinomial Naive Bayes, Logistic Regression, Linear Support Vector Machine, and Decision Tree—with a convolutional neural network, a bidirectional long short-term memory network, and DistilBERT on the text-only, six-class LIAR benchmark. All models used the official training, validation, and […]
  • Aerospace Engineering, Engineering
    Abstract Rotating detonation engines (RDEs) have emerged as a promising propulsion technology, offering superior thermodynamic efficiency over conventional Brayton-cycle engines due to their utilization of detonation combustion. Annular combustors (with inner and outer walls) are typically used in RDEs. However, hollow—lacking an inner wall—combustors are emerging as a promising alternative due to their potential mitigation […]
  • Biology, Earth & Environment, Environmental Science, Life Sciences
    Abstract Background/Objective: International wildlife trade monitoring relies on accurate reporting from both importing and exporting nations. However, regulations built on this data often prove inadequate, as gaps in records allow disease risks to slip through the cracks. This study investigated discrepancies in United States live mammal import tracking data in the CITES Trade Database between […]
  • Mathematics, Mathematics & Statistics
    Abstract Sudoku, with its complex combinatorial structure, provides a natural NP-hard benchmark for testing optimization models. In this study, we quantitatively compare Quadratic Unconstrained Binary Optimization (QUBO) and Higher-Order Binary Optimization (HOBO). While QUBO is compatible with quantum computers, it is often evaluated on classical simulators known as quantum simulators which calculates with tensor network […]
  • Social Sciences, Social Sciences, Humanities & Policy
    Abstract Sporting activities have proven to be another important site of learning, displaying and challenging hegemonic masculinities. Previous research has, however, mostly considered hegemonic masculinities, inclusive masculinities and the fitness culture of women as separate topics. This study examines relationships among masculine pressures, inclusive masculinity, perceived benefits of participating in sport-related activities and attitudes towards […]

Share Your Research

NHSJS publishes original research by high school students across all scientific disciplines. Our peer-review process provides valuable feedback and mentorship from experts in the field.

Submit a Manuscript