Researchers are developing preclinical microgrippers that could be deployed throughout the upper urinary tract to grab tiny pieces of tissue and facilitate early detection of disease.
More by Karen Olsen
NIBIB-funded researchers are working to bring in vivo gene editing to the fore. Through rational engineering of lipid nanoparticles, this collaborative team developed a way to effectively target specific organs in the body to precisely deliver therapeutic cargo, including gene-editing molecules. Their research demonstrated that a one-time treatment with their nanoparticles resulted in durable gene editing in mouse lungs for nearly two years. Further, their technique showed promise in correcting a mutation present in a currently untreatable form of cystic fibrosis in several models of the disease.
As AI is deployed in clinical centers across the U.S., one important consideration is to assure that models are fair and perform equally across patient groups and populations. To better understand the fairness of medical imaging AI, researchers trained over 3,000 models spanning multiple model configurations, algorithms, and clinical tasks. Their analysis of these models reinforced some previous findings about bias in AI algorithms and uncovered new insights about deployment of models in diverse settings.
A team of NIBIB-funded researchers recently developed an AI platform that can analyze 3D pathology images to predict disease outcomes. Their method had improved performance in predicting prostate cancer outcomes when compared with traditional pathology approaches, such as analysis by expert pathologists using 2D images.
NIBIB-funded researchers are working to make bladder surgeries better, tackling the issue from two vantage points: improving bladder function using a biodegradable construct that facilitates tissue regeneration, and enhancing patient monitoring by developing an implantable bladder sensor.
A collaborative NIH-funded team is using AI to mine common chest CT scans to predict mortality. Their research identified a collection of cardiac factors that were predictive of death in a large group of patients, potentially setting the stage for improved cardiac screening.
NIBIB-supported researchers have developed a smart nanoprobe designed to infiltrate prostate tumors and send back a signal using an optical imaging technique known as Raman spectroscopy. The new probe, evaluated in mice, has the potential to determine tumor aggressiveness and could also enable sequential monitoring of tumors during therapy to quickly determine if a treatment strategy is working.
A multidisciplinary group of NIH-funded scientists have successfully captured real-time, high-resolution images of the developing mouse placenta during the course of pregnancy. Their technique, which combines a surgically implanted window with a next-generation imaging system, provides key insight into placental development under both healthy and pathological conditions.
What if bacteria—which love to grow deep inside tumors—could guide cancer therapies directly to their target? NIH-funded researchers have engineered a bacterial strain to “light up” tumors so that reprogrammed T cells, drawn like a moth to a flame, can find and destroy them. Their preclinical treatment could potentially be effective against any solid tumor type.
A collaborative team of NIH-funded researchers is developing a way to obtain DNA shed from brain tumors using focused ultrasound. Their first-in-human study could be an important step towards improving the way brain tumors are diagnosed.
NIH-funded researchers have outlined a method to print biocompatible structures through thick, multi-layered tissues using focused ultrasound.
Dendritic cells are key orchestrators of the immune response, but most vaccination strategies don’t effectively target them. NIBIB-funded researchers have developed biodegradable nanoparticles that are designed to deliver mRNA cargo to dendritic cells in the spleen. Combined with another type of immunotherapy, their vaccine had robust antitumor effects in multiple mouse models.
Nanozymes—artificial enzymes that can carry out pre-determined chemical reactions—could selectively activate a cancer drug within a tumor while minimizing damage to healthy tissue in a mouse model of triple negative breast cancer.
This fully wireless ultrasound patch, which can capture detailed medical information and wirelessly transmit the data to a smart device, could represent a major step forward in at-home health care technology.
This interview with Maryellen Giger, PhD, delves into the creation of the MIDRC imaging repository, how its data can be used to develop and evaluate AI algorithms, ways that bias can be introduced—and potentially mitigated—in medical imaging models, and what the future may hold.