When you were in medical school, your stethoscope likely felt like an extension of your hand – always by your side and your go-to tool for conducting exams and assessing your patients. For the next generation of medical students, point-of-care ultrasound could take that coveted spot.

At least, that is the goal of a burgeoning partnership between the College of Osteopathic Medicine at Touro University in Nevada and point-of-care ultrasound manufacturer Vave Health, said Amin Nikoozadeh, Vave founder and chief executive officer.

“We want to empower doctors, and ultrasound is the best tool to do that,” he said. “Ultrasound has a lot of utility. In a lot of senses, it is better than the stethoscope in that it allows students to see instead of just hear.”

Starting this semester, Vave will outfit Touro’s medical students with wireless pocket-sized ultrasound devices – that can sync with tablets and smartphones – that they can use for hands-on training in a variety of settings, including clinical rotations, in-person lab classes, or remotely from their home. In addition, each student will also receive a Vave EduPlus Membership that gives them access not only to ultrasound experts, but also to in-app content that can help with medical exam preparation and offer clinical context. The details of what educational details will be included in the app are still under discussion.

But, by putting a portable ultrasound in the hands of every medical student, they can begin learning – from day one – how what they read about in their textbooks actually appears in real patients.

“Putting…ultrasound in the hands of our students provides them with cutting-edge technology to further their hands-on education, something that is so critical, especially in a time when traditional forms of learning have been rattled by the pandemic,” said Wolfgang Gillar, D.O., dean of the College of Osteopathic Medicine. “We’re proud to provide our students with a probe that allows them to practice in a variety of environments, whether in a hospital, rural area, or an underserved community.”

Integrating ultrasound during the first two years of training could fundamentally change medical education, said Renee Dversdal, M.D., Vave’s chief medical officer.

“It’s taking advantage of different educational concepts, longitudinal learning, and applying ultrasound,” she said. “For example, when they’re studying the carotid artery, we can show them the three layers, as well as show them where plaque accumulates and what happens with turbulent flow.”

The idea, she said, is to weave point-of-care ultrasound seamlessly into medical training and patient care. While receiving didactic training, students will also gain instruction and experience in image acquisition, image interpretation, and clinical integration.

“It’s a cooler way to learn anatomy and highlight the clinical exam,” she said. “You’re simultaneously sneaking in image acquisition and image interpretations skills. They’re subtly learning the basics of these skills. The goal is to have ultrasound so woven into their education that by the time they graduate, they won’t even have to think about doing ultrasound – it’ll just be second nature.”

Most importantly, she said, this partnership is not one-size-fits-all. It can be customized to meet the needs and adhere to the curriculum designs of individual medical schools. Basically, administrations can create their own programs.

And, from a partnership perspective, Nikoozadeh said, this is not a one-time interaction. The business model is designed to be low-cost for the medical school, and Vave will routinely provide updates and upgrades to the software as needed to ensure students have access to the most current information.

“We want our partners to know that they are getting a platform that evolves with them – that they’re not going to get left behind,” he said. “They will have access to the latest technologies in hardware, software, and artificial intelligence. This is a change that will lead to faster adoption of ultrasound technology at both the medical school and clinical levels.”

For more coverage based on industry expert insights and research, subscribe to the Diagnostic Imaging e-Newsletter

Articles

A partnership between one medical school and an ultrasound manufacturer is intended to re-design how students learn.

Is Ultrasound the New Stethoscope?

There is more disagreement between radiologists and more errors identified on secondary interpretations of body MRI scans than previous studies have reported, leading some experts to call for more sub-specialty interpretations.

Nearly 70 percent of body MRI interpretations have at least one discrepancy, according to researchers at the University of Vermont and the University of Southern California Medical Center. And, since most of these errors are cognitive – a misidentification of a finding – lead study author Danielle E. Kostrubiak, M.D., a diagnostic radiology resident at the University of Vermont Medical Center, said there should be a push for having sub-specialty trained providers read these studies.

“The data suggest that sub-specialty interpretations should be encouraged at tertiary care centers,” she said, “and institutions should provide adequate resources for these interpretations to occur.”

It is the first study to focus on secondary interpretations of body MRI evaluated by type of likely error, they said.

Diagnostic Errors: Lessons Learned and Mitigation Strategies

Interpretation errors are common in radiology, particularly with MRI scans, and those mistakes frequently lead to delayed or incorrect treatment plans. According to existing research, discrepancy rates can range from between 2 percent and 6 percent for resident reports over-read by attending providers to up to 56 percent for secondary interpretations. Most of these problems occur with pelvic and abdominal imaging, they said.

To determine the actual discrepancy rate, as well as why these mistakes happen, Kostrubiak’s team retrospectively reviewed 357 secondary body MRI reports that were captured between January 2015 and December 2018. They also analyzed the initial reports and divided the discrepancies by the most likely underlying reason.

Based on their evaluation, they identified at least one discrepancy in 246 cases (68.9 percent), and they pinpointed a second discrepancy in 54 of those cases. Cognitive errors accounted for most discrepancies – 68.8 percent of errors – perceptual errors accounted for 59.3 percent of secondary discrepancies.

