The U.S. Food & Drug Administration has granted imaging artificial intelligence company Qure.ai its first 510(k) clearance for its emergency room head CT scan product qER.

This move marks the first time the industry has seen a four-in-one clearance. With the FDA’s decision, 

 can be used to triage radiology scans with intracranial bleeds, mass effect, midline shift, and cranial fractures, covering nearly all critical abnormalities seen on routine head CT scans. Two capabilities – cranial features and midline shift – are exclusive to the Qure.ai project, company official said.

With more than 75 million CT scans performed in the United States annually and more than 10,000 people dying within seven days of emergency room discharge, there is a need in the industry for an AI imaging tool that can help providers prioritize and interpret images, said company leaders.

AI Model Uses MRI To Identify Brain Lesions Causing Mass Effect

“Patient outcomes depend directly on the onset-to-treatment time, especially for brain injuries. Every day, doctors are required to weigh the benefits of a potentially life-saving surgery versus the risks of an intracranial bleed or other complication,” said Pooja Rao, co-founder and head of Qure.ai’s research and development. “The sooner they have in-depth information that helps them make that decision, the better for the patient. This is where qER plays a key role.”

According to company information, because time is critical in dealing with many emergency room situations, such as intracranial hemorrhage, the goal is to continue to shorten the 30-minute turnaround time goal that exists for acutely critical conditions. To help reach that aim, the qER suite plugs directly into the radiology workflow, accelerating critical case prioritization.

Articles

510(k) clearance applies to four critical abnormalities; two are exclusive to Qure.ai’s product.

Qure.ai Secures 4-in-1 FDA Clearance for AI-Powered CT Imaging

. I’m senior editor Whitney Palmer. We’re back with you this holiday week to bring you some of the highlights of this week’s radiology news.

This week saw a flurry of news about developments with MRI.

 researchers from Germany used 3T MRI scanners to scan the neocortex, the heavily grooved outer layer of the brain that is integral to language, conscious thought, and other functions, for iron levels. Historically, the neocortex has been difficult to impact because its anatomy prompts distortions, signal decay, and artifacts in MRI scans. However, in this study, investigators were able to show, by comparing patients with Alzheimer’s to healthy participants, that iron levels were higher in the whole cortex for patients with Alzheimer’s, revealing the association between high iron levels in the brain and cognitive deterioration independent of the loss of brain volume. Additionally, the researchers said, their findings point to the possibility of using chelators – iron-reducing drugs – as a potential therapy for Alzheimer’s.

 detailed how an existing MRI technique called T1 mapping could potentially be used to whether children have a particularly aggressive type of cancer – neuroblastoma – and whether the treatment is working at an early stage. Though their current research was conducted in mice, investigators from the Institute of Cancer Research in London said their findings could be applied clinically. In the study, the team used T1 mapping MRI scans to map different tumor cell populations and compared them to maps created with artificial intelligence. They discovered areas with high T1 values corresponded to areas with more aggressive cancer cells, and lower T1 value spots were more benign. They also used T1 mapping to determine the efficacy of two neuroblastoma-targeting drugs, alisertib and vistusertib, and they found that when the drugs succeeded, T1 measures dropped. Their hope, they said, is that this technique can be used to both improve treatment for aggressive cancer and limit the medication side effects children experience by minimizing their exposure to drugs that might not be working.

 a team from the University of Maryland, Baltimore County, detailed a new image analysis method called independent vector analysis for common subspace extraction (IVA-CS) that can be used to categorize fMRI data based solely on brain activity. Clinically, it can help providers treating patients with schizophrenia by grouping patients based on their brain imaging alone. Treatment for schizophrenia is not a one-size-fits-all endeavor, so this data, they said, can be used to just whether a treatment is working, potentially leading to a change in therapy is a patient is seeing no improvement. The investigative team has a plan to researcher this method’s efficacy with addiction and substance abuse in adolescents. 

Across the country, academic medical centers and community hospitals are beginning to see a steep spike in their numbers of COVID-19 patients. Consequently, triaging and managing these patients has become an ever-more critical task as resources and personnel begin to stretch thin. One institution in this position is University of Texas Health in San Antonio. Within the past two weeks, the health system has seen a quadruple increase each in their number of diagnosed COVID-19 cases, the number of patients hospitalized, and the number of patients on ventilators. 

 spoke with Dr. Nilam Soni, director of the Point-of-Care Ultrasound Training Program in the South Texas Veterans Health Care System associated with University of Texas Health, about the role this bedside modality has increasingly played in how they are handling this growing patient volume. Here’s what he shared.

