The cutting edge of CT

Johns Hopkins Hospital | www.hopkinsmedicine.org

No Comments

Dr. Elliot Fishman, Director of Diagnostic Radiology and Body CT at Johns Hopkins Hospital, on how technological advances in CT scanning improve diagnoses, and lead to better management and cheaper cost.

Advances

For the last two years, 64-slice CT has evolved to become the state-of-the-art of computer tomography (CT). We are talking about scanners that rotate three times a second, that have the spatial resolution of under 0.4 millimeters, and a temporal resolution of under 160 milliseconds. They can scan any part of the body very quickly, and even freeze the heart without motion. We are in an era where data can be acquired that is very thin, typically 0.75 millimeters, and we have isotropic datasets. CT, from the days of EMI, has classically been axial, with the radiologist looking at slice after slice for each patient. While previously we looked at 50 to 100 slices, today we look at volumes of data because we get thousands of slices on a single patient.

The new technology has enabled us to opacify blood vessels at a very high resolution, without motion, and with isotropic data. We are able to create angiograms just like the classic ones that were catheter-based studies. But today we do them for one quarter of the cost, with the same or greater accuracy, and with none of the complications or potential complications. We are creating angiographic maps of vital organs like the pancreas, kidney or liver that determine whether a patient with cancer is resectable for cure. If a patient is resectable, the surgeon uses the images for 3D planning. Chest pain patients with atypical chest pain who typically would get admitted to the hospital, get worked up and usually get a nuclear stress test, today are able to go directly to CT. It has been shown that CT in the cardiac patient has a 99 percent negative predictive value, which means when normal, you can send the patient home with confidence that the symptoms are not cardiac related.

Benefits

CT is a non-invasive, low-risk study that takes typically around 10 seconds, replacing more invasive, expensive, and high-risk studies. And yet you don’t decrease the quality of the answers, you usually increase the information you can get. This helps the patient because doctors are able to make diagnoses faster and more accurately. It also helps the health system because we are making better diagnoses, which leads to better management, at a lower cost.

Disease areas

64-slice CT has been implemented at Johns Hopkins Hospital across a range of disease areas. The volume goes up 10 to 20 percent every year. CT is becoming in many ways a physical examination. For example, patients with pancreatic masses will not get operated on until they had a 3D CT scan. We are also doing an increasing number of cardiac patients. And we are doing virtual colonoscopy. Patients with abdominal pain, acute abdomen, appendicitis, or diverticulitis are routinely sent
to CT. And so are patients with pulmonary issues, whether it’s interstitial lung disease or unexplained shortness of breath. CT has replaced nuclear medicine studies for suspected pulmonary embolism. Classic nuclear medicine studies lacked sensitivity and specificity, where CT has a very high sensitivity and specificity. An important thing is that we are replacing studies with MDCT, which provides more accurate information, not just adding but one more exam. We are able to save cost particularly with the studies being more accurate and more timely on a 24/7 basis. It’s really changing how we practice medicine.

Prevention

Many operations that aren’t necessary can now be prevented. In the past, it was felt to be reasonable that even if you were a good surgeon, you might operate on 20 to 30 percent of patients who turn out to have a perfectly normal appendix. Now it’s very rare for patients with a normal appendix to undergo surgery. With newer scanners, we are now looking at tumor perfusion. We are looking at dual energy, which will give better tissue characterization. And we are looking at a number of things that will potentially improve our diagnostic capabilities even further.

Potential

It is phenomenal that although CT is very good, it is even getting better. If you look at scanners, they are getting faster. There are new technologies being developed even within CT. For the most part, CT looks at anatomy. But now we are looking at things where we use PET CT, for example, or tissue perfusion. With new software, and new scanners, you will be able to look at the perfusion of tumors. You may be able to potentially determine the tumor type. With PET/CT you may be able to determine drug and chemotherapy response as early as the first dose of chemotherapy. We are going to see continued growth in the use of MDCT. The better CT gets, the more applications we will see in practice.

Cost-efficiency

There is a tendency to say that new technologies automatically cost more to the health system. However, although CT scanners are getting better, they have not increased in price. In 1980, the typical CT scanner cost a little over $1 million. This is still the case today. Studies that have looked at cost issues concluded that CT scanning is cost-effective because one of the most expensive things is having somebody in the ER or in the hospital. If you use CT instead of the standard tests you will save millions of dollars a year in the typical emergency room as well as in the hospital setting.

Challenge

The challenge that remains in trying to make advances in the field of CT scanning is creating new scanners that are faster with higher resolution and that can see smaller structures. But CT uses radiation. We want to minimize radiation dose to patients. So the challenge going forward is to develop new scanners and new technologies while decreasing the dose of radiation to patients.

Education

Technology is crucial. But you need to know how the technology works in practice. I help develop ways to train other people along with computer programs that help make the process easier. On our website www.ctisus.com, which is used by about 60,000 plus people worldwide, we provide information, vodcasts and podcasts, for free. I have also run over 100 CME courses for Johns Hopkins. From the public’s perception, the exciting thing is that technology, when used correctly, has the opportunity to advance public health, to really improve their daily lives.

