Ayooshi Mitra, Amity University Kolkata
Deep Learning, a popular artificial intelligence method, is a powerful tool for surveying and categorizing biological data. Deep-learning tools could also assist researchers in stratifying disease types, understanding disease subpopulations, discovering new treatments, and matching them with suitable patients for clinical testing and treatment.
A deep learning (DL) algorithm that uses data from low-dose CT scans to screen for lung cancer accurately predicts the risk of death from cardiovascular disease. In a study published in Radiology: Cardiothoracic Imaging, researchers used DL to test a faster, automated method for predicting five-year cardiovascular disease mortality with the only minor extra workload. The researchers trained the method to quantify six types of vascular calcification using data from 4,451 participants, with a median age of 61 years, who underwent low-dose CT over two years in the National Lung Screening Trial. They then put the method to the test using data from 1,113 participants. The prediction model based on calcium scores outperformed the baseline model, which relied solely on self-reported participant characteristics such as age, smoking history, and illness history.
The method works in two stages. DL is used in the first stage to determine the amount and location of arterial calcification in the coronary arteries and aorta. The second stage predicts mortality using a more traditional statistical approach. The second stage also shows which characteristics are most predictive of five-year mortality. The method relies solely on image data, is fully automated, and is quick. In less than a half-second, the method obtains calcium scores in a full chest CT. This implies that the method should be simple to incorporate into routine patient workups and screenings. Most importantly, the method may aid in identifying individuals in a population of heavy smokers who may be at increased risk of death from cardiovascular disease-related causes. According to researchers, lung screening studies show that heavy smokers die from cardiovascular disease just as much as from lung cancer.
The researchers created several automatic calcium scoring methods that can be applied to a wide range of data. They are now developing a calcium scoring method that detects arterial calcification accurately in low-quality data, such as data affected by cardiac motion, low image resolution, or high noise levels. They devised a method for detecting coronary calcifications even when the lesions are below the clinically accepted threshold. They hope that by doing so, they will be able to improve the reproducibility of calcium scoring and enable more accurate prediction.
Also read: Reflectin and Embryonic kidney cells with Cephalopod-like features!!!
Reference:
- de Vos, B. D., Lessmann, N., de Jong, P. A., &Išgum, I. (2021). Deep Learning–Quantified Calcium Scores for Automatic Cardiovascular Mortality Prediction at Lung Screening Low-Dose CT. Radiology: Cardiothoracic Imaging, 3(2), e190219.
- Algorithm Predicts Risk of Death from Heart Disease Using Information from Lung Cancer Screening CT https://appliedradiology.com/communities/Artificial-Intelligence/algorithm-predicts-risk-of-death-from-heart-disease-using-information-from-lung-cancer-screening-ct-exam
The Pacific Ocean shows brand new octopus species
Rohit Bhattacharjee, Amity University, Kolkata. The dumbo or finned octopuses are a group of remote ocean staying octopuses that incorporate around 45 species. Among cephalopods, they establish moderately uncommon creatures, despite shaping up a huge piece of the megafauna in remote ocean living spaces of the World Ocean down to in any event 7,000 m (4.3 […]