Videos

Diagnostic Imaging's Weekly Scan: July 3, 2020

Point-of-Care Ultrasound and COVID-19, Plus MRI and Schizophrenia, Pediatric Cancer, and Alzheimer's

There is a significant link between the virus that causes COVID-19 and strokes, newly published research reveals, indicating providers should adjust their treatment of these patients accordingly.

 investigators from Ichan School of Medicine at Mount Sinai published the results of an evaluation of more than 120 patients admitted for possible stroke during the pandemic to six New York City hospitals. Their findings indicate that clinicians should more actively monitor these patients to identify any problems as soon as possible.

“This is the first major peer-reviewed study to show that COVID-19 infection is a risk factor for acute strokes,” said lead study author Puneet Belani, M.D., assistant professor of radiology and neurosurgery. “Patients with COVID-19 should be evaluated early for acute neurological changes and timely workup should be performed in patients suspected to have stroke to reduce morbidity and mortality.”

Belani’s team conducted a retrospective, case-control study with 41 patients. Each patient had a stroke confirmed with non-contrast head CT followed by vascular imaging, and the researchers compared that analysis to 82 control participants who, with the same image confirmation, had no evidence of stroke.

Based on their evaluation, they determined that more than 46 percent of patients with acute ischemic stroke also had COVID-19 infection. The same was true for only 18 percent of the control group. After they adjusted for sex, age, and other risk factors, the association between the virus and stroke was clear.

“We found that COVID-19 is an independent risk factor for imaging-confirmed acute ischemic stroke during stroke alert evaluation, after controlling for traditional vascular risk factors,” the team said. “This finding suggests that COVID-19 infection is associated with increased morbidity and mortality that transcends the primary cardiopulmonary sequelae from the infection.”

After adjusting for traditional vascular risk factors, radiologists encourage aggressive monitoring of COVID-19 patients for blood flow changes to the brain. 


COVID-19 Is an Independent Stroke Risk Factor

Estimating blood flow across a variety of systems and facilities with a 3D ultrasound approach is both effective and non-invasive, according to newly published research.

Being able to accurately and quickly assess blood flow measurements and blood flow to organs, such as the heart and brain, is critical in emergency situations. These measurements also play a role in chronic disease management, such as monitoring blood flow to the hands and feet in patients with diabetes. But, getting reliable measurements is difficult.

“Right now, we just don’t have anything better to quantify blood flow,” said Oliver D. Kripfgans, Ph.D., associate professor of radiology at Michigan Medicine and lead study author of the article published in 

Current tactics, including blood pressure measurements and 2D ultrasound, such as spectral Doppler, provide surrogate measures rather than actual volumetric flow. They are also error prone and can vary significantly between facilities and operators.

To fix this problem, Kripfgans’ team has spent several years developing a 3D approach that could yield quantitative measurements. For this study, the team partnered with volunteers from the Quantitative Imaging Biomarkers Alliance (QIBA), a healthcare research alliance launched by the Radiological Society of North America to improve current biomarkers and investigate new ones.

Together, the group tested their 3D strategy on three clinical scanners using a custom flow phantom, a device that mimics human blood flow. In seven different labs, they altered eight distinct testing conditions, including changing flow rate, imaging depth, and other parameters, to evaluate the approach’s efficacy and reliability. For each scanner, they monitored and recorded 3D quantitative blood flow measurement dependence on color gain. This tactic, the team said, decreased user and scanner measurement variability.

According to Kripfgans, their analysis revealed that blood flow volume that is estimated by 3D color-flow ultrasound is accurate. The findings, he said, were reproduced across all seven labs.

“We had less than 10-percent error or variation,” he explained. “For some of the systems, we were down to only 3-percent to 5-percent difference between labs. These are fantastic results that show that, from a technology point-of-view, some systems could be ready to go to the clinic.”

Because the system is easy to use, it reduces the variation between operators and facilities that routinely pop up with existing blood flow assessment tactics, he said. The range of tested flows means that the 3D approach could have a wide range of clinical applications, including peripheral vascular flow and cerebral blood flow estimation other than perfusion and cardiac output. And, blood flow estimation could allow for:

More accurate measurements in hemodialysis arteriovenous fistulas and grafts

Evaluating stenosis and function in transjugular intrahepatic portosystemic shunts

Alongside these possible clinical applications, having QIBA involved, with its goal of accelerating the development and adoption of hardware and software standards to facilitate accurate and reproducible quantitative imaging method results, will make implementation of the 3D approach in patient care much easier, he said.

“Because of QIBA and this study, I’m confident that this 3D ultrasound technology is on a path to the clinic,” Kripfgans said.

Monitoring color flow gain in 3D quantitative blood flow measurements provides accurate and reliable information.


