In this study, the authors developed and assessed the accuracy of a machine learning algorithm to identify skin cancers using images of biopsies.

Dental resins can be a source of reactive oxygen species (ROS) which in unruly amounts can be toxic to cellular and overall health. In this report, the authors test whether the consumption of antioxidant rich foods like avocado and asparagus can protect against the effect of dental resin-derived ROS. However, rather than testing humans, they use fruit flies and their climbing abilities as an experimental readout.

The authors test the ability of planarians to remember conditioned stimuli following regeneration.

Since the discovery that thousands of different bacteria colonize our gut, many of which are important for human wellbeing, understanding the significance of balancing the different species on the human body has been intensely researched. Untangling the complexity of the gut microbiome and establishing the effect of the various strains on human health is a challenge in many circumstances, and the need for simpler systems to improve our basic understanding of microbe-host interactions seems necessary. C. elegans are a well-established laboratory animal that feed on bacteria and can thus serve as a less complex system for studying microbe-host interactions. Here the authors investigate how the choice of bacterial diet affects worm fertility. The same approach could be applied to many different outcomes, and facilitate our understanding of how the microbes colonizing our guts affect various bodily functions.

The authors compare the predicted versus actual geyser eruption times for the Old Faithful and Beehive Geysers at Yellowstone National Park.

The authors use pictures of candy wrappers and neural networks to improve nutritional accuracy of diet-tracking apps.

In this study, the authors characterized the trends of cigarette use amongst people who do and don't use electronic nicotine delivery systems (or ENDS). This was done to help determine if the use of ENDS is aiding in helping smokers quit, as the data on this has been controversial. They found that use of ENDS among people either with or without previous cigarette usage were more likely to continue using cigarettes in the future. This is important information contributing to our understanding of ways to effectively (and not effectively) reduce cigarette use.

The authors explore how diversity in data sets contributes to bias in artificial intelligence.

Breast cancer is the most common cancer in women, with approximately 300,000 diagnosed with breast cancer in 2023. It ranks second in cancer-related deaths for women, after lung cancer with nearly 50,000 deaths. Scientists have identified important genetic mutations in genes like BRCA1 and BRCA2 that lead to the development of breast cancer, but previous studies were limited as they focused on specific populations. To overcome limitations, diverse populations and powerful statistical methods like genome-wide association studies and whole-genome sequencing are needed. Explainable artificial intelligence (XAI) can be used in oncology and breast cancer research to overcome these limitations of specificity as it can analyze datasets of diagnosed patients by providing interpretable explanations for identified patterns and predictions. This project aims to achieve technological and medicinal goals by using advanced algorithms to identify breast cancer subtypes for faster diagnoses. Multiple methods were utilized to develop an efficient algorithm. We hypothesized that an XAI approach would be best as it can assign scores to genes, specifically with a 90% success rate. To test that, we ran multiple trials utilizing XAI methods through the identification of class-specific and patient-specific key genes. We found that the study demonstrated a pipeline that combines multiple XAI techniques to identify potential biomarker genes for breast cancer with a 95% success rate.

Machine learning algorithms are becoming increasingly popular for data crunching across a vast area of scientific disciplines. Here, the authors compare two machine learning algorithms with respect to accuracy and user-friendliness and find that random forest algorithms outperform logistic regression when applied to the same dataset.