After taking a nearly 70-percent hit to imaging volume during the COVID-19 pandemic, mammography service lines are beginning to churn toward full volume again. 

Best Practices for Safely Reopening Mammography Services

Since the COVID-19 pandemic began, questions have been swirling about any similarities or differences between COVID-19 pneumonia and pneumonia linked to the flu. A new study, based on findings from CT scans, now shows that, depending on the type of pneumonia, lesions appear differently in the lungs.

 investigators from China’s First Affiliated Hospital of Wenzhou Medical University revealed, with the help of artificial intelligence software, that most COVID-19 lesions appear in the peripheral zone close to the pleura. In contrast, lesions from influenza virus pneumonia are more apt to show mucoid impaction and pleural effusion.

These details highlight the supplemental role this knowledge can play in helping radiologists identify which patients are fighting COVDI-19. Even with these findings, though, the team, led by Liaoyi Lin, from the First Affiliated Hospital’s radiology department, cautioned that it will still be tough to differentiate between the two pneumonia types.

“CT manifestations of COVID-19 pneumonia and influence virus pneumonia have a large amount of overlap, such that even with the characteristics evaluated using AI software, no significant differences were detected,” they wrote. “Distinguishing between these two types of viral pneumonia with imaging alone is difficult.”

To pinpoint any differences between pneumonia caused by COVID-19 and the influenza virus, the team conducted a retrospective analysis with 97 patients – 49 women and 48 men – who had positive nucleic acid test results for the respective viruses. Their patient records also included CT images and detailed clinical information. Of the group, 52 patients had COVID-19 pneumonia and 45 had influenza virus pneumonia.

Overall, the team did not observe any significant differences in CT scores, length of the largest lesion, mean density, volume, of lesion mass. But, while most features overlap, they said, some statistical differences did emerge.

“Between the group of patients with COVID-19 pneumonia and the group of patients with influenza virus pneumonia, the largest lesion close to the pleura (i.e., no pulmonary parenchyma between the lesion and the pleura), mucoid impaction, presence of pleural effusion, and axial distribution showed statistical difference,” the team wrote.

For axial distribution, the team determined, for COVID-19 pneumonia, it was mostly in the peripheral zone (67.3 percent). Lesions in influenza virus pneumonia had a distribution that was more diffuse (35.6 percent) and random (33.3 percent). The difference in axial distribution suggests, they said, that COVID-19 pneumonia might originate from the subpleural mesenchyme and gradually spread inward.

More specifically, this is what they found:

 For patients with COVID-19 infection, the most common main CT feature of the largest lesion was ground-glass opacity (GGO), present in 30 out of 52 patients (57.7 percent), and the largest lesions were close to the pleura in 36 patients (69.2 percent).

Consolidation, 28 patients (53.8 percent)

Mosaic attenuation, 10 patients (19.2 percent)

 Bronchial wall thickening, 6 patients (11.5 percent)

Centrilobular nodules, 12 patients (23.1 percent)

Interlobular septal thickening, 14 patients (26.9 percent)

Crazy paving pattern, 7 patients (13.5 percent)

Air bronchogram, 18 patients (34.6 percent)

 GGO was also the most common main CT feature of the largest lesion, appearing in 21 out of 45 patients (46.7 percent), with the largest lesions, again, appearing close to the pleura in 18 patients (40 percent). Still, most lesions were not close to the pleura, the team pointed out. Additionally, they noted that this viral pneumonia has more mucous composition and pleural effusion, likely because the influenza virus affects the large and small airways and lung parenchyma.

Consolidation, 26 patients (57.8 percent)

Mosaic attenuation, 10 patients (22.2 percent)

Bronchial wall thickening, 7 patients (15.6 percent)

Centrilobular nodules, 5 patients (11.1 percent)

Interlobular septal thickening, 10 patients (22.2 percent)

Crazy paving pattern, 6 patients (13.3 percent)

Air bronchogram, 17 patients (37.8 percent)

Mucoid impaction, 6 patients (13.3 percent)

Pleural effusion., 10 patients (22.2 percent)

There were also some differences in lesion distribution among patients, the team said. Influenza virus pneumonia patients presented more unilateral lung lesions distribution than did COVID-19 patients – 44.4 percent and 26.9 percent, respectively. However, COVID-19 patient exhibited more bilateral distribution – 73.1 percent compared to 55.6 percent.

Still, even with these differences in findings, the team noted the existing similarities between the two pneumonias make imaging alone an insufficient option for distinguishing between the two.

“CT examinations needs to be combined with clinical indicators for comprehensive evaluation,” they said. “The more important role for CT in the pandemic is in finding lesions and evaluating effects of treatment.”

Articles

Scans reveal differences in lesion locations, mucoid impaction, and pleural effusion.

