Greg Freiherr, Industry Consultant

Greg Freiherr has reported on developments in radiology since 1983. He runs the consulting service, The Freiherr Group.

Blog | Greg Freiherr, Industry Consultant | Artificial Intelligence | February 03, 2016

Will Artificial Intelligence Find a Home in PACS?

PACS

Images courtesy of Pixabay.

Time is brain. The longer a brain injury goes untreated, the higher the chance of permanent damage or death. And no brain injury is more insidious than the ones that cause hemorrhage. If unchecked, intracranial hemorrhage (ICH) and the edema that goes with it literally squeeze the brain to death.

Although ICH accounts for no more than 13 percent of all strokes, it claims 20,000 lives annually in the United States. And it comes not just from stroke but also blunt force trauma, the kind that often accompanies car accidents and assaults.

Recognizing this condition is a high priority in emergency medicine and, consequently, at vRAD, which interprets medical images for thousands of ERs across the U.S. On average, the results of emergency computed tomography (CT) of patients with intracranial bleeds are distributed, read and interpreted by vRAD neuroradiologists in 10 to 15 minutes.

A smart algorithm, stoked with artificial intelligence, promises to cut that time in half.

vRAD, also known as Virtual Radiologic (an affiliate of MEDNAX), is grooming the currently experimental algorithm as a traffic cop for its picture archiving and communication system (PACS) workflow. The algorithm, which has been taught to recognize the signs of intracranial bleeding in CT brain scans, is being trained to direct suspicious images to vRAD’s neuroradiologists.

Shannon Werb, vRAD chief information officer, explains that the algorithm, working with vRAD’s PACS, is being developed to channel the images to neuroradiologists licensed to practice in the state where the patient was scanned. The first of these neuroradiologists to become available would immediately open the study. This could cut the time elapsed from exam to interpretation from the current 10 to 15 minutes to 5 or 10 minutes, Werb said.

But before the algorithm can enter day-to-day operation, it must pass review at the U.S. Food and Drug Administration (FDA). The company is now preparing an FDA application, proposing to use the algorithm to smarten up its already FDA-cleared PACS. The algorithm’s patent filing will provide the basis for the application, he said.

Unlike other AI applications being developed as diagnostic assistants, the vRAD algorithm is not about changing the practice of medicine. It is about making medicine more efficient, according to Werb.

“This is not about replacing doctors,” he said. “We want to help physicians do a better job and do it more rapidly and more accurately.”  

The algorithm will leave the practice of medicine to vRAD’s legion of radiologists, who provide preliminary interpretations during the day, and specialty and after-hours final interpretations.

Contracting for these services are about 40 percent of the healthcare facilities in the United States, according to Werb, as well as facilities in ten more countries.

The company orchestrates the interpretation of client images by assigning patient cases to the worklists of its radiologists. Assignments are made on the basis of radiologists’ special training or availability.

“We do this 20,000 times a day across 2,000 facilities,” he said. 

Development of the algorithm for intracranial hemorrhage is part of an effort by vRAD to make its operations more efficient, using AI to speed diagnosis and, thereby, improve patient care. And that effort is only just getting started.

Eventually the algorithm now in development may evolve to one that points neuroradiologists to specific areas in the CT images where they should look for signs of hemorrhage. And the company has plans to develop algorithms focused on life-threatening conditions besides intracranial bleeds. High on the list are pulmonary embolism and aortic tears. The goal, Werb said, is to “move technology close to the patient” so as to “tap the doctor on the shoulder a lot more rapidly.”

One day the company may completely bridge the gap between exam and interpretation. Working in concert with equipment manufacturers, vRAD hopes to embed AI algorithms in the scanners themselves.

This would allow the algorithms to flag suspicious images — while the patient is still on the scanner table.  

Watch the VIDEO “Examples of Artificial Intelligence in Medical Imaging Diagnostics.” This shows an example of how AI can assess mammography images.
 

Watch the VIDEO “Development of Artificial Intelligence to Aid Radiology,” an interview with Mark Michalski, M.D., director of the Center for Clinical Data Science at Massachusetts General Hospital, explaining the basis of artificial intelligence in radiology.

Editor’s note: This is the first blog in a series of four by industry consultant Greg Freiherr on Working With Smart Machines. To read all of Greg's blogs, click here.

Related Content

A, Initial conventional axial CT image shows no noticeable lung damage (within red box) in right upper lobe. B, Electron density spectral CT image obtained at same time as image in A shows lesions (within red box) in right upper lobe. C, Follow-up conventional axial chest CT image obtained 5 days after images in A and B confirm presence of lesions (within red box) in right upper lobe.

A, Initial conventional axial CT image shows no noticeable lung damage (within red box) in right upper lobe. B, Electron density spectral CT image obtained at same time as image in A shows lesions (within red box) in right upper lobe. C, Follow-up conventional axial chest CT image obtained 5 days after images in A and B confirm presence of lesions (within red box) in right upper lobe. Image courtesy of the American Roentgen Ray Society (ARRS), American Journal of Roentgenology (AJR)

News | Coronavirus (COVID-19) | October 22, 2020
October 22, 2020 — According to an open-...
The fMRI hyperscanning environment.