3D Ultrasound Yield Accurate Volumetric Blood Flow Measurements Non-Invasively

Working remotely can be a challenge for any radiologist under the best of circumstances, but the current environment controlled by the COVID-19 pandemic has highlighted the need for information technology (IT) solutions that can help providers and their staff maximize their productivity in this situation.

During SIIM2020, company leaders from vendor Konica Minolta Healthcare Americas, Inc., highlighted the challenges radiologists and their IT colleagues face when trying to streamline remote workstations and workflow, as well as some solutions to overcome these obstacles.

According to Chris Van Houten, Konica Minolta’s vice president of sales, radiologist and IT professionals face a myriad of problems when trying to create a seamless workflow that allows providers to download, access, and send out image files with ease. Every facility should be aware of these stumbling blocks so they can ensure any solution they choose addresses the pain points.

In addition to the cost of updated technology, radiologists could find themselves facing these obstacles:

 Without sufficient bandwidth, radiologists can find themselves waiting for images – particularly advanced images – to load into their system. This can bring productivity to a screeching halt, Van Houten said, as some files, such as 3D mammograms, can take up to 20 minutes to download. 

 It is commonly said in the industry that it is imperative for the right radiologist to read the right study at the right time. Without a way to filter through studies in the worklist, meeting that goal can be difficult. For a streamlined workflow system, a worklist must include a filter so providers can sift through the waiting studies for the most urgent ones within their specialty area, he said.

 Even the most up-to-date, tech savvy office can be stymied when working with referring clinicians or other collaborators that are relying on older technologies. In these situations, he said, it is harder to share all types of files. 

IT professionals also encounter their own stumbling blocks, Van Houten explained. In most situations, upgrading a workstation’s software means physically working with the machine. With many providers currently working remotely, IT departments have had to find ways to access and upgrade each system, and the process can be time consuming. Also, if an office purchases any new software, they must make sure that the workstations can be updated to leverage it appropriately.

Fortunately, said Kevin Borden, Konica Minolta’s vice president of product, the industry does offer two solutions to help radiology offices overcome these challenges. Pairing server-side rendering with zero-footprint viewers can pave the way for rapid, smooth workflow.

 This solution eliminates the need for traditional workstations to process all the calculations needed associated with reading images. Instead of downloading images to the radiologist’s browser, SSR renders the images in the server and send it to the workstation, alleviating the stress from the workstation itself and allowing for faster downloading and processing.

“Server-side rendering does all of the heavy lifting that a traditional workstation had to do,” he said. “It performs all of the calculations, so your expensive powerful server-based workstations don’t really have to exist.”

Not only does this solution allow for large files and prior studies to be downloaded and accessed quickly, but it also side-steps the need to push whole sets of DICOM data to a workstation.

 Implementing this technology requires no installation on workstations, and it functions on any operating system or browser. The high diagnostic quality and full tool set is available for use on any modality at any workstation, Borden explained, even for radiologists working remotely.

As with the challenges, radiologists and IT professionals experience their own unique benefits, especially when SSR and ZFP are combined, Borden said, in a product, such as Konica Minolta’s Exa® Enterprise Imaging.

For the radiologist, a combined platform makes accessing larger files, including 3D mammography or echocardiograms, much faster without any pre-fetching or moving data from workstation to workstation. It also frees the provider up – he or she is no longer strapped to a specific workstation because they can now view any modality from any location.

“The radiologist can pick up a laptop with an approved monitor, and as with any workstation before, hit a URL, log into the application, and begin reading remotely,” he said.

IT professionals also enjoy several benefits with an SSR+ZFP platform, Borden explained. By not having to physically touch workstation, their workload is significantly lighter. And, without installing applications or moving data around frequently, there is less strain on a department or office’s network, reducing utilization.

The benefit that has been most helpful during the pandemic, however, he pointed out, is that with this type of platform – with no installations or downloads required – once the appropriate equipment is in place, radiologists can begin to immediately read remotely.

“In today’s environment, that’s very important,” he said. “Before the pandemic and post-pandemic, the future needs to be now downloads or installations necessary.”

Konica Minolta leaders outline the challenges to – and discuss tech solutions for -- provider and IT productivity during SIIM2020.

Technology for Improving Productivity and Efficiency

In a direct comparison, a new radiopharmaceutical has emerged as a superior imaging agent for patients with neuroendocrine tumors who have cancer that has spread to their liver, according to newly published research.

In testing the radiopharmaceutical 68Ga-DOTA-JR11, researchers from China determined the agent does a better job detecting liver metastases than another somatostatin receptor (SSTR) 68Ga-DOTATATE PET/CT. They published their results in the 

“What we’ve learned from this study is that peptides matter,” said lead study author Wenjia Zhu, M.D., a nuclear medicine physician at Peking Union Medical College Hospital in Beijing. “For patients with different metastatic patterns, different peptides should be used. In liver-dominant disease, 68Ga-DOTA-JR11 may be a better choice in tumor staging and re-staging compared to 68Ga-DOTATATE.”