CT Reveals Differences Between COVID-19 and Influenza Pneumonias

 Weekly Scan. I’m senior editor, Whitney Palmer, bringing you some of the most popular radiology news of the week.

As we enter the fourth full month of the COVID-19 pandemic, radiology practices, in many ways, are still trying to find the best workflow plans that will allow for quick, accurate, and standardized diagnostics. Although each practice must find the strategy that works best for them, radiologists from a 48-center network in Moscow shared the facets of a pre-existing plan that has given way to better workload management and better COVID-19 outcomes. Specifically, they laid out nine steps: infrastructure preparation, including digital equipment and uniform resource identifiers operation, quality training of radiologists and X-ray technologists, the development of methodological materials and diagnostic protocols based on international experience, fast learning through distance education, new protocol integration into the RIS, regular discussion of complicated cases either online or at conferences, using a special dashboard to monitor performance, using a remote study reporting system, and implementing AI algorithms. They also divide their radiology departments into green and red zones. Radiologists who read reports with no face-to-face contact with patients remain in the clean green zone. Radiographers, medical volunteers who greet and escort patients, and workers who disinfect the equipment remain in the red zone. Through this system, the radiologists said, images of a patient with suspected COVID-19 infection can be captured, analyzed, and reported within 20 minutes, giving clinicians the ability to isolate the patient quickly, if needed.

Radiology investigators also revealed some important findings about breast imaging this week. In the 

 investigators from Boston’s Children’s Hospital determined that annual breast cancer screenings for women who were survivors of a pediatric cancer cuts the death rate in this group by more than half. Specifically, they recommended beginning annual MRI screening, with or without mammography, at the age of 25. Overall, women in this group have a 10-percent to 11-percent risk of dying from breast cancer. Implementing the routine screening sooner rather than later slices that risk of mortality by 56 percent to 71 percent. The researchers did say, though, that postponing the screenings until age 30 might be more cost effective for the patient, and it would not have any deleterious effects on the patient’s health or outcomes.

In addition, radiologists from Massachusetts General Hospital published a study this week in the 

, detailing that pre-operative breast MRI can catch 11 percent more cancers in women who have ductal carcinoma 

. Even though breast MRI is more sensitive than mammography and does a better job of identifying both calcified and non-calcified DCIS better, whether to use the scan is still up for debate. Consequently, the team set out to determine whether using pre-operative breast MRI would have a positive impact on surgical planning and outcomes. They looked at data from 963 women who had core biopsy-confirmed DCIS, 236 of whom also had breast MRI. Among that group, they found an additional cancer identification rate of 11.2 percent. They did find that women who received breast MRI did have more mastectomies at initial surgery, but the team cautioned that it could be due to non-MRI related factors, such as patient age, breast density, personal breast cancer history, and ordering physician assessment. In addition, women who had the scan had a similar re-excision rate compared to women who did not have breast MRI. Overall, the team said, these findings point to the positive role pre-operative breast MRI can play in surgical decision-making and overall treatment planning.

New findings about the impact of COVID-19 on the body continue to roll in. This time, a study, published in 

 has revealed that COVID-19-positive patients are more likely to experience a lung injury while ventilated than are patients who do not have the viral infection. According to the study that evaluated 601 intubated COVID-19-positive patients, as well as 285 intubated patients with acute respiratory distress syndrome, 15 percent of the COVID-19 patients experienced at last one barotrauma – an injury associated with increase air pressure in the lungs. They found 31 patients with left pneumothorax, 35 with right pneumothorax, 11 with bilateral pneumothoraces, 54 with either unilateral or bilateral pneumothorax, 59 with pneumomediastinum, and 14 with pneumopericardium. In contrast, only 11 percent of the acute respiratory distress patients developed a barotrauma. Ultimately, the team said, they hope their findings support emerging theories around lung damage from COVID-19 infection.

 spoke this week with Dr. Richard Duszak, professor and vice chair for health policy and practice in the radiology and imaging sciences department at Emory University, and Dr. Danny Hughes, executive director and senior research fellow at the Harvey L. Neiman Health Policy Institute, about their research published this week in the 

Journal of the American College of Radiology

 that examined the decline on non-invasive diagnostic imaging completed in community practices during the pandemic. They discussed some of the specific findings of the work, as well as the implications the results could have on workflow and financial planning as the pandemic continues. Here’s what they shared.

Videos

Diagnostic Imaging's Weekly Scan; July 10, 2020

9-Point COVID-19 Prep Plan, Breast MRI, COVID-19 Lung Injury, and the Pandemic's Community Imaging Impact

When is a shipping container not a shipping container? When it is mobile CT or X-ray pod.

