Diagnostic Imaging’s Advances in AI monthly roundup is your chance to catch up on the latest AI news, AI-powered imaging modalities and emerging AI research affecting the field of radiology. Review the slideshow below to see highlights from the past month.
Deep learning algorithms have been shown to detect lung cancer on low-dose computed tomography scans with great accuracy. Researchers trained a deep learning model on a dataset of 4,960 low-dose CT scans, and it was able to detect 96.7% of lung cancers, compared to 78.7% for human radiologists. This has significant implications for early detection and treatment of lung cancer. Diagnostic imaging techniques such as low-dose CT scans are becoming more common and AI-powered detection methods like this one are set to change the field of radiology.
In recent news, a team of researchers from the University of Oxford has developed an AI-powered system that can detect diabetic retinopathy in eye exams. Diabetic retinopathy is a common complication of diabetes that can cause serious vision loss if left untreated. The AI system analyzed 868 retina scans and was able to detect more than 97% of cases. While the study was small, its results are promising and could potentially be applied to large-scale health screenings.
Using AI to improve diagnostic imaging doesn’t just mean better detection of diseases; it can also reduce radiation exposure for patients. Researchers at the University of California, Los Angeles (UCLA) have developed an AI-powered system that can reduce radiation exposure from CT scans by as much as 70%. The system works by identifying the areas of the body that don’t need to be imaged in detail. By leaving those areas out of the scan, the system reduces radiation exposure without compromising image quality.
Another area of AI research in diagnostic imaging is the development of AI-powered imaging modalities. A new study published in the Journal of Medical Imaging has explored the use of machine learning algorithms to reconstruct MRI images more quickly. MRI scans provide detailed images of internal organs, but they often take a long time to acquire. The researchers developed a machine learning algorithm that was able to reconstruct MRI images in just a few seconds, compared to the traditional method which can take up to 30 minutes.
Researchers from the University of Chicago and the University of Pennsylvania have also made significant advances in breast cancer detection using AI. They developed a deep learning model that was able to detect invasive breast cancer in mammography images with great accuracy. The model performed better than human radiologists in some cases, highlighting the potential of AI in diagnostic imaging.
Researchers at Google Health have been working on an AI-powered system that can detect breast cancer from mammography images. They developed a deep learning model that was trained on a dataset of 2.5 million breast cancer screening exams. The model was able to detect breast cancer with great accuracy and was even better than human radiologists in some cases. This has significant implications for early detection and treatment of breast cancer.
About The Author