In the treatment of neuroendocrine tumors, SSTRs are the main targets for imaging and radionuclide therapies. They, typically, are imaged with 68Ga-labeled peptides, agonists that link with SSTRs to prompt a response. But, newly developed peptide antagonists have revealed better pharmacokinetics, improved image contrast, higher tumor uptake, and better residence time in recent studies.

The researchers made this discovery through a two-day study with 31 patients. Patients received a 68Ga-DOTATATE intravenous injection on day one and a 68Ga-DOTA-JR11 on the second day. Investigators ran a 40-to-60 minute whole-body time-of-flight PET/CT after each injection on the same scanner. After the scans, they compared physiologic normal-organ uptake, lesion numbers, and lesion uptake for both imaging agents.

Based on that comparison, Zhu’s team determined that 68Ga-DOTA-JR11 had significantly lower physiologic normal-organ uptake of the spleen, renal cortex, adrenal glands, pituitary glands, stomach wall, normal liver parenchyma, small intestine, pancreas, and bone marrow. Additionally, 68Ga-DOTA-JR11 also detected more liver lesions with a higher target-to-background liver lesion ratio, but 68Ga-DOTATATE found more bone lesions.

With 68Ga-DOTA-JR11, the team said, more options now exist with imaging, and their work provided greater clarity about how the imaging agents performs and compares to other agents.

“With antagonists, we now have an alternative to agonists,” Zhu said. “However, there is still not much evidence about the performance of PET/CT imaging with SSTR antagonists. Hence, we designed this prospective study.”

Newly created peptide antagonist outperforms standard agonist imaging agent in head-to-head comparison.

New Radiopharmaceutical Offers Better Tumor Imaging and Staging for Patients with Liver Cancer

Iron accumulation in the outer layer of the brain – identified with MRI scans – is linked to cognitive decline in people who have Alzheimer’s disease, according to new research.

 show that MRI can be used to determine how much iron has collected in the neocortex, the heavily grooved outer layer of the brain that is integral to language, conscious thought, and other functions. Until now, being able to assess the presence of iron in this layer has been difficult because its anatomy can create distortions, signal decay, and artifacts in MRI scans.

The association between high iron levels in the brain and Alzheimer’s is already known, but most research has focused on how the iron collects in the deep gray matter structures of the brain. This is the first study to look at the relationship between iron changes in the neocortex and cognitive decline.

But, given the complexities presented with the neocortex itself and the limitations of scanning time in the clinical environment, researchers needed to find another way to adequately assess iron levels, the team indicated.

PET, MRI Identify What Increases Women’s Likelihood of Alzheimer’s

"The best solution to minimize these artifacts would be using ultra-high-resolution scans,” said study co-author Reinhold Schmidt, M.D., professor of neurology and chair of the neurology department at the Medical University of Graz in Austria. “However, in the clinical setting, scan time is a limiting factor, and a compromise has to be found.”

To achieve this compromise, Schmidt’s team used a 3T MRI scanner with a 12-channel passed-array head coil that allowed for a balance between resolution and scan time. It also enabled any impacts from distortions to be corrected during post-processing. Initially, they enrolled 100 patients with Alzheimer’s and 100 health control participants. Among the Alzheimer’s group, 56 were able to fully complete the study with subsequent neuropsychological testing and brain MRI scans conducted after an average of 17 months.

Using this technique, they created a brain iron map that revealed the iron levels in several parts of the brain.

“We found indications of higher iron deposition in the deep gray matter and total neocortex, and regionally in temporal and occipital lobes, in Alzheimer’s disease patients compared with age-matched healthy individuals,” Schmidt said.

FDA Approves First Radiopharmaceutical for Imaging Tau

Specifically, a comparison of R2* values in the total and regional brain matter between patients with Alzheimer’s and the healthy control participants revealed these values were higher in the whole cortex, including the neocortex and deep gray matter of patients with Alzheimer’s. In addition, these patients had higher media R2* levels in their temporal and occipital lobes, as well as the caudate nuclei, a structure that plays a vital role in how the brain learns.

Overall, the team said, the collection of higher levels of iron in the brain were associated with cognitive deterioration independent of the loss of brain volume, adding that changes in iron levels in the temporal lobes did correlate with cognitive decline in patients who had Alzheimer’s.

Ultimately, they said, these findings bolster the existing belief that high iron levels in the brain directly facilitate amyloid beta deposits and neurotoxicity in Alzheimer’s. The results also point to the possibility of using iron-reducing drugs, known as chelators, potential therapy for Alzheimer’s.