In an effort to keep radiologists, technologists, and patients as safe as possible while still conducting necessary imaging for the growing number of patients who are testing positive for COVID-19, radiology vendor Philips is outfitting shipping containers to be used as imaging pods in the Philippines. These units will be mobile and can be placed where they are needed most – within hospitals, on facility grounds, or in the community, company officials said.

Plans are currently in place, according to company details, for officials in the Philippines to begin using these CT and X-ray cabins to triage patients. The individual nature of these pods means that healthcare providers can instantly isolate a patient if their scans point to COVID-19 infection.

 Walk-Up X-ray Booth Offers Quick Scans to COVID-19 Patients

“Limiting the chance of cross-infection is especially important during triage, when infected and uninfected patients share the same diagnostic pathway,” company officials said. “When you add in the need to upscale and deploy diagnostic facilities in an agile way to cope with increasing patient numbers and localized COVID-19 outbreaks, it makes sense to find new ways to deploy and install diagnostic solutions.”

Each cabin will be equipped with Philips CT solutions, including Access CT and Incisive CT, or X-ray solutions, including DuraDiagnost and DuraDiagnost F30. Additionally, every cabin will have an integral lead shield to reduce stray radiation, as well as ultraviolet lamps that can be used to sterilize the workspace between patients and a laboratory-grade computer for in-the-moment results analysis.

Everything, company officials said, will be configured to let radiologists conduct diagnostic imaging procedures with as little patient contact as possible.

It is possible, company officials siad, that these pods will have functionality beyond the COVID-19 pandemic. They could also potentially be used for emergency rescue, as well as in disaster relief zones.

Philips thinks outside-the-box by turning shipping containers into individual CT and X-ray scan rooms.

Imaging-in-a-Box

The neurological outcomes for COVID-19 patients who are unresponsive might not be as bleak as previously thought.

According to a case study, published by investigators at Massachusetts General Hospital (MGH) last week in the 

 it is possible these patients have a better shot at functional recovery than standard imaging alone would indicate.

Brain Alterations Seen Via MRI in COVID-19-Positive Technologist

By using resting-state functional magnetic resonance imaging (rs-fMRI), the team, led by Brian Edlow, M.D., director of the Laboratory for NeuroImaging of Coma and Consciousness and associate director of the Center for Neurotechnology and Neurorecovery at MGH, discovered that an unconscious COVID-19-positive patient had “robust functional connectivity” in the default mode network (DMN), a brain network that is believed to be involved in human consciousness.

"Because there are so many unanswered questions about the potential for recovery in unresponsive patients who have survived severe COVID-19, any available data that could inform prognosis are critical," Edlow said. "Our unexpected observations do not prove that functional MRI predicts outcomes in these patients, but they suggest that clinicians should consider the possibility that unresponsive survivors of severe COVID-19 may have intact brain networks. We should, thus, exercise caution before presuming a poor neurologic outcome based on our conventional tests."

Having a greater understanding -- and being able to provide some neurological context around possible prognosis -- could be helpful for patients' families as they are trying to make healthcare decisions for a loved one, added lead author David Fischer, M.D., neurocritical care fellow at MGH.

The patient, a 47-year-old man with hypertension and asthma, presented to the hospital with fever and shortness of breath. He was placed on a ventilator, and for roughly 60 days, he floated in and out of a coma and a minimally conscious state. On the 40th day of his hospitalization, he underwent both 3T structural and functional MRI scans.

Acute Stroke Predicts Negative Outcome in COVID-19 Patients with Neuroimaging Findings

The structural scan revealed substantial damage that is often associated with hypoxic-ischemic brain injury – symmetric T2 hyperintensity of the bilateral basal ganglia, medial thalami, and parahippocampal gyri, and diffusion restriction of the basal ganglia. However, the rs-fMRI revealed a different picture. It showed his DMN connectivity was comparable to healthy patients, pointing to his potential for recovery, the team said.

Those positive outcomes began to emerge 20 days later, the team reported, as the patient began to follow commands, such as blinking his eyes and opening his mouth upon instruction, intermittently over a week timespan. By day 66, he could follow four out of four vocalization commands and could successfully complete gaze tracking to visual and auditory stimuli.

Ultimately, the team said, these results indicate that further investigation is needed into how the virus impacts unresponsive patients.

“The long-term outcomes for patients with disorders of consciousness following severe COVID-19…are unknown,” the team said. “Longitudinal cohort studies in this population will be crucial for understanding their disease course and developing prognostic biomarkers. Therefore, any available data that may inform prognosis in this novel disease are critical.”

Case study results show patient has brain connectivity similar to healthy individuals, pointing to positive prognosis possibilities.