“This difference [between cognitive and perceptual errors] may be explained by the fact that abdominal and pelvic MRI is frequently not the first imaging modality used and is often performed to characterize an abnormality detected on CT or ultrasound,” they said. “Therefore, the abnormality has already been perceived by the time MRI is performed.”

When they drilled down in the underlying reasons behind these discrepancies, they found that faulty reasoning – a misclassification of the abnormality – was responsible for 34.3 percent of all instances, as well as 37.8 percent of primary discrepancies. In addition, satisfaction of search occurred with 37 percent of second discrepancies, and 15 percent of overall discrepancies. In these instances, the team said, an MRI scan has likely been ordered to answer a specific question, and once that answer has been determined, the radiologist likely did not examine the rest of the scan for any other abnormalities.

There could be several reasons why these discrepancy rates are higher than what has been previously reported, the team said. Existing research already shows that body imaging frequently has highest error rates, and double-reading by sub-specialists also increases the discrepancy rate, they added. In addition, general radiologists may be unaware of MRI’s high sensitivity can be used to determine specific diagnoses.

It is possible that similar studies will be limited by the innate subjectivity of error classification, but the team said related research should be easier to conduct in the future as medical practices expand their use of electronic medical records.

“The next step,” they said, “would be to explore how these discrepancies may impact patient outcomes and overall cost to the system associated with these radiologic errors.”

Mistakes highlight the need for more sub-specialty reads with MRI, particularly abdominal and pelvic scans.

Discrepancy Rates and Errors a Bigger Problem with Body MRI Than Realized

A new ultrasound technique has been shown to quantify lung scarring and detect lung fluid. It is a strategy, researcher said, that can be a non-invasive, more affordable option than CT.

IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control

 researchers from North Carolina State University and the University of North Carolina at Chapel Hill outlined the technique that they have already shown is effective in rats. A study in humans is currently underway, and if successful, it could give healthcare providers a tool to better determine the efficacy of medical interventions aimed at reducing pulmonary fibrosis or pulmonary edema. 

“CT scans use radiation, so you want to limit their use. They are also expensive and require a trained radiologist,” said co-senior author Marie Muller, Ph.D., associate professor of mechanical and aerospace engineering at N.C. State. “Ultrasound is a good solution because it does not pose a cancer risk. It’s portable, it’s relatively inexpensive, and our technique effectively gives users a quantitative assessment of the fibrosis.”

Giving providers an easy-to-use tool to evaluate treatments could have a positive impact on patient care, they said, because the most common form of pulmonary fibrosis affects 200,000 Americans with 50,000 new cases diagnosed annually. The team is also investigating, with support from a National Institutes of Health grant, how well their technique can quantify pulmonary edema in heart failure patients. Making that assessment is critical, they said, 75 percent of those patients experience pulmonary edema.

“Being able to monitor pulmonary edema in patients with heart failure would also be very useful,” Muller said. “This is often done by assessing fluctuations in a patient’s body weight in order to estimate how much fluid has collected in the patient’s lungs – which is not as specific as we’d like it to be.”

With their automated assessment technique, the ultrasound transducer collects data and feeds it through a computational model that can calculate the density of healthy alveoli in the lung and offer a quantitative assessment of how much fibrosis is present in the lungs. The team said they are also testing whether their strategy can correctly identify how much water has collected in the lungs.

Their technique is different, they said, because previously existing strategies could only provide qualitative assessments, alerting providers to whether lung health was “bad or good” without offering a measurable gradient. The output of this new strategy is a number that can be captured and understood by providers without an expertise in ultrasound.

“The automated quantitative assessment would allow the technology to be used by personnel with minimal training and would allow healthcare providers to compare data across time,” Muller said. “For example, caregivers would be able to tell if a patient’s edema is getting better or worse.”

The strategy could be a less expensive alternative to CT when evaluating pulmonary fibrosis and pulmonary edema.

New Ultrasound Technique Could Offer Cheaper Option for Lung Health Assessments

’s Weekly Scan. I’m senior editor, Whitney Palmer.

Before we get to our featured interview with Dr. Christoph Lee about the latest trends and innovations in breast imaging, here are the top stories of the week.

For all its growing popularity throughout the industry, artificial intelligence has not yet gained the confidence of women in the general public. Nearly 80 percent say they do not want an AI tool to be the only interpreter of their screening mammograms. This level of opposition means the industry should pump the brakes on implementing AI with mammography and double-down on efforts to educate women about AI’s capabilities and benefits. Researchers from The Netherlands reached this conclusion by surveying 922 women between the ages of 16 and 75. They published their results in the 

Journal of the American College of Radiology

Based on their findings, even having AI as a second read on select studies was not very popular – 42 percent disagreed with it, and 27 percent were undecided. But, there is room for AI to earn more acceptance, the researchers said. Only 17 percent were categorically opposed to using the tool in collaboration with the radiologist. Even though there is resistance to using the technology now, it is critical to gauge women’s opinions and listen to their concerns, the team said. Proper and effective implementation of artificial intelligence tools cannot occur if their worries are not addressed and they are not willing to participate.