“Our study provides support for the hypothesis of impaired iron homeostatis in Alzheimer’s disease and indicates that the use of iron chelators in clinical trials might be a promising treatment target,” Schimdt said. “MRI-based iron mapping could be used as a biomarker for Alzheimer’s disease prediction and as a tool to monitor treatment response in therapeutic studies.”

Scans used, for the first time, to examine relationship between neocortex and cognitive decline.


MRI Shows Link Between Higher Brain Iron Levels and Cognitive Decline in Alzheimer’s Patients

A custom-designed app made specifically for radiologists can successfully encourage providers to submit reports of incidents that could negatively impact patient safety, workflow, and quality, according to research presented at 

Historically, incident reports have been largely submitted by radiology technologists. To broaden participation in submitting these reports, researchers from Thomas Jefferson University Sidney Kimmel Medical College created an internal incident and issue reporting app to not only capture more episodes, but to also increase radiologist involvement with the hopes of acting upon issues within six months.

“Customized reporting tools are a useful supplement to existing systems to capture issues that may impact quality and workflow, may have downstream effects on patient safety, and may otherwise go unreported,” investigators said. “A custom application built for radiologists can increase their engagement in the reporting process, by providing a tool for discussing and resolving latent but important issues.”

Their six-month intervention proved the new tool could successfully reach that goal. According to their analysis, radiologist involvement spiked from 5 percent prior to the tool launch to 97 percent of all submissions. They, consequently, accounted for 46 percent of all submissions across both the enterprise and internal tools.

The findings also showed several other positive outcomes. While the existing enterprise tool collected roughly the same number of incident reports (215 pre-launch and 203 post-launch), the internal tool received an additional 168 submissions – a 72-percent increase. Additionally, submissions via the new tool spanned seven categories with a 62-percent rise in the types of issues captured. Workflow, protocol, IT, and communications issues were the most frequently reported.

By the end of the six-month intervention, 37 percent of reports submitted via the internal app had been either triaged or resolved.

The research team collected their data from March 1, 2019, to Sept. 1, 2019 via the new mobile-friendly app. Radiology staff voluntarily submitted confidential and optionally anonymous reports. The team compared pre- and post-launch data for the number of submissions, incident categories, severity scores, and staff roles. In addition, they took steps to ensure the new tool did not replace the existing enterprise system, and they sent out reminders for the staff and providers to use the app, encouraged user engagement through a feedback loop, created a volunteer committee to resolve submissions.

Internal reporting app, designed for radiologists, increased provider participation and captured incidents in a greater number of categories.


Improving Incident Reporting: There’s an App for That

I’ve been writing this column, believe it or not, since 2011. It seems safe to say I have some skills when it comes to putting words together. However good or bad at various facets of it I may be, I feel comfortable in calling myself a writer.

I have also, from time-to-time, taken on a related role: that of an editor, for instance with my med school’s student newspaper. I don’t think anyone would be surprised at the notion that there is considerable overlap between the skill sets of writers and editors.

Even if one assumes no difference in aptitudes between the two, their focuses and goals differ. As a writer, I might want to put forward page after page in the name of sharing my every thought with whichever readers might be interested. As an editor, I might trim someone’s submission to be more digestible.

In last week’s column, for instance, I referenced (humorously, I hope) a number of movie-quotes with relevance to my practice of radiology. One that might have been the funniest was from a Samuel L. Jackson character in 

 I substituted the oft-misused “R/O” in healthcare for the word “what,” in one of his mini-tirades (his first scene, in which he and Travolta are interrupting some young folks’ breakfast).

I make these specifications lest readers want to imagine what would have been in the column…because it wasn’t. That particular quote was edited out of the piece. I guess because the quote had some salty language in it, and even though I liberally replaced letters in the naughty words with asterisks, I’m guessing that was considered insufficient.

Some writers might be disappointed, even outraged, when they see the work they have done being so altered. They can lose sight of the (I cringe to use the trite term) “value added” by an editor other than themselves. Not just in terms of massaging a piece to better appeal to a desired target-audience: There are things that only a different set of eyes (and associated brain) will notice when reviewing a piece.

Self-editing is unavoidably subjective; you need somebody else if you want an objective appraisal. I typically review an entry of my column multiple times before emailing it to my editor for publication. I almost always trim or adjust something on each pass. No matter how many times I go through this, however, if I subsequently see it online, I might still wish I’d said something differently. And, then there are any number of things that might seem fine to me, even with a dozen re-readings, that my editor might recognize at a glance as needing tweaking for better consumption by others.

I’ve had more than a few thoughts along these lines when contemplating the numerous pages’ worth of radiology reports I create in a typical working-day. Far greater volume is generated there, under (partially self-imposed) pressure for better speed and productivity…yet how carefully am I proofreading those reports? Maybe one speedy going-over before hitting “Sign?”