Functional MRI Reveals Better Brain Activity in COVID-19 Patients That Previously Believed

Without question, the rapid spread of the COVID-19 pandemic pushed teleradiology into the spotlight. Practices that had previously dabbled in the technology rushed to expand their efforts, and those who had resisted altogether found themselves quickly trying to put plans in place. But, those practices and departments with established teleradiology systems discovered they were in the position to best weather this continuing storm.

St. Mary’s Hospital in London is one such facility. COVID-19 was not even a whispered concern when they implemented a pilot teleradiology program in June 2019. But, the system they created led to exponential growth in their productivity, setting them up for continued success during the outbreak.

The team published the results of their experience recently in 

To create the teleradiology program, St. Mary’s leaders reached out to and employed a 17-year veteran radiologist located in Sydney, Australia who was charged with overnight reads for the hospital and checking after-hours on-call reports from radiology registrars. The three-month initiative was designed to examine the change in reporting productivity over the four weeks prior to the pilot launch and during the study.

According to their analysis, the department saw a 140-percent spike in reporting productivity compared to the same number of in-hospital sessions that occurred during the four weeks before the pilot began. The remote reads included routine musculoskeletal and general CT and MRI studies. In addition, department leaders said they saw a 40-percent drop in the time needed for the consultant to authorize the report.

All of these benefits paved the way for the department to easily slip into more widespread remote reading for all providers once the pandemic hit, said the team led by E.A. Dick, a consultant radiologist with St. Mary’s larger organization Imperial College National Health Services Trust, and they indicated these changes could be here to stay.

COVID-19 Spotlights Need for Virtual Services, Including Teleradiology

“There has been a national realization across healthcare, as for other workplaces, that remote working is desirable for the duration of the pandemic and that remote working should be optimized to provide a service as good as that seen in traditional in-hospital Trust settings,” they wrote. “Additionally, there is a growing realization that working patterns will likely be permanently changed.”

Alongside the uptick in productivity, provider feedback was also positive. Most indicated they noticed improved service even though they simultaneously expressed disappointment that they could not easily communicate with the remote radiologist to discuss cases.

“A remote reporting service has to not only deliver a technologically seamless service, but also to deliver communication with the clinical teams at least as good as that given in person,” the team said. “The remote reporter has to work harder to maintain good communication – provide easy-to-access telephone, email, and message services – and, most of all, reach out to clinicians.”

Even with this drawback, though, the team said the pilot program, overall, clearly demonstrates the value and efficacy of a teleradiology program both for the current pandemic circumstances and beyond.

“This pilot study has relevance for all radiologists and radiology service providers in the COVID-19 and post-COVID-19 era, not only because it showed better prodcuvity and shorter time to consultant authorization,” said Dick’s team. “The positive feedback from clinicians and opportunities for increased support of registrars suggest potential benefits to training, and therefore, patient care.”

An existing program set one London hospital up for success in managing images remotely during COVID-19.

Teleradiology Veterans See 140-Percent Spike in Productivity

 (DCIS) can benefit from undergoing pre-operative breast MRI, new research has revealed. These scans pinpoint 11 percent more cancers, indicating the images could augment surgical planning and lead to better outcomes.

 investigators from Massachusetts General Hospital determined incorporating pre-operative breast MRI can positively impact the surgical management of women with DCIS.

“Pre-operative identification of invasive disease and delineation of disease extent in patients with DCIS is necessary to inform surgical decision-making,” wrote the team led by Leslie Lamb, M.D., radiology instructor at Massachusetts General Hospital. “For example, if invasive cancer is identified prior to surgery rather than during the initial surgery, then sentinel lymph node biopsy can be performed at the initial surgery.”

DCIS is typically first identified via mammography, and it accounts for roughly 20 percent of all breast cancers. But, because breast MRI has higher sensitivity than mammography and can find both calcified and non-calcified DCIS better, it is increasingly being used as a pre-operative tool with women who have invasive breast cancer.

However, using preoperative breast MRI is still a subject up for debate, Lamb’s team said. Alongside finding more cancers, it may also contribute to false-positive care need biopsies, overestimate the extent of disease compared to mammography, and prompt more mastectomies without lower re-excision rates.

Post-Mastectomy Bilateral Breast MRI Detects Asymptomatic Cancers

Consequently, to determine whether pre-operative breast MRI has positive effects on surgical planning and outcomes, Lamb’s team conducted a study with 963 women who had mammography-identified calcifications that were determined to be DCIS by core biopsy between 2007 and 2016. Of the group, 236 women (24.5 percent) also received breast MRI. These women shared three characteristics: they were younger; they were less likely to have a previous breast cancer; and they were more likely to have dense breast tissue.