(A) The fMRI hyperscanning environment. The clinician (1) and patient (2) were positioned in two different 3T MRI scanners. An audio-video link enabled online communication between the two scanners (3), and video images were used to extract frame-by-frame facial expression metrics. During simultaneous acquisition of blood oxygen level–dependent (BOLD)–fMRI data, the clinician used a button box (4) to apply electroacupuncture (EA) treatment (real/sham, double-blind) to the patient (5) to alleviate evoked pressure pain to the leg (6; Hokanson cuff inflation). Pain and affect related to the treatment were rated after each trial. (B) Study overview. After an initial behavioral visit, each individual participated in a Clinical-Interaction (hyperscan preceded by a clinical intake) and No-Interaction condition (hyperscan without a preceding intake), in a counterbalanced order, with two different partners. (C) Experimental protocol. Each hyperscan was composed of 12 repeated trials (four verum EA, four sham EA, and four no treatment) in a pseudo-randomized order. After a resting period (far left), both participants were shown a visual cue to indicate whether the next pain stimulus would be treated (green frame) or not treated (red frame) by the clinician. These cues prompted clinicians prepare to either apply or not apply treatment while evoking corresponding anticipation for the patient. Following the anticipation cue, moderately painful pressure pain was applied to the patient’s left leg, while the clinician applied or did not apply treatment, respectively. After another resting period, participants rated pain (patients), vicarious pain (clinicians), and affect (both) using a visual analog scale (VAS).

News | Clinical Trials | October 22, 2020
October 22, 2020 — The potential impact of the patient-clinician relationship on a patient's response to treatment is
The FDA clearance, Quantib’s 6th to date, marks the first time a comprehensive AI prostate solution will be available to radiologists in the United States
News | Prostate Cancer | October 21, 2020
October 21, 2020 — Quantib, a market leader in...
Lesion was originally reported as indeterminate enhancing mass, and outside report recommended biopsy. Classic features of benign hemangioma are shown. Error was attributed to faulty reasoning. A, Axial MR image obtained 5 minutes after contrast agent administration shows peripheral nodular discontinuous enhancement. B, Axial MR image obtained 10 minutes after contrast agent administration shows centripetal progression of enhancement (arrow). C, Axial fast imaging employing steady-state acquisition (FIESTA)

Lesion was originally reported as indeterminate enhancing mass, and outside report recommended biopsy. Classic features of benign hemangioma are shown. Error was attributed to faulty reasoning. A, Axial MR image obtained 5 minutes after contrast agent administration shows peripheral nodular discontinuous enhancement. B, Axial MR image obtained 10 minutes after contrast agent administration shows centripetal progression of enhancement (arrow). C, Axial fast imaging employing steady-state acquisition (FIESTA) MR image shows lesion is homogeneously hyperintense compared with liver parenchyma. Image courtesy of American Roentgen Ray Society (ARRS), American Journal of Roentgenology (AJR)

News | Magnetic Resonance Imaging (MRI) | October 21, 2020
October 21, 2020 — According to an artic...
According to an inquest, a man with a heart disorder and chest pain died two days after a doctor viewed the wrong scan and sent him home
News | Computed Tomography (CT) | October 21, 2020
October 21, 2020 — The BBC News
Flowchart of patient inclusion and exclusion.

Figure 1. Flowchart of patient inclusion and exclusion.

News | Coronavirus (COVID-19) | October 20, 2020
October 20, 2020 — A new multi-institutional study published in the journal ...
Rensselaer, First-Imaging, and GE Global researchers develop a deep neural network to perform nearly as well as more complex dual-energy CT imaging technology
News | Computed Tomography (CT) | October 20, 2020
October 20, 2020 — Bioimaging technologies are the eyes that allow doctors to see inside the body in order to diagnos
Ezra, a NY-based startup transforming early cancer screening using magnetic resonance imaging (MRI), announced that it has received FDA 510(k) premarket authorization for its Artificial Intelligence, designed to decrease the cost of MRI-based cancer screening, assisting radiologists in their analysis of prostate MRI scans. It is the first prostate AI to be cleared by the FDA.
News | Artificial Intelligence | October 20, 2020
October 20, 2020 — Ezra, a NY-based startup transforming early cancer screening using...
Lesion was originally reported as indeterminate enhancing mass, and outside report recommended biopsy. Classic features of benign hemangioma are shown. Error was attributed to faulty reasoning. A, Axial MR image obtained 5 minutes after contrast agent administration shows peripheral nodular discontinuous enhancement. B, Axial MR image obtained 10 minutes after contrast agent administration shows centripetal progression of enhancement (arrow). C, Axial fast imaging employing steady-state acquisition (FIESTA)

56-Year-Old Woman With Benign Hemangioma: Lesion was originally reported as indeterminate enhancing mass, and outside report recommended biopsy. Classic features of benign hemangioma are shown. Error was attributed to faulty reasoning. A, Axial MR image obtained 5 minutes after contrast agent administration shows peripheral nodular discontinuous enhancement. B, Axial MR image obtained 10 minutes after contrast agent administration shows centripetal progression of enhancement (arrow). C, Axial fast imaging employing steady-state acquisition (FIESTA) MR image shows lesion is homogeneously hyperintense compared with liver parenchyma.

News | Magnetic Resonance Imaging (MRI) | October 16, 2020
October 16, 2020 —