Granted, in that context I’m not trying to be witty or thought-provoking. As long as I’m accurately conveying what an imaging study showed and what it means, my mission is accomplished. Grammar, even spelling, don’t have to be perfect. Indeed, in the name of efficiency and giving the voice-recognition software less to mess with, I have embraced the usage of sentence fragments, as much as it initially offended my sensibilities to do so.

I have nevertheless come to wish that we had some sort of editor other than ourselves, in the radiological workflow. As it stands, the only editing we get tends to be when someone has an issue with reports we already signed, and, then, the best we can do is to make an addendum…which leaves the original content untouched (and highlights it as an imperfection to subsequent readers).

Some places I’ve worked have innovated a layer of software for this purpose. If you’re trying to sign off a report on a right-sided extremity, for instance, and you dictated the word “left,” you get interrupted with a warning. Or, if you’re reading a male patient, and your report mentions the uterus or ovaries. Even if you mention metacarpals in a study of the foot, or metatarsals in the hand.

One problem with such things is that there are always exceptions, and they irritate us would-be productive workers every time they trip us up. For instance, a study was of the right hip, but the left one was also in the field-of-view, and unless I try to fool the software by saying “contralateral” instead of “left,” I’m going to get interrupted when I try to sign off.

Or, it is a female patient, and I dutifully reported on the gynecological viscera. But, the tech erroneously entered the study as being on a male patient, so the software interrupts to alert me to the discrepancy. I also once got hassled by such software when I was reading a scan on a transgender patient.

Even spellcheckers can hinder, as well as help. They can miss a great many medical terms from their dictionaries, and we radiology-drones don’t always have access to an “Add” button to tell the checker that terms like “sphenopalatine” should, henceforth, not prompt a pop-up window. Not to mention the abundance of proper names we might need to include, say when documenting that we spoke to the referrer, Dr. X, or suggesting that a patient might have Lady Windermere syndrome. Also, some dictation software auto-populates certain fields with things the techs or referrers wrote, and anytime they make spelling errors or use unsanctioned abbreviations, we’re the ones who get interrupted by software alerts at signing time.

So, I have come to fantasize about what it might be like to have an honest-to-goodness human editor going over my radiology reports before I officially sign them. And, thinking about it more than once in a blue moon, I’ve moderately fleshed out the idea. I’ll go into how such a thing might practically work next time.

Contemplating an editor for radiology reports.

Your Friend, The Editor (Part 1)

An already available MRI technique could help providers determine whether a child has a particularly aggressive cancer, as well as pinpoint early whether a treatment is working, new research has revealed.

 researchers from The Institute of Cancer Research in London described how using an MRI technique, called T1 mapping, could reveal critical details about neuroblastoma in children. It is the first study, the team asserted, to evaluate the benefit of this MRI technique as a “smart” cancer biopsy.

According to existing data, neuroblastoma is the most common cancer identified in children under age 1. It accounts for approximately 6 percent of pediatric cancers with 800 new cases diagnosed annually in the United States.

Although their work, to date, has been in studying neuroblastoma in mice, the investigators said the non-invasive T1 mapping scans could potentially have wide clinical applicability, allowing for treatments to be tailored to individual patients.

“Our findings show that an imaging technique readily available on most MRI scanners has the potential to pick out children with aggressive cancer and give us early signs of whether a treatment is working,” said lead study author Yann Jamin, Ph.D., Children with Cancer UK Research Fellow at the Institute of Cancer Research in London. “We’ve shown in mice that this technique can give us detailed insights into the biology of neuroblastoma tumors and help guide use of precision medicine, and next we want to assess its effectiveness in children with cancer.”

T1 mapping scans – studies that measure how water molecules interact at microscopic levels inside cells, offering a greater understanding of the cellular tissue composition – are currently used to analyze heart muscle tissue damage. In this study, the team used those T1 mapping MRI scans to map different tumor cell populations and compared them to maps created with artificial intelligence.

T1 Mapping with Cardiovascular MRI Detects Fabry Disease, Heart Damage

They found that T1 values directly corresponded to changes in the aggressiveness of the cancer cells. High T1 value regions – where water molecules behave “more freely” – were locations of more aggressive cancer cells, and lower T1 value spots were more benign or even had dead tissue.

As an additional part of the study, investigators looked into whether T1 mapping could help them determine how mice with neuroblastoma responded to two drugs – alisertib and vistusertib – that target MYCN, a key protein linked to aggressive forms of neuroblastoma.

Given this analysis, they found that when the drugs succeeded in stopping tumor growth, T1 measures decreased, indicating the cancer cells had died. This outcome, they said, suggests that T1 measures could be a potential biomarker.