Based on their analysis of women undergoing breast MRI, the team determined 66 women (28 percent) had 80 findings for which further work-up or biopsy was suggested. An additional 21 (8 percent) had 23 findings that did not receive follow-up, and 45 women (19) percent) had 57 findings that prompted biopsy alone. Among the women who had biopsies, 37.6 percent were benign, 17.5 percent indicated high-risk pathology, and 43.9 percent were malignant.

According to Lamb’s team, pre-operative MRI identified 26 cancers in 24 women – an additional cancer identification rate of 11.2 percent. Half were found to be invasive; half were DCIS. And, in reference to the concern over pre-operative MRI and mastectomies, the investigators discovered more women in the breast MRI group did have a mastectomy at initial surgery. However, they cautioned that rate could be due to more than just MRI results – there were other differences in the patient populations, as well, including patient age, breast density, personal breast cancer history, at ordering physician assessment.

Ultimately, in light of these results and a re-excision rate similar to women who did not undergo breast MRI, the team determined that pre-operative breast MRI should be arrow in the quiver for imaging and treatment women who receive a DCIS diagnosis.

“Ideally, pre-operative imaging with MRI, combined with other advanced diagnostic tools, can inform surgical decision-making and the overall treatment plan, in order to attain improved therapeutic precision,” the team concluded.

Incorporating this imaging catches 11 percent more cancers, potentially affecting surgery plans and outcomes.

Pre-Operative Breast MRI Diagnoses More Cancers in Women with DCIS

Annual breast cancer screenings can reduce mortality among women who were treated with chest radiotherapy for a childhood cancer. In fact, according to new research, implementing yearly screening as early as age 25 can cut breast cancer deaths by more than half.

 a multi-institution team of investigators compared the benefits, harms, and cost effectiveness of various breast cancer screening strategies among women who survived a childhood cancer. The study, funded by the U.S. National Institutes of Health and the American Cancer Society, is the first to provide a benefit-harm ratio estimate for this patient group.

The analysis from the team, led by Jennifer Yeh, Ph.D., an associate scientific research in the general pediatrics division in Boston Children’s Hospital, supports the recommendation from the Children’s Oncology Group (COG) – women who have previously undergone chest radiation benefit from having annual mammography or breast MRI screening starting at age 25.

“Our findings underscore the importance of making sure that young women previously treated with chest radiation are informed about their elevated breast cancer risk and the benefits of routine screening,” the team wrote. “Both primary care providers and oncologists who care for survivors should discuss breast cancer screening with these patients.”

According to existing research, there are 55,000 women who received chest radiotherapy at doses of 20 Gy or more to treat a pediatric cancer. Approximately 30 percent of this group will develop breast cancer by age 50 – a rate similar to women with the BRAC1 mutation.

To determine what breast cancer screening options offered the greatest benefit, the researchers examined data from two established breast cancer models from the Childhood Cancer Survivor Study and the Cancer Intervention and Surveillance Modeling Network. Using the models, they evaluated three strategies:

Digital mammography with MRI screening starting at age 25, 30, or 35, continuing to age 74

MRI-only screening starting at age 25, 30, or 35, continuing to age 74

Overall, women in this group had between a 10-percent and 11-percent lifetime risk of dying from breast cancer, and the authors determined that all of the annual screening strategies truncate that mortality rate by more than 50 percent.

Specifically, though, an annual mammogram paired with breast MRI starting at age 25 offers a 56-percent to 71-percent drop in mortality. The life-years gained were also significant – the screenings produced a rise from 1,317 to 2,544 per 1,000 women. In addition, the false-positive screening rate and number of benign biopsy results produced a harm-benefit ratio for these childhood cancer survivors that was better than that for women of average breast cancer risk.

But, when taking cost-effectiveness into consideration, the team discovered that delaying the mammography-MRI screening combination to age 30 might lower the cost to the patient. And, the impact on a patient’s quality of life from putting off screening for five years is marginal, they said.

The study did have limitations, however. Because there is a dearth of research on women who are pediatric cancer survivors, the team based its analysis of MRI and mammography performance on research conducted with women who were BRAC1 and BRAC 2 mutation carriers. Additionally, they did not take into account any risk for radiation-induced breast cancer from repeated mammography screening because the risk between the ages of 25 and 39 is small compared to the radiation doses women received during cancer treatment.

Ultimately, the team said, their findings highlight the importance of incorporating breast MRI into screening for these women.

“Identifying effective policies and interventions to reduce barriers to screening should be a priority for policymakers to ensure comprehensive and coordinated care for these high-risk survivors,” they said.

Annual MRI screening – with or without mammography – can prevent at least 50 percent of early deaths.

Women Who Survive Childhood Cancer Benefit From Earlier Breast Screening

The Radiological Society of North America (RSNA) announced Tuesday that it has approved $4 million in research and educational grants.

Through the RSNA’s Research and Education (R&E) Foundation Board of Trustees, the group funded 28 percent of grant applications with awards going to radiologists at all stages of their careers.