 revealed a connection between impaired blood flow and the accumulation of tau-protein in patients who have Alzheimer’s disease. Until now, the connection between vascular dysfunction and tau pathology has not been well understood. To provide clarity, a team from the University of Southern California in Los Angeles examined MRI and PET scans from 206 individuals – 68 USC patients who had normal to mild cognitive impairment and 138 patients with mild cognitive impairment or Alzheimer’s disease who were enrolled in the Alzheimer’s disease Neuroimaging Initiative. Alongside the scans, the patients also completed neuropsychological testing and were assessed for executive brain function, attention, and memory. Based on these tests, the team discovered spots in the brain with more significant tau build-up also had decreased blood flow, particularly in the inferior temporal gyrus – the area of the brain that typically experiences tau build-up in Alzheimer’s patients. And, this correlation between tau and vascular function was strongest in patients with greater cognitive impairment and higher amyloid-β levels. It appeared in more brain regions as the disease progressed. Consequently, the team posited, targeting vascular function could be key to avoiding and treating Alzheimer’s.

There is a “clear and present” danger facing this year’s Match – particularly radiology residency programs, experts said this week in an opinion article in the 

. As the second interview season without in-person interviewing rolls around, medical students are likely to apply to more programs because there will be no travel cost involved if they win an interview spot. In the past, 41 percent of students reported foregoing an application because it was too expensive to travel for the interview. With this obstacle removed, radiology programs could face a problem. Now, more students who view radiology as a “back-up” option instead of being truly vested in the specialty could apply, potentially taking interview spots from students who are actually interested in a career in radiology. To help avoid this problem, a group of students launched SingleToken, the Residency Application Preference Signaling Platform. Functional now, this platform lets students pay $25 dollars to register their interest in the 12 top programs of their choice, and residency programs can opt to participate at no cost. The goal, the authors said, is to make a dent in the increased application congestion that could form during the interview season. And, at the very least, they advised, programs that choose not to take part in SingleToken should increase the number of applicants they interview. This year, they said, radiology residencies cannot afford to be complacent.

spoke with Dr. Christoph Lee about the trends and innovations surfacing in breast imaging. Dr. Lee, who is a breast radiologist at Seattle Cancer Care Alliance and a professor of radiology at the University of Washington, discussed the latest developments with modalities, as well as artificial intelligence. Here’s what he had to say.

Videos

Opposition to AI in Mammography; Alzheimer's Disease and Vascular Dysfunction; Clear & Present Danger for The Match; Trends & Innovations in Breast Imaging

Diagnostic Imaging Weekly Scan: Oct. 16, 2020

Radiology extenders who read chest X-rays save attending radiologists more time during the day than radiology residents do, potentially streamlining workflow and alleviating provider burnout.

At least that has been the experience for researchers at the University of Pennsylvania. Radiologists in their department read more cases per hour when the drafts came from radiology extenders than from residents, resulting in nearly an hour – 51 minutes – of provider time saved each day. The authors shared their experience on Oct. 13 in the 

“Interpreting these radiographs entails a disproportionate amount of work (eg., retrieving patient history, completing standard dictation templates, and ensuring proper communication of important findings before finalization of reports). Given low reimbursement rates for these studies, economic necessities push radiologists to provide faster interpretations, contributing to burnout,” said the team led by Arijitt Borthakur, MBA, Ph.D., senior research investigator in the Perelman School of Medicine radiology department. “Deployment of [radiology extenders] in the thoracic imaging section of our academic radiology practice improved efficiency of chest radiography.”

Taking these more routine exams off of the radiologist’s plate can also help improve his or her sense of job satisfaction, the team pointed out.

To determine the kind of impact radiology extenders would have on daily workflow, Borthakur’s team recruited two radiologic technologists to be radiology extenders and trained them over two months to interpret one- or two-view chest radiographs and to dictate using standardized templates. The extenders subsequently drafted an average of between 130 and 160 cases each day for six months. These cases were compared with those created by radiology residents.

Attending radiologists used the RADPEER scoring system to judge the quality of the drafted cases, assigning each case a score of 1, 2, or 3. A 1 score indicated agreement with the interpretation; a 2 score referred to a discrepancy with an understandable miss; and a 3 score indicated a discrepancy with a miss that should not have occurred. The RADPEER scores between the two groups were largely similar, and the team said any variability was likely due to case mix more than whether the scan was read by an extender or a resident.

Artificial Intelligence vs. Radiology Residents: Who Reads Chest X-rays Better?

But, according to their analysis, radiology extender-drafted cases were finalized faster than those drafted by residents. On average, four attending radiologists were able to finalize 93.6 cases per hour when reviewing cases from radiology extenders, but only 72.6 cases when reviewing resident-drafted cases. When they worked independently, they read just 62.4 cases per hour, +/ 10.2 cases.