Based on the results of this study, the team said, the next step is to launch a clinical study to determine the benefit of T1 mapping in children.

“It’s vital that we find ways to improve treatments for aggressive childhood cancers like neuroblastoma,” said Paul Workman, Ph.D., chief executive of The Institute of Cancer Research, “and also that we spare children unnecessary side effects by minimizing exposure to drugs that do not seem to be working.”

T1 mapping has the potential to determine neuroblastoma aggressiveness, as well as early efficacy of treatment.

Readily Available MRI Technique Could Shed Light on Common Pediatric Cancer

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

Patients with COVID-19 pneumonia who must be placed on a ventilator are more likely than other patients to experience an injury from increased air pressure in their lungs, placing them at a higher risk of death, new research has found.

 investigators from New York University (NYU) Langone Health discussed their observation that more patients with COVID-19 infection on ventilators developed barotrauma – an injury associated with elevated air pressure in the lungs – than did other patients who required intubation. Being aware of this risk can be vital for optimizing patient care, the researchers said.

“High barotrauma rates in patients with COVID-19 infection on [invasive mechanical ventilation] is associated with longer hospital stay and is a risk factor for higher mortality,” said the team led by Georgeann McGuinness, M.D., senior vice chair of the NYU Langone Health radiology department. “Barotrauma risk is particularly important to recognize as these critically ill patients may be managed by staff less familiar with the management of ventilator settings.”

Chest X-rays Can Predict COVID-19 Severity in Young and Middle-Age ER Patients

To determine the rate of barotrauma in COVID-19 patients, the team evaluated chest X-rays from 601 patients over age 18 who were positive for the virus and were also put on a ventilator. Of that group, 89 patients – 15 percent – developed at least one barotrauma event. All total, this group experienced 145 barotrauma episodes for an overall barotrauma rate of 24 percent. Specifically, the team identified:

11 patients with bilateral pneumothoraces

54 patients with either unilateral or bilateral pneumothorax

59 patients with pneumomediastinum (the most common barotrauma event)

They compared these results to the incidence of barotrauma in a group of 196 intubated patients who were negative for the virus. In that group, investigators identified only one event. The team also compared these findings to 285 intubated patients with acute respiratory distress syndrome (ARDS) from 2016 to 2020. During that time, 28 patients had 31 barotrauma events – a rate of 11 percent.

These findings fall in line with previous research that examined barotrauma rates during the SARS and MERS outbreaks. According to that data, between 12 percent and 34 percent of SARS patients developed barotrauma, and 30 percent of intubated MERS patients in the intensive care unit experienced pneumothorax. These rates, the team said, beg the question of whether coronavirus infections present a unique barotrauma risk.

X-ray as a New Frontline Tool in COVID-19 Pandemic

Overall, the team determined that the COVID-19 patients who develop barotrauma were younger than intubated ARDS or COVID-19-negative patients (63 years versus 68 years), suggesting that younger age in these patients after ventilation is an independent barotrauma risk factor. Additionally, a multivariate analysis revealed that alongside barotrauma, mortality in COVID-19-positive patients in this study was associated with greater age, shorter hospital length of stay, and Hispanic ethnicity.

Ultimately, the team said, they hope these findings can be used clinically to improve patient outcomes.

“Our observed high rate of barotrauma in patients with COVID-19 infection on [invasive mechanical ventilation] may support emerging theories of lung damage in SARS-CoV-2 infection,” they said.

Patients positive for the virus are more likely to experience barotrauma, putting them at risk for longer hospital stays and mortality.


X-ray Identifies More Air Pressure Lung Injuries in COVID-19 Patients

Advanced stroke imaging company RapidAI announced today it has secured 510(k) clearance from the U.S. Food & Drug Administration for its neuroimaging analysis device Rapid ASPECTS.

This is the first device to receive clearance in the computer-aided diagnostic software (CADx) category, and, according to a company statement, it is the only product confirmed to improve physician interpretations of non-contract CT using a standardized APSECT score. It can be particularly useful for less experienced readers, company officials said.

“The FDA’s advanced CADx clearance has never been granted to a neuroimaging solution before,” said Jim Rosa, senior vice president of regulatory and clinical at RapidAI. “Through a combination of clinically validated advances in AI and machine learning, Rapid ASPECTS offers a glimpse into the future of stroke imaging.”

CT Predicts Stroke Patients Benefitting From Endovascular Thrombectomy

Rapid ASPECTS is based on the Alberta Stroke Program Early CT Scoring (ASPECTS) system, and it uses a validated machine-learning algorithm to identify affected regions of the brain and create an ASPECT score than can pinpoint early signs of stroke in non-contrast CT scans. In doing so, it can help providers identify areas of the brain that have irreversible injury.