“We are deeply grateful for the ongoing and generous support of our individual, practice group, and corporate donors who make supporting these researchers and educators possible,” said Thomas M. Grist, M.D., chair of the R&E Foundation Board of Trustees. “Because of the generosity of our donors, the RSNA R&E Foundation is able to support research and education efforts that elevate our specialty and place radiology at the forefront of innovation in medicine.”

All total, the Foundation approved grants to 93 recipients from 47 different institutions across North America. Of the group, 49 are new grant recipients, 26 are new medical student grants, and 18 are continuing grants.

The Foundation also launched an initiative this year to provide additional grants with medical students who identify as being from racial and ethnic groups who are underserved in medicine (UIM). This year’s UIM grants will fund students conducting research on optimizing immune responses for liver and skin cancer patients, as well as investigating how artificial intelligence can be used to classify bone lesions.

More information about the RSNA R&E Foundation and the funded 2020 projects can be found 

The RSNA Research and Education Foundation awards 93 grants from 47 institutions. 


RSNA Announces $4 Million in Research, Education Grants

Artificial intelligence (AI) tools are not the Dooms Day machines for radiology that many naysayers would have you believe, said one industry expert.

In an opinion article published recently in 

 John Banja, Ph.D., professor and medical ethicist in the Center for Ethics at Emory University, addressed the often negative ballyhoo that has surrounded AI in radiology for several years – that the technology will soon replace the radiologist, putting you out of a job.

“What seems ethically imperative at present…is a steady and informed rebuttal of AI hype, especially as it is aimed at image-dependent technologies like radiology,” he said. “Today’s hospitals simply cannot function without radiologists, who are core to their diagnostic functions.”

It’s True – Interest in AI Is Growing

The past decade has seen an explosion of development around AI tools, Banja said, particularly in research. In 2008, there were only 100 publications in radiology literature that focused on AI tools specific to the industry. But, by 2018, that number had ballooned to approximately 700 papers.

In the same vein, investment in AI has also taken off – since 2010, the market has seen applications for 154,000 patents. Between in 2018 and 2019 alone, organizations that use AI expanded from 4 percent to 14 percent. And, according to the Worldwide Semiannual Artificial Intelligence Systems Spending Guide, the fastest AI growth between 2018 and 2022 is expected to be in healthcare with image recognition software being some of the most familiar models.

Radiology Leaders Urge Proceeding with Caution with Autonomous AI

But, that still does not mean AI is in a position to upset the radiologist.

“It is a long way from observing the success of an AI model in a research setting to implementing it in routine clinical practice,” he explained. “And, it is a much longer way still to replacing human radiologic expertise.”

Even though there is vast potential for AI in radiology, the technology still has some significant limitations. In most cases, the algorithms used in the industry have not been tested either on heterogeneous patient populations or across a variety of modalities. Consequently, the strength of their generalizability is in question.

This means the applications for AI tools are relatively narrow within radiology, lending the technology to be a radiologist’s “amiable apprentice” rather than an “awful adversary,” Banja said.

Additionally, AI’s current function is to identify potential problems and bring them to a radiologist’s attention. The next steps of decision-making, though, are outside its capabilities, keeping the radiologist’s position secure.

“Even if a plethora of models appeared that could detect and evaluate subtle lesions as well as an experienced radiologist,” he said, “those applications wouldn’t be able to determine whether to image or not, how to best provide protocol for an examination, teach, and communicate findings to clinicians and patients.”

It is possible that one of the biggest obstacles to wider – and more complex – AI use is technology itself, Banja said. According to Wim Naude, a business professor from The Netherlands, the long-standing pattern of computer power doubling roughly every two year is largely winding down, potentially making it harder for AI to advance to the point where it can take over the wide variety of tasks that fall to the average individual radiologist.

Currently, the technology needed to achieve this level of multi-tasking does not functionally exist, Banja said. Reaching the goal would mean replacing AI’s silicon-based transistors with technologies that are still nascent, such as organic biochips or carbon nanotubules.

Overall, he concluded, there are many unanswered questions about how radiology, as a whole, will be able to seamlessly mesh AI solutions into current workflows and successfully navigate many of the other issues that currently face the industry.

“It remains anyone’s guess as to how AI applications will be affected by their integration with PACS, how liability trends or regulatory efforts will affect AI, whether reimbursement for AI will justify its use, how mergers and acquisitions will affect AI implementation, and how well AI models will accommodate ethical requirements related to informed consent, privacy, and patient access,” Banja said.

Once these issues are addressed, he said, it is possible that, rather than succumb to an AI take-over, radiology could ride the wave to a new height of professional excellence.

