Diagnostic Imaging Weekly Scan: July 24, 2020


Lung cancer, X-ray improvements, Post-Pandemic Success, and Minority Patients and COVID-19

Welcome to Diagnostic Imaging’s Weekly Scan. I’m Whitney Palmer, senior editor, here with the top stories of the week.

Even though lung cancer screenings are known to be an effective way to identify lung cancer and initiate early treatment, not all patients who could benefit from them stick with the annual screening. In a study, published in the American Journal of Roentgenology, investigators from the University of Pennsylvania Perelman School of Medicine discussed a strategy that can improve adherence to screening initiatives. Based on an analysis of 260 patients who returned for follow-up lung cancer screening, 43 percent stuck with the initiative. They found former smokers were more likely to stay with it than current smokers with 50 percent of former smokers returning and only 36.2 percent of current smokers doing so. To improve adherence, the team suggested focusing on current smokers and patients with negative lung CT scans since a negative scam does not rule out possible cancer. The team recommended creating electronic medical record tracking systems that detect patients who miss their lung cancer screening appointments or offering health insurance premium reductions to promote participation.

Also in lung cancer news, a new deep learning-based automatic detection algorithm can improve the sensitivity of chest X-rays for detecting the disease. Not only could its use decrease the number of unnecessary follow-up CT scans, but it could also reduce the amount of radiation exposure to patients. In a study, published in Radiology, investigators from South Korea outlined the efficacy of their algorithm that is designed to detect overlooked lesions. To evaluate how well it works, the team analyzed chest X-rays from 117 people who had lung cancer, as well as 234 healthy counterparts. From the group who had lung cancer, there had been 105 lesions that were initially overlooked. Nine observers reviewed the X-rays twice – once with the algorithm and once without it. Across the board, the detection of lesions improved with the use of the algorithm – the observers pinpointed 53 percent more cancers with it compared to 40 percent more without it. Based on these findings, they were able to more accurately suggest follow-up CT scans for actionable lung cancers – recommendations for additional images rose from 54 percent without the algorithm to 71 percent with it. Ultimately, the team said, the AI tool could be of the most benefit to early-career radiologists, as well as those who do not have a thoracic imaging specialization.

What happens next with the pandemic is still unclear, but that doesn’t mean that there are not certain factors that practices must consider if they want to have the most successful post-pandemic future. In an article published in the Journal of the American College of Radiology, past-chair of the American College of Radiology’s Board of Chancellor’s Dr. Geraldine McGinty and her colleague from Weill Cornell Medicine, Dr. Robert Min outlined five specific indicators that will be pivotal as the pandemic recovery continues. They touched on changes to reimbursement and how practices should handle sensitive financial conversations, as well as the regulatory changes that will remain after the pandemic subsides, including broader telehealth, consolidation efforts, and downward pressure on fee-for-service rates. In addition, they discussed the supplemental – not replacement – role of artificial intelligence tools, as well as the need for radiologists to identify new ways to interact, network, and conduct research and educational efforts in the future. And, lastly, they pointed to the need for increased cybersecurity to protect both providers working from home and the patients whose scans they read.

But, even among all of the planning and forward-thinking, providers are still focusing heavily on learning more about COVID-19 and how to best treat it. In an article published in Ultrasound in Medicine & Biology, researchers from Gemelli University Policlinic in Rome looked into whether point-of-care ultrasound can be used to effectively identify which patients are more likely to have severe outcomes from the viral infection. The results of their study, they said, showed that bedside ultrasound in the emergency department can, in fact, predict on first evaluation the overall prognosis of COVID-19-positive patients. To reach this conclusion, they analyzed bedside ultrasound scans from 41 COVID-19-positive patients in their emergency department, concentrating on 14 lung areas. They gave each scan a score from 0 to 3 with 3 indicating the most severe disease, including white lung with or without subpleural consolidations. Overall, they said, more than 90 percent of these patients had at least one lung area with an abnormal finding, and among those who died, every scanned area contained a pathological finding. Of those patients who recovered, similar findings appeared in 50 percent of scanned areas. Although the study was conducted at one institution, the team said they hope their results could be generalized and improve the triaging of COVID-19 patients.

And, finally, this week, Diagnostic Imaging spoke this week with Dr. Efren Flores, a radiologist at Massachusetts General Hospital about his recent work, examining the prevalence of more severe COVID-19 disease in patients from racial and ethnic minority groups. According to the research he conducted with colleagues, there are a number of socioeconomic factors that contribute heavily to this larger disease burden in this population. He spoke with us about those elements and also discussed the active role that radiologists can play in mitigating those problems and ultimately improving patient outcomes. Here’s what he shared.

Related Videos
Nina Kottler, MD, MS
The Executive Order on AI: Promising Development for Radiology or ‘HIPAA for AI’?
Practical Insights on CT and MRI Neuroimaging and Reporting for Stroke Patients
© 2023 MJH Life Sciences

All rights reserved.