Overall, according to company information, Rapid ASPECTS offers three main features:

Standardized, rapid assessment within a few minutes to help physicians determine a patient’s eligibility for thrombectomy

Fast measurements via machine-learning algorithm validated by clinical research studies

Collaboration between facilities and specialists for faster triage and transfer decision

Ultimately, company officials said, Rapid ASPECTS is designed to facilitate better patient outcomes.

“Rapid ASPECTS represents the next AI-powered step forward in stroke imaging, and the groundbreaking CADx clearance from the FDA puts it in a class by itself,” said RapidAI co-founder Greg Albers, M.D., professor of neurology at Stanford University and director of the Stanford Stroke Center. “In addition to helping stroke teams with faster triage and transfer decision, Rapid ASPECTS also improves the accuracy of typical readers.”

Rapid ASPECTS is already available internationally and will be commercially available in the United States later in July.

Rapid ASPECTS receives clearance as first neuroimaging analysis device in computer-aided diagnostic category.

 FDA Approves First AI-Powered Stroke Imaging Solution

Artificial intelligence (AI) tools that operate autonomously in radiology are not yet safe, and pursuing their use should be done carefully, two industry leadership organizations have told the U.S. Food & Drug Administration (FDA).

 detailing their reticence, the American College of Radiology and the Radiological Society of North America leaders expressed concerns that these AI tools have not been sufficiently tested to ensure their safety – and that a framework is not currently in place to address that problem. This message comes in response to a two-day FDA workshop convened earlier this year to assess AI’s integration into medical imaging.

“The ACR and RSNA believe it is unlikely the FDA could provide reasonable assurance of the safety and effectiveness of autonomous AI in radiology patient care without more rigorous testing, surveillance, and other oversight mechanisms throughout the total product life cycle,” said Howard Fleishon, M.D., ACR chairman, and Bruce Haffty, M.D., RSNA chairman, in the letter. “We believe this level of safety is a long way off, and while AI is poised to assist physicians in the care of their patients, autonomously functioning AI algorithms should not be implemented at this time.”

American College of Radiology Offers Suggestions for Federal Artificial Intelligence Oversight

To avoid potential problems with patient safety, care, and outcomes, both leaders urged the FDA to proceed slowly, specifically waiting until more radiologists are comfortable using AI algorithms.

Much of their concern stems from the relatively small use of AI algorithms in the industry to date. According to a recent ACR survey, only 30 percent of radiologists use AI in clinical practice, averaging 1.2 algorithms per radiologist. Instead, most providers use the tools solely for research.

Approved algorithms on the market are not intended to work without human oversight. Instead, they assist in image interpretation and exam prioritization, as well as help with some administrative tasks. In most cases, the tools were not tested to be generalizable across a heterogenous patient population or across varying modalities, the leaders said.

“It is not surprising that 93 percent of radiologists using AI in our survey said that the results of AI in their practices are inconsistent, and 95 percent said they would not use the AI algorithms without physician overread,” they wrote.

If the goal is to continuing pursuing autonomous AI in radiology, Fleishon and Haffty said, there are various steps the FDA should consider. They recommended:

All AI algorithms undergo testing using multi-site heterogenous data sets to ensure a minimum level of generalizability across patient populations, imaging equipment, and protocols.

Rigorous pre-market approval process to ensure patient safety across heterogeneous patient populations and equipment types.

Develop post-market monitoring requirements, possibly using trusted third-party registries, to protect patients and the public.

Post-market oversight mechanisms should be in place to ensure algorithms function longitudinally.

Requirements should be developed for continuous monitoring of all algorithms used in clinical practice, regarding patient demographics, equipment type, and imaging protocols. These should be monitored by interpreting physicians.

Establishing a close relationship between the FDA and radiology leadership organizations to pinpoint practical use cases for autonomous AI.

Most importantly, Fleishon and Haffty pointed out, safe image interpretation, such as the analysis of screening mammography, still requires active radiologist involvement, and removing the provider and the context he or she brings to each interpretation could have negative consequences.

“We are not confident that these examinations can be safely excluded from radiologist interpretation while maintaining the current level of patient safety,” they wrote.

Overall, they recommended the FDA focus its efforts in algorithms that help providers with population health as a way to integrating autonomous AI into radiology care, such as concentrating on algorithms that can incidentally detect and quantify potentially undiagnosed chronic disease. The human touch is still necessary when examining, recognizing, and characterizing disease, they said.

“The value that human interpretation with independent medical judgement brings to patient care cannot currently be replaced,” they said.

ACR and RSNA leaders request the FDA slow its momentum with AI tools that work without radiologist oversight.