As a result, the team said, radiology extender-drafted cases improved the flow rate by 52 percent compared to just 17 percent by resident-drafted cases. In addition, attending radiologists were also able to finalize radiology extender cases 36 percent faster than they were resident cases. This difference, they said, suggests that resident cases needed more edits.

The increased productivity can also have a direct impact on how departments organize their workflow and schedules. The team recommended strategically scheduling radiology extenders to work during times when study volumes are higher, potentially leading to more efficiency gains.

“We are confident that the [radiology extender] job function makes more business sense if radiology extenders are deployed in a strategic and meaningful manner,” they said.

Alongside accelerating study interpretations, the team pointed out that radiology extenders can also streamline workflow by stepping in to field reading room phone calls and communicate previous reporting findings to referring physicians. And, as artificial intelligence continues to assume a larger role in patient care, such as identifying lung nodules on CT or distinguishing COVID-19 pneumonia from other types, the radiology extender role has the ability to grow and morph with that expansion.

“We foresee the [radiology extender] role evolving in conjunction to serve as a data curator,” they said, “leveraging their technical expertise to assess image quality and guide reacquisition and performing hyperparameter tuning during model training to accelerate the adoption of machine learning algorithms in radiology workflow.”

Cases drafted by radiology extenders saved attending radiologists nearly one hour a day.

Radiology Extenders Outperform Radiology Residents with Chest X-ray Interpretations

Current lung cancer screening guidelines could be overlooking African American smokers who are at risk for developing lung cancer.

According to a study published in the Sept. 15 

 the current guideline based on the 30-pack-year threshold sets the bar too high, as most African American smokers self-report a less than 30 pack-year history. This determination sounds an alarm bell because African Americans have a high incidence of and worse outcomes from lung cancer.

“There is growing concern that there are healthcare disparities in lung cancer screening guidelines and programs,” said lead study author Anupam Basu, M.D., a diagnostic radiologist in Cook County Health. “Evidence suggests that the National Lung Screening Trial (NLST) and Preventive Services Take Force inclusion threshold of a 30-pack-year smoking history excludes a portion of African American patients at risk for lung cancer.”

Even though African Americans are more significantly impacted by lung cancer, the team said, they still only account for 4 percent of the NLST. Based on existing evidence this group does have a smoking duration history similar to their white counterparts – but, they actually smoke fewer packs a day. Consequently, they have a shorter pack history – the benchmark providers use to determine who should undergo lung cancer screening.

The team, therefore, suggested lung cancer screenings would be much more accurate and impactful if they adopted a 20-pack-year threshold.

To subscribe to The Reading Room podcast, click 

To identify a threshold that could better identify more African American smokers who could benefit from lung cancer screening, the team retrospectively reviewed low-dose CT scans from patients in their public hospital between September 2017 to December 2019. 

All total, they reviewed data from 784 patients, 57.5 percent of whom were men with an average age of 62. The majority were African American (66.2 percent), followed by white (20.4 percent), Hispanic (8.3 percent), Asian (4.8 percent), and Native American (0.3 percent).

Of the group, three-quarters – 75.5 percent – were active smokers, and just more than half – 51.5 percent revealed a pack history of fewer than 30 years. Total years smoked did not vary much by race, the team said. But, the African American experience was somewhat different, and it painted a grim picture.

“African Americans smoked fewer packs per day and had fewer pack-years compared with other races,” the team said. “African Americans were more likely to be active smokers but also more likely to report [less than] 30 pack-years compared with other races.”

Low-dose CT findings determined that lung cancer incidence did not vary by race, but African American had an incidence of 2.5 percent while whites experience 1.9 percent. Fortunately, the majority of cancers identified (75 percent) were Stage 1, and all patients had the cancerous tissues successfully removed with surgery.

Looking at lung cancer diagnosis specifically in the context of smoking brought more bad news. African Americans were the only race diagnosed with lung cancer when their self-reported pack-year history was less than 30. For smokers in this group with more than 30-pack-year histories, the incidence rate was 2.2 percent. For those with less than 30 years, it was 2.7 percent.

“We are concerned that the 30-pack-year threshold in the African American population may exclude a portion of patients who are still at risk for lung cancer,” the team said. “We propose prospective study of African Americans with [less than] 30-pack-year histories to determine the potential benefits and harms of [lung cancer screening] in this unique underserved population.”

Podcasts

In this podcast, Dr. Anupam Basu from Cook County Health in Chicago discusses the 30-pack-year threshold for lung cancer screening that overlooks at-risk African American smokers.

The Reading Room: Lung Cancer Screening Misses At-Risk African American Smokers

Dr. Long is a breast imaging radiologist with Raleigh Radiology in North Carolina.

Although the importance of breast cancer screening should always be emphasized, the current issues surrounding the coronavirus disease 2019 (COVID-19) pandemic have led to a decrease in the number of women who are continuing to follow their usual mammogram schedule.

According to experts, neglecting to have vital cancer screenings puts an individual at a greater risk. Moreover, recent research has shown that, at the onset of the COVID-19 pandemic, nearly 50 percent of the American women who are breast cancer survivors have experienced disruptions in their care, revealing critical flaws in disaster preparedness within the nation’s current healthcare system.

