A freely available computer program that takes into account specific local conditions enables users to predict the impact of adopting different diagnostic strategies on the spread of tuberculosis in their region.
Application of machine learning to serum miRNA profiles generated through next generation sequencing identifies a biologically relevant miRNA signature which can be deployed as a qPCR test to assist the diagnosis of epithelial ovarian cancer.
Mass spectrometry on plasma from patients with typhoid fever and other febrile disease identified and validated 24 metabolites that can distinguish typhoid from other febrile diseases, providing a new approach for typhoid diagnostics.
PCR testing for SARS-CoV-2 from samples taken from smartphone screens, Phone Screen Testing, provides a sensitive, cost-effective, simple, and non-invasive new method that could boost COVID-19 mass test screening.
Applying deep learning technology for the large-scale curation of symptoms from unstructured EHR clinical notes accurately predicts the differential signals of COVID-19 diagnosis over the week preceding typical PCR testing.
A model based on empirical parameter estimates predicts that arresting cancer cell growth by less than 1% per day will produce optimal outcomes in preventing life-threatening cancers, and that such preventive measures are generally more successful than post-diagnostic interventions.