Researchers developed a new artificial intelligence (AI) tool known as a deep learning convolutional neural network (CNN), and found it is able to diagnose skin cancer (specifically melanoma) more accurately than a group of 58 international dermatologists with varying levels of expertise. The tool was created using Google’s AI technology and was trained using 100,000 skin images.
Breaking Research News - cancer
A new research approach screens a library of 200,000 chemical compounds with more than 100 well-characterized, human lung cancer cell lines to identify 'therapeutic triads': potential drug candidates that would kill the cancer cells, biomarkers that would predict who may respond to the candidate drug, and the molecular targets of each active chemical.
Researchers recently developed a three-dimensional cellular model of lung cancer encased within a clear gelatinous matrix that allowed them to study the interaction between the tumor cells, immune cells, and the tumor microenvironment.
Researchers used a mathematical method to convert genome-wide data on gene expression and epigenetic modifications (i.e., DNA modifications that regulate gene expression) from a human cancer stem cell line to a readily analyzable format and applied it to identify critical genes relating to cancer development and drug resistance.
Using gastric cancer biopsies from 77 patients, scientists recently identified a panel of proteins (BRF1, BRCA1/2, and MPO) induced by alcohol consumption that can predict the recurrence of the cancer, the overall survival, and the response to platinum-based adjuvant chemotherapy.
Using a cervical cancer cell line, scientists demonstrated that nutrients from blueberries can boost the cancer-fighting power of radiation treatments from 20-70 percent by increasing cancer cell death and inhibiting cancer cell growth.
Researchers recently used a lung-on-chip model to study human lung cancer cell growth in response to drugs and breathing.
Chemotherapy is a nonspecific way of killing off cancer cells with drugs that is often accompanied by many toxic side effects. About 70 percent of breast cancer patients are diagnosed with a tumor containing estrogen receptors (ER), which requires treatment with hormones but not necessarily chemotherapy. Whether a patient needs chemotherapy depends on the aggressiveness of the tumor, which is currently determined by an expensive laboratory test that requires tissue biopsy and time to ship and run the test.
Defining in advance whether breast cancer might invade other organs of the body (a process called metastasis) will help guide treatment options (i.e., conservative versus aggressive treatments). A group of researchers has recently developed a new cellular (in vitro) and computational model that could help predict the risk of metastasis for patients affected by breast cancer.
Breast cancer is characterized by progressive modifications of the microenvironment of mammary tissue. Being able to monitor these tissue changes can be critical to improving clinicians’ diagnoses and designing early intervention strategies.