Study Finds Delays in Breast Cancer Treatment During Pandemic

To determine how the COVID-19 pandemic has affected breast cancer survivors' abilities to continue their cancer treatment, two University of Chicago providers created a questionnaire that  focused on the prevalence of the delays that these survivors experienced.

The questionnaire was distributed to breast cancer survivors in the United States vis email and social media. All total, 608 women, with an average age of 47, and one man responded to the survey. 

Respondents were asked if their appointments were re-scheduled for a later date even after they received a breast cancer diagnosis, as well as which aspects of their care were being delayed. Many patients stated that, rather than having their treatments canceled, they were modified in some way.

Patient-Reported Treatment Delays in Breast Cancer Care During the COVID-19 Pandemic

 study results, approximately 44 percent of responders reported delays in their care with diagnostic imaging tests and follow-up office appointments being the two most frequently postponed services.

Breast Cancer Treatments and Care Negatively Affected

Within the group of responders, 63 percent received cancer treatment, and many reported the pandemic had a negative effect on the care they received:

Nearly 80 percent stated that they had to re-schedule their doctor's office visits (routine).

Almost 70 percent who were scheduled to have reconstruction surgery needed to re-schedule their procedure.

For imaging, 60 percent were re-scheduled.

Fifty percent had to delay their lab tests.

In addition, 30 percent of respondents said they experienced a pause in their therapy at a clinic or hospital: 

Surgery for the removal of a tumor/lump (26 percent)

Consequently, the pandemic has had a serious impact on diagnosing, treating, and researching cancer. And, it is has created a serious cause for concern.

According to Philip Salem, M.D., president of the Salem Oncology Center in the Texas Medical Center, the pandemic has significantly impacted the cancer community. Due to their fears of catching COVID-19, many people are neglecting their cancer screenings.

However, with cancer being much deadlier than COVID-19, the director emeritus of cancer research at Baylor Medical Center said, people actively undergoing cancer treatment, as well as those who need their breast cancer screening, should not fear contracting the COVID-19 virus while at the hospital. Precautions against transmission are being taken.

Multi-disciplinary Recommendations for Treating Breast Cancer During a Pandemic

When considering disaster preparedness, there is a serious gap present in the U.S. healthcare system. This gap has led to the creation of recommendations for treating breast cancer patients during a pandemic.

 outlines the recommendations drafted by a group of expert medical societies, including the American College of Radiology and several other groups, for breast cancer treatment during the current pandemic.

These guidelines are designed to modify patient care in ways that will minimize the risk of exposure to the virus while preserving the resources necessary for treating breast cancer patients that require the most immediate care.

New Recommendations for Prioritizing Patients

This multi-disciplinary group recommends categorizing patients with breast cancer into one of the treatment priority levels (A, B or C):

Patients with symptomatic, life-threatening, completely intolerable or clinically unstable conditions fall within this category. In addition, patients whose prognosis would be altered significantly in the event of a short delay are included in the Priority A category.

This category includes most breast cancer patients. These women are not currently experiencing a life-threatening condition; however, they do require services or treatment in the near future. Care for these patients can be delayed, just not indefinitely.

The overall thought is that, if a geographic location only has enough resources available to accommodate Priority A patients, treatment for Priority B patients can be postponed from six weeks to 12 weeks without a negative impact on the patient’s health.

Certain services and treatments for these patients can be delayed indefinitely without any concerns about negatively affecting the patient’s outcome. To view recommendations for this group in their entirety, 

Overall, following these recommendations will make it easier to treat patients with breast cancer during this pandemic, or any other type of disaster that may arise. By categorizing these patients according to their current condition, medical professionals can provide breast cancer care strategically, undoubtedly saving lives.

Delays in the diagnosis and treatment of breast cancer have led to the creation of multidisciplinary recommendations for breast cancer care during the COVID-19 pandemic. 

Fewer Breast Cancers Being Diagnosed During COVID-19 Pandemic 

Artificial intelligence (AI) algorithms designed to independently interpret mammograms might be gaining ground when it comes to radiologist confidence, but the public-at-large does not share this sentiment. In fact, nearly 80 percent of women say they oppose letting AI interpret their studies without a radiologist being directly involved somehow.

a team of investigators from the University of Groningen, in The Netherlands, revealed that, when asked whether they were comfortable with having AI alone analyze their mammograms for signs of breast cancer, 78 percent of women surveyed said no.

“Despite recent breakthroughs in the diagnostic performance of AI algorithms for the interpretation of screening mammograms, the general population currently does not support a fully independent use of such systems without involving a radiologist,” said the team led by Yfke Ongena, Ph.D. assistant professor with the university’s Center of Language and Cognition.

Artificial Intelligence & Mammography: Where It’s Been & Where It’s Going

Given those advancements and study results that show AI can outperform radiologists, this response is surprising, they said. But, it shows that radiology as an industry should pump the brakes on implementing any AI algorithms as the sole interpreter of mammograms, as well as actively work to educate women about the capabilities and efficacy of these algorithms. Ultimately, improving the public’s perceptions about AI is critical because the industry currently has a shortage of radiologists in many regions who are qualified to interpret mammograms, the team pointed out.