One expert makes the case that the negative hype around artificial intelligence is not warranted and the radiologist’s role is secure.

AI Is Not the End of Radiology’s World

COVID-19 can present a myriad of challenges to the clinician, but adopting a methodical, practical approach can make it easier for radiologists and technologists to render services to these patients. A nine-step plan released by providers in Russia details how.

Between mid-April and mid-June 2020, a network of 48 outpatient CT facilities in Moscow conducted more than 170,000 CT scans and identified more than 82,000 cases of COVID-19 pneumonia. A pre-existing plan, developed between 2012 and 2018 at the Moscow Center for Diagnostics and Telemedicine (CDT), has made it simpler for radiologists to read remotely.

“The goals of this innovation were to reduce workload on medical facilities under the conditions of rapidly spreading disease,” explained Sergey Morozov, chief regional radiology officer of Moscow, “and to reduce mortality rates from COVID-19.”

, designed to foster quick, accurate, and standardized radiology diagnostics, includes:

Infrastructure preparation (digital equipment and uniform resource identifiers operation)

Quality training of radiologists and X-ray technicians

Development of methodological materials and diagnostic protocols based on international experience

Integration of new protocols into the RIS in the form of study-report templates

Regular discussion of complicated cases at online meetups and conferences

Using a special dashboard to monitor performance

Implementing AI algorithms to help doctors

In an effort to streamline image reading – and keep technologists and radiologists safe – the CDT divided departments into “green” and “red” zones. Radiologists who read reports and do not have face-to-face contact with patients stay in the clean “green” zone. The CT rooms and all the staff who work there – radiographers, medical volunteers who greet and escort patients, and workers who disinfect equipment – comprise the “red” zone. These staff members are outfitted with Class Three personal protective equipment, and those coming off shift are careful not to overlap duties with those coming on shift.

With more than 1,000 diagnostic devices currently connected to Moscow’s Unified Radiological Information Service (URIS), CDT has designed a workflow that streamlines the diagnostic process. When a radiographer performs a study, a personalized referral is created, and patient-specific images are uploaded to URIS. After the radiologist writes the report, URIS-connected clinicians can access both the report and images in real time.

“This interoperability between different facilities has proven to be invaluable in a pandemic and to isolation restrictions, when a transfer of physical image mediums, [such as CD or films,] became impossible,” Morozov said.

Within 20 minutes, he said, a patient with suspected COVID-19 infection could be imaged, and a referring physician could have a completed radiology report in hand. And, patients can be hospitalized without re-examination. Other benefits include reduced radiation exposure, shortened waiting times, and study reports available in any format.

As part of the effort to facilitate a smoother workflow and prevent further COVID-19 spread, CDT also launched a telemedicine program. This initiative has not only helped facilities that have a shortage of radiologists, but it has also benefited those environments where the number of studies is overwhelmingly high, Morozov said. In these situations, images with suspected COVID-19 are flagged and placed at the top of the worklist.

Given the large number of studies generated during the pandemic, artificial intelligence tools have been vitally important, Morozov explained.

URIS is equipped with algorithms to process CT scans as soon as they are received, and the radiologist has the opportunity to review both the original study and the one processed by artificial intelligence which includes pathological findings marked in red. The tools can also generate a report on spots of probable viral pneumonia, but final diagnostic and treatment conclusions still come from the radiologist, he said.

A network of nearly 50 CT facilities and the Moscow government reveal how their collaboration is working to promptly diagnose patients with COVID-19.


A 9-Step Plan to Confront COVID-19

COVID-19 has wreaked havoc on imaging volumes across radiology. New research reveals that community-based practices have felt the brunt of the loss, particularly with mammography screenings.

In a new study published this week in the 

 investigators examined imaging data from nine community-based practices nationwide. What they discovered is a crippling impact of the pandemic on providers – work relative value units (RVUs) dropped approximately 52 percent, falling from a zenith 120,000 in February to 55,000 in April.

While this analysis reveals a currently bleak picture, providers could also make the findings actionable, using them to craft recovery strategies as the industry continues to grapple with the pandemic.

“Our findings support widespread anecdotal, as well as early published reports of massive declines in imaging volumes and revenue that have led to great concern across radiology practices,” wrote lead study author Richard Duszak, M.D., professor and vice chair for health policy and practice in the radiology and imaging sciences department at Emory University. “With no end in sight of the current pandemic, such information could prove actionable for both radiologist workforce and practiced financial planning, particularly if a predicted potentially disastrous ‘second wave’ of coronavirus disease, indeed, materializes.”

To determine the impact on community practices, Duszak’s team evaluated imaging metadata collected between January 2019 and May 2020, looking at volume trends by modality, site of service, and body region.

