machine learning

Researchers created a model that uses clinical testing data to locate the primary site of cancer cells with no known origin, likely improving survival.

Researchers integrate scRNA-seq, spatial transcriptomics, and histology imaging data to show that spatial cellular architecture predicts glioblastoma prognosis.

A new machine learning model promises fast prediction of drug-target interactions.

The signature could one day be used to improve treatment planning for people with substance abuse disorders.

A method appears to be the first to noninvasively reconstruct language from brain activity.

A study in mice finds that for certain genes, one parent’s allele can dominate expression and shape behavior—and which parent’s allele does so varies throughout the body.

The Scientist spoke with Ohio State University microbiologist Matthew Sullivan about a recent expedition that identified thousands of RNA viruses from water samples and cataloged them into novel phylogenic groups.

Armed with improved imaging techniques and supercomputers, researchers are generating detailed three-dimensional images of cellular structures that anyone can explore.

The experimental system, developed and tested in just one patient so far, relies on brain signals associated with handwriting to achieve the fastest communication yet seen with BCI.

While machine learning could improve detection of tumors at their earliest stages, it also risks identifying malignancies that would never cause the patient any harm.

Good girl, Florin!

In a proof-of-concept study, researchers used dogs’ diagnoses of prostate cancer to inform a machine learning algorithm with the goal of one day detecting cancers with canine-level accuracy.

Some researchers have turned their focus to artificial intelligence to study an underused symptom common to most respiratory pathogens—cough.

A small study shows artificial intelligence can pick out individuals with coronavirus infections, but ophthalmologists and AI experts say the approach is far from proven to be capable of distinguishing infections with SARS-CoV-2 from other ills.

The company had helped develop a tool to aid decision-making in distributing limited medical equipment among coronavirus patients, but two high-profile retractions call into question the validity of Surgisphere’s work in toto.

The University of Pittsburgh scientist, who was studying SARS-CoV-2, was shot to death in his home.