The team’s survey results also showed when it comes to using AI as a second read on select studies, 42 percent disagreed, 31 percent agreed, and 27 percent were undecided.

Ongena’s team reached their conclusion after conducting a survey of 922 women, ages 16 to 75, in two waves. They included women ages 16 to 40 because they represent the next generation of women who will undergo routine screening mammography. First, they collected responses from December 2018 and, then, from April 2020. The team asked women for their feeling about the necessity of a human check, AI as a second reader, AI as a selector for second reading, and whether developers or radiologists were responsible for errors.

Their goal was to determine how women actually feel about having AI interpret their mammograms in light of how the mainstream press has portrayed these tools in a largely positive manner. They initially postulated that women’s responses would be in alignment.

Based on the survey responses, they were wrong. Still, the team concluded, there is room for AI to gain acceptance as a second reader that supports the radiologist. Only 17 percent of women categorically opposed using the tool in collaboration with her radiologist.

“The combination of a radiologist as a first read and an AI system as a second reader seems to be the most acceptable approach to the population at present, although still not fully embraced by the entire population,” they said. “Improved information supply and education about the development, possibilities, and limitations of AI algorithms in screening mammography may potentially overcome some of the perceived obstacles and increase acceptance of this new technique in clinical practice.”

The most important step to improving the acceptance of AI in the general population, they said, is bolstering patient education about the possibilities of the technology. On the whole, women who opposed the use of AI also reported having lower level of education.

The study did have a significant limitation, however, the team reported. Survey responses were collected in a country where mammography is routinely offered to women between ages 50 and 75, and those scans are interpreted by at least two radiologists. Other countries might not have the resources available to process mammograms at this level, and in the case of the United States, breast cancer screening is only routinely covered by Medicare.

Overall, the team said, it is critically important for women’s opinions of these tools to be considered.

“The voice of the population who will undergo AI-based diagnostic tests is crucial in this context, because it is a determining factor for the boundaries within which an AI system is allowed to operate,” they said. “The success of a breast cancer screening program depends on the willingness of subjects to participate, and the willingness may be affected if AI systems are used without taking into account the population’s wishes, concerns, and objections.”

More than three-quarters of women prefer a radiologist to be involved with reading their screening studies.



Most Women Reject AI As the Only Interpreter of Their Mammograms

Third-trimester ultrasound is typically used to determine whether a fetus is large for its gestational age, potentially predicting shoulder dystocia, one of the most common complications associated with delivering large babies. But, new research is casting doubt on whether the scan can reliably make this assessment.

Associated with infants who weigh more than 8 lbs. 13 oz. at birth (macrosomia), shoulder dystocia is a major perinatal complication. Ideally, an ultrasound would be able to pinpoint a fetus that will likely have a larger birth weight to help avoid should dystocia. But, in a study published on Oct. 13 in 

 a team of investigators from the University of Cambridge reported that ultrasound fails consistently do so.

“We conclude that ultrasonically suspected [large for gestational age] in the general population has quite good diagnostic effectiveness for macrosomic birth weight,” said the team led by Alexandros Moraitis, M.D., an obstetrics and gynecology trainee at the University of Cambridge. “However, it is not strongly predictive of the risk of associated complications, such as shoulder dystocia.”

Moraitis’s team made the determination after conducting a systematic review of 112,000 pregnancies. Their evaluation showed that, while prenatal ultrasound is highly effective in identifying whether a fetus is large for its gestational age, the scan is only weakly associated with correctly predicting should dystocia. This is particularly true for women who have low- and medium-risk pregnancies.

In particular, the team examined 41 studies that looked at whether size measurements captured via ultrasound at 24 weeks gestation could be effectively predictive. The team avoided published studies that focused on high-risk pregnancies as they are the focus of another ongoing U.K. trial.

Based on their evaluations, the team discovered that ultrasound-derived size measurements – calculated either with abdominal circumference or estimated fetal weight – can reliably determine which women with low- or medium-risk pregnancies will have a baby with a large birth weight. They also identified a significant relationship between shoulder dystocia and large size measurements identified on ultrasound, but the relationship is weak. Large estimated fetal weight is linked to only a 15-percent greater probability of this complication, they said.

Ultrasound in Third Trimester Doubles Detection Pre-Term Birth Risk

This finding, they added, underscores the actual incidence of shoulder dystocia.

“The fact that ultrasonic [estimated fetal weight] is relatively poor as a predictor of shoulder dystocia is not unexpected, given that the actual birth weight of the baby is also not strongly predictive of the outcome,” they said. “The majority of cases of shoulder dystocia involve a normal birth weight infant.”

The team also highlighted that simply having ultrasonic evidence of a large baby could lead to complications. Existing studies show that women who erroneously believe – due to ultrasound-derived measurements – that they will deliver a larger infant are more likely to undergo an emergency caesarean section.

“This finding underlines the potential for harm caused by screening low-risk women,” they said.

