Browse Articles

Pancreatic ductal adenocarcinoma (PDAC) is the most common form of pancreatic cancer, with early diagnosis and treatment challenges. When any of the genes KRAS, SMAD4, TP53, and BRCA2 are heavily mutated, they correlate with PDAC progression. Cellular stress, partly regulated by the gene SERPINA6, also correlates with PDAC progression. When SERPINA6 is highly expressed, corticosteroid-binding globulin inhibits the effect of the stress hormone cortisol. In this study, the authors explored whether there is an inverse correlation between the expression of SERPINA6 and PDAC-linked genes.

Here seeking to develop a method to diagnose, hypertrophic cardiomyopathy which can cause sudden cardiac death, the authors investigated the use of a convolutional neural network (CNN) and long short-term memory (LSTM) models to classify cardiac magnetic resonance and heart electrocardiogram scans. They found that the CNN model had a higher accuracy and precision and better other qualities, suggesting that machine learning models could be valuable tools to assist physicians in the diagnosis of hypertrophic cardiomyopathy.

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

Mammographic screening is a common diagnostic tool for breast cancer among average-risk women. The authors hypothesized that adherence rates for mammographic screening may be lower among minorities (non-Hispanic black (NHB) and Hispanic/Latino) than among non-Hispanic whites (NHW) regardless of the guideline applied. The findings support other studies’ results that different racial/ethnic and socio-demographic factors can affect screening adherence. Therefore, healthcare providers should promote breast cancer screening especially among NHW/Hispanic women and women lacking insurance coverage.

Using a convolution neural network, these authors show machine learning can clinically diagnose breast cancer with high accuracy.

The authors test the effectiveness of machine learning to predict onset of cardiovascular disease.

Skin cancer is a common and potentially deadly form of cancer. This study’s purpose was to develop an automated approach for early detection for skin cancer. We hypothesized that convolutional neural network-based models using transfer learning could accurately differentiate between benign and malignant moles using natural images of human skin.

Here, recognizing the difficulty associated with tracking the progression of dementia, the authors used machine learning models to predict between the presence of cognitive normalcy, mild cognitive impairment, and Alzheimer's Disease, based on blood DNA methylation levels, sex, and age. With four machine learning models and two dataset dimensionality reduction methods they achieved an accuracy of 53.33%.

In this study, a deep learning model is used to classify post-traumatic stress disorder patients through novel markers to assist in finding candidate biomarkers for the disorder.

Cholesterol is a major component of neuronal cell membrane and myelin sheath. In this study, the authors either transfected Neuro 2A cells with CYP46A1 cDNA or treated the cells with 24SHC. Cells expressing CYP46A1 had significantly less viability compared to the negative control. Up to 55% reduction in cell viability was also observed in 24S-HC-treated cells. This work supports that CYP46A1 and 24S-HC could directly trigger cell death. The direct involvement of 24S-HC in cell death provides further evidence that 24S-HC can be a promising biomarker for diagnosing brain damage severity.