According to their results, outpatient imaging fell by 66 percent between February and April of this year. When examined as a percentage of total all-practice work RVUs, CT and radiography/fluoroscopy took the biggest hits, dropping 31 percent and 30 percent, respectively. In addition, abdominal/pelvis imaging fell 25 percent, and breast imaging declined 19 percent, both experiencing the most significant body part drops.

Most significantly, the team pointed out, mammography plummeted the most across the different modality-regions groups. This service line saw a 92-percent plummet.

Alongside these changes in imaging volume, the team also pinpointed some downstream effects that could impact radiology is the effects of the pandemic continue or worsen. For example, they said, if revenues continue on a downward trajectory, the specialty could become less attractive to medical students looking to choose their career path or it could push some radiologists getting close to retirement age to pull the trigger early, leaving practices and departments without senior leadership.

From another financial and management perspective, this situation could further bolster the culture of consolidation already thriving in radiology, they said.

Ultimately, they said, this analysis could inform decision-making around radiologist manpower, clinical coverage, and financial planning as the pandemic lingers on.

“In light of predictions of an upcoming, potentially disastrous ‘second wave’ of coronavirus disease, such information could provide actionable for radiology practice planning,” they said.

From February to April 2020, outpatient imaging in rural areas fell 66 percent.

COVID-19 Decimates Community-Based Imaging Volumes, Hitting Mammography Hardest

Therapixel captured 510(k) clearance from the U.S. Food and Drug Administration this week for MammoScreen™, an explainable, actionable artificial intelligence (AI) tool designed to help with mammography interpretation.

The clearance comes after the release of findings from a multi-case study, company officials said, that showed radiologists had better cancer detection performance on mammograms when they used MammoScreen versus when they analyzed images solo.

“We believe MammoScreen will provide quick and reliable confirmation of radiologists’ suspicions as they read,” said Matthieu Leclerc-Chavlet, Therapixel chief executive officer. “This AI solution will ensure a more certain assessment by radiologists and a speedier reassurance of women having breast cancer screening exams, resulting in a more efficient workflow and reduced costs for the healthcare system.”

According to company information, MammoScreen can automatically detect and characterize suspicious soft tissue lesions and calcifications on mammography images while simultaneously assessing their likelihood of malignancy. Radiologists receive a summary report that categorizes each lesion’s suspiciousness on a scale of 1-to-10 with 10 being the most likely to indicate malignancy.

Artificial intelligence tool designed to improve radiologist cancer detection performance.

FDA Clears MammoScreen™ from Therapixel

“Image Gently” is a familiar standard when it comes to controlling the parameters around imaging children. A group of pediatric radiologists are taking the concept a step further by looking at several aspects of MRI imaging that can play a role in limiting the use of anesthesia during diagnostic scans.

In an effort to identify any relationship between MRI scan time and total propofol dose, a team of investigators from Massachusetts General and Boston Children’s Hospitals examined approximately 500 images from children under age 18. The ultimate goal was to determine whether they could limit the use of anesthetics in kids and potentially determine a range of propofol doses that could be used in clinical practice.

 the team pinpointed three possible factors that could predict propofol exposure: 1.5T field strength, IV contrast administration, and the imaging of multiple body parts.

“Pediatric MRI anesthetic exposure can be quantified and predicted based on imaging and clinical variables,” said the team led by Michael S. Gee, M.D., Ph.D., chief of pediatric radiology at Massachusetts General. “We belive the results presented in this study serve as a valuable baseline for future efforts to reduce pediatric MRI anesthetic medication doses and MRI scan times.”

MRI Technique Could Reduce Radiation Dose in Assessing Pediatric Cancer Treatment

To identify those three factors, Gee’s team retrospectively examined the electronic health records of pediatric patients to capture all MRI exams conducted between 2016 and 2019. All total, they examined 501 exams from 426 patients, including images of the brain, brain and spine, brain and abdomen, and the brain/head/neck.

Based on their analysis, they found patients who had single body parts examined had shorter MRI scans with less propofol than did patients who were having multiple body parts imaged. But, there were several factors that lengthened scan time and increased anesthesia use for patients having single studies: a higher American Society of Anesthesiologists classification, oncologic diagnosis, 1.5T magnet strength, and IV contrast administration.

MRI is already considered a more preferable option than CT for imaging children due to the lack of ionized radiation. Ultimately, knowing how the scan parameters can affect anesthesia use can valuable, particularly with the youngest patients, they said.

“In children under 6 years of age, MRI often requires general anesthesia that may have long-term detrimental effects,” they wrote. “This study helps identify considerations for minimizing MRI anesthesia use. Higher field strength scanners, unenhanced technique, and single body part imaging will all likely reduce patient anesthesia dose.”

Paying attention to scanner strength, IV contrast, and body parts imaged can contribute to reducing anesthesia use.