Moraitis’s team did acknowledge that the quality of the studies included in their review posed a limitation. Many were largely heterogenous and unblinded, and those unblinded studies indicated a stronger link between ultrasound weight predictions and the shoulder dystocia complication. Additionally, only a handful of reviewed studies examined the relationship between size predictions based on ultrasound and newborn morbidity, and they failed to find a link between the prediction of large birthweight and neonatal unit admission.

Overall, the team said, their findings do not support the implementation of universal screening for large birth weight.

“We recommend caution prior to introducing universal third trimester screening for macrosomia, as it would increase the rates of intervention, with potential iatrogenic harm, without clear evidence that it would reduce neonatal morbidity,” they said. “We believe that future studies should address the other factors which help differentiate those suspected [large for gestational age] fetuses which are at the greater risk of complications.”

Late pregnancy scans cannot reliably predict one of the most common delivery complications.



Third-Trimester Ultrasound Fetal Size Predictions in Doubt

There is a connection between impaired blood flow and the build-up of tau protein in patients who have Alzheimer’s disease, according to findings revealed on MRI and PET scans.

a team of investigators from the University of Southern California (USC) in Los Angeles shed more light onto the vascular component of Alzheimer’s, noting that their findings point to the potential of a combination treatment.

“Our results demonstrate vascular-tau association across the [Alzheimer’s disease] spectrum and suggest that early vascular-tau associations are exacerbated in the presence of amyloid, consistent with a two-hit model of [Alzheimer’s disease] on cognition,” said the team led by Daniel Albrecht, Ph.D., a neuroimaging research fellow. “Combination treatments targeting vascular health, as well as amyloid-β and tau levels, may preserve cognitive function more effectively than single-target therapies.”

To date, the connection between vascular dysfunction and tau pathology and how it affects cognition has not been well understood. Still, there is a growing body of evidence that shows vascular dysfunction plays a large part in Alzheimer’s pathophysiology.

“Like the proverbial chicken and egg, it remains unclear if impaired blood flow causes or is caused by errant protein building,” they said, “or if these two symptoms occur for unrelated reasons.”

 FDA Approves First Radiopharmaceutical for Imaging Tau

To dig deeper into this connection, Albrecht’s team used MRI and PET scans to compare blood flow and tau build-up in older adults. They examined 68 patients from USC who ranged from having normal cognition to showing signs of mild cognitive impairment, as well as a validation group of 138 individuals with mild cognitive impairment or Alzheimer’s disease who were enrolled in the Alzheimer’s disease Neuroimaging Initiative. Alongside their MRI and PET scans, all patients completed neuropsychological testing and were assessed for executive brain function, attention, and memory.

Based on their analysis, the team discovered that spots in the brain with more significant tau build-up also had decreased blood flow. This was particularly apparent in the inferior temporal gyrus – the area of the brain which typically experiences tau build-up in patients with Alzheimer’s disease. The team pointed out that this relationship also held true for the validation group.

“The correlation between tau and vascular function was stronger in people with greater cognitive impairment and higher amyloid-β levels,” they said. “It also appeared in more brain regions as the disease progressed in severity.”

Consequently, Albrecht’s team said, their results suggest that targeting vascular function could be key to avoiding and treatment Alzheimer’s.

“Results from the current study provide the first evidence of associations between elevated tau PET signal and vascular dysfunction,” they said. “Take together, combination treatments targeting vascular health, as well as amyloid-β and tau levels, may be more effective in preserving cognitive function than single-target therapies.”

As Alzheimer’s disease progresses, more brain regions exhibit evidence of this link.



MRI and PET Reveal Link Between Blood Flow and Tau Accumulation in Alzheimer’s Patients

Even though CT scans have not been widely used to diagnose COVID-19 infection, new research shows that using spectral CT with electron density (ED) imaging can make it much easier to assess the extent of lung lesion in patients who have early-stage disease.

Using a small group of case studies, published Oct. 21 in the 

a team of clinicians from Antony’s Private Hospital in France showed that electron density (ED) imaging can improve ground-glass opacity (GGO) visualization compared to conventional CT studies.

“ED imaging may improve the detection of early-stage COVID-19, a stage for which conventional CT has shown limited sensitivity for detection,” said lead study author Beatrice Daoud, an associate radiologist at Antony's. “Spectral CT may also provide a better assessment of the extent of the lung lesions.”

Deep Learning Plus CT Produces Spectral Images Without Higher Radiation

Although GGOs do show up on conventional CT scans, the visualization is subtle, Daoud’s team said, revealing only very small increases in attenuation compared with normal lungs. Consequently, accurate identification can be difficult, particularly during early infection. This study marks the first time investigators have looked into whether spectral CT scans can improve providers’ abilities to assess lung lesion extent.

To make this determination, the team retrospectively examined data from spectral CT scans conducted in four patients who have had confirmed COVID-19 cases with an RT-PCR test, they said. They reconstructed the spectral CT images using the same standard soft kernel (filter B) and a similar iterative method that is used to capture conventional CT images. They, then, compared initial conventional CT images with the follow-up conventional CT scans.

Based on their analysis, all 45 GGOs identified in the four patients showed up better on the ED images than the conventional CT scans, and the follow-up conventional CT images confirmed their presence. In addition, the lesion extent, which they assessed with semi-quantitative reporting scale denoting surface area involvement for each lobe, was much easier to determine on ED images, the team said.

“We reviewed conventional chest CT images obtained with a parenchyma kernel and standard lung window setting, as is usually the case in everyday radiology practices,” Daoud said.

The team compared the ED images to conventional images obtained using a soft mediastinum kernel and standard lung window setting, as well as conventional images captured with a soft mediastinum kernel and narrow lung window setting, virtual low-monoenergy images, virtual high-monoenergy images, and electron density images.

Based on these comparisons, the team discovered ED images made a difference in evaluating each of the four patients:

 A 39-year-old man diagnosed via conventional CT with lesions in both the right upper and lower lobes was found – with ED imaging – to have lesions in the upper left lobe that were barely noticeable previously.

 A 41-year-old woman with GGOs in both lower lobes discovered with conventional CT had additional lesions identified in the upper right lobe with ED imaging.

 A 64-year-old man had a conventional CT that showed peripheral bilateral diffuse GGOs. ED imaging revealed the GGOs to be more extensive and conspicuous than initially believed.

 An 80-year-old man had barely noticeable peripheral GGOs in the left upper lobe. Follow-up ED imaging showed the extent of these GGOs.

Overall, these results show that ED imaging offers greater detail for early assessment of GGOs that can be overlooked with conventional CT. The results are most promising, they said, by enhancing the contrast of GGOs compared with the normal lung.

“Our results suggest that the better ground-glass opacity visualization obtained using electron density imaging may be chiefly related to the increased visual noise in the image with soft kernel reconstruction and narrow lung window setting compared with electron density imaging, for which narrowing the window does not impair image quality,” they said.

Still, they said, a more systematic analysis of the different available image reconstruction is needed.

Spectral CT provides greater detail about the extent of ground-glass opacities than conventional CT scans.

Spectral CT Makes Detecting Early-Stage COVID-19 Easier

CT colonography (CTC) exams occur less often in patients over age 65, a new study has found, and this downward trend could affect racial and ethnic minority patients the most.

 a team of investigators from Emory University School of Medicine, led by Courtney Moreno, M.D., associate professor of radiology and imaging sciences in the Winship Cancer Institute, revealed the impetus for this drop could be because Medicare does not cover the exam.

“We observed that the relative utilization of screening CTC in individuals typically eligible for Medicare coverage decreased after age 65, suggesting access barriers related to a more restrictive Medicare coverage policy,” the team said.

Despite being a preventable condition with appropriate screening, colorectal cancer remains the second leading cause of cancer death in the United States. CTC is designed to detect colorectal cancer, as well as other types of colon cancer, but the Center for Medicare & Medicaid Services (CMS) announced in 2009 that evidence on the efficacy of CTC was lacking.

CT Colonography Has Higher Cancer Detection Rate Among S

Given the agency’s position, Moreno’s team set out to determine whether this stance is detrimentally impacting the quality of care given to patients. To do so, they examined CTC use, pulling 12,648 screening exams from the American College of Radiology’s CTC registry. They looked at both the age of patients, as well as any association between use of the exam and Medicare-eligible patient race.

Based on their analysis, the team identified a significant uptick in colon cancer rates in patients over age 50 – an annual increase of 5.3 percent in individuals between ages 52 and 64. On the flip-side, though, CTC screening rates began a noticeable decline after age 65, falling off by 6.9 percent for each additional year of age.

In addition, they noticed a difference in the screening rates between white and African American patients. For white patients, the CTC screening rate was highest between ages 65 and 69, but the test was administered most often in African American patients between ages 55 and 59.

This difference, the team said, could be attributed to Medicare’s lack of CTC coverage for patients over age 65. The average out-of-pocket expense for CTC screening, according to Costhelper.com, is $2,400, they said. Consequently, by not covering the test, the agency is likely making it harder for patients who cannot afford to pay for the test without insurance to secure the screening.

The factors behind this disparity are largely social and structural in nature, the team concluded, and they equate to a reduced ability to afford effective screening services. Not only is the average annual income lower for African American patients than it is for white patients, but African American Medicare beneficiaries also have, on average, lower levels of savings.

Due to these factors, the team determined, Medicare should change its course and move to cover CTC.

“Medicare coverage of screening CTC is needed so that Medicare patients who cannot afford to pay for this test out of pocket can undergo screening CTC,” they said.

African American, as well as other racial and ethnic minority, patients likely have lower levels of CTC screening due to out-of-pocket costs.

Lack of Medicare Coverage Limits CT Colongraphy Use in Minority Patients

New York-based start-up Ezra, a company focused on early cancer screening with MRI, has secured 510(k) clearance for its artificial intelligence (AI) tool for prostate cancer screening.

Algorithm cleared to help radiologists analyze and segment prostate MRI.