Dr Elliot Fishman: “With an interest in medical imaging and medical post-processing, Dr Elliot Fishman has developed 3D Imaging since the mid-80s. He is focusing on cardiac CT, earlier diagnosis and better diagnosis, and better management. He is also concentrating his efforts on vascular imaging, CT angiography for oncology applications, earlier detection of tumors, better staging of tumors, and helping manage
patients surgically. Dedicated to education, he offers up-to-date information on his webiste www.ctisus.com.”

SOMATOM Definition Scanner: “The Somatom Definition dual source CT scanner is using two x-ray tubes and two detector rows at the same time. It is more powerful and has a higher resolution. Faster than every beating heart, it delivers full cardiac detail at half the radiation dose.”

CT Market Snapshot: US Sales – Total Market

Revenues (2006)

up to 10 Slice: $153.5 million
16 Slice: $413.4 million
32/40 Slice: $194.1 million
64 Slice: $933.4 million
Total: $1,694.3 million

Units Shipped (2006)

up to 10 Slice: 363
16 Slice: 523
32/40 Slice: 180
64 Slice: 734
Total: 1800

Source: Frost&Sullivan Medical Imaging Industry Outlook 2006 - 2007

CT Demand Analysis (2005)

Source: Frost&Sullivan Medical Imaging Industry Outlook 2006 - 2007

CT Technology trends

Wider detector coverage
o State of the art: 64-slice CT has 3.2 to 4.0 centimeters detector width
o Work-in-progress: 128-slice (expected 2007-2008) and 256-slice CT (expected 2008-2009) for up to 16-cm-wide detector coverage
Multi-source CT
o Industry-first: Siemens dual source CT
o With gantry rotation speed near its maximum, this solution may be the only way to improve temporal resolution
Workflow
o Enterprise distribution of images and clinical applications, PACS
o Post-processing software: Computer-aided reading (CAR) and Detection (CAD), Advanced visualization and 3-D/4-D imaging
o More efficient scanner software
Improved clinical applications, in particular cardiac imaging packages
Dose reduction / efficiency
o Improvements in hardware allow faster procedures, noise reduction
o Advanced techniques and algorithms: Automatic exposure control, tube current modulation, ECG pulsing
Multi-energy imaging
o Using different tubes, e.g. Siemens’ dual-source CT
o Using a single X-ray source and multi-energy detector: Work in progress, e.g. Philips, GE
Tomosynthesis using flat-panel detector (FPD) technology expected to merge with computed tomography (CT) technology in the long-term, pending improvements in FPD technology

Making the move to a PACS

EHM spoke to Doug Ricci, Manager of Johns Hopkins Medicine Image Archive, about the benefits and challenges of switching to a Picture Archive and Communication Systems (PACS).

PACS is an electronic imaging archive. Could you explain PACS in more detail?

It was developed in the early 1980’s, when physicians and scientists began working together on new applications as accessories to the new digital modalities such as CT and MRI. PACS was originally developed for the Department of Defense and the Veterans Affairs agency using federal funds to improve the continuity of care for military deployments. The idea for PACS came out of the need to easily view, store, and manage images from those new digital modalities that were beginning to be used extensively. Film digitizing technologies were also developed, which made it possible to make copies of patient x-rays, and to store them along side of the digital images. Today, the focus is on improving productivity by reducing turn around time with faster retrievals, faster transfers, easier use, more portability, and on more advanced features like 3D reconstruction, teaching tools, and multi-institutional support. Radiologists and referring physicians are now using PACS in filmless environments, and are demanding much better service than was possible in the past.

What are the benefits of PACS for clinicians and for patients?

Improved patient outcomes are realized by faster reading and electronic reporting. PACS reduces the turn-around time between when the patient is imaged and when that exam is available to Radiologists for reporting, and to the Clinicians in the operating rooms and in the hospital wards for viewing the images and the reports. And so, decisions can be made much earlier.

What are the challenges involved in making a move to a PACS?

The initial move to a PACS is expensive and very challenging from a workflow and training perspective. It is also difficult to reign in the tendency for the scope to expand. For example: We had experimented with a mini-PACS for several years before we finally purchased a larger system for the new Oncology Center in the Fall of 2000. This system was fully integrated with our information systems, and the quality of the image, the accuracy of the information, and the widespread availability were very compelling. The demand to expand services exploded and far exceeded the ability of the capital administrators to allocate funds to expand the infrastructure. The result was a system that was too small and ill-equipped to handle the demand. Small less expensive mini-PACS devices were purchased by several divisions within our department as band-aids because the large integrated PACS was undersized and could not support their needs. This increased the complexity of the imaging network and reduced reliability within the environment, which took years and a great deal of money to fix.

What advice would you give to anyone who is making the move to a PACS?

It is very important that the there is a commitment to fund a full PACS implementation plan even though it may only be partially implemented in the beginning.This includes commitment for power and network infrastructure, people, and space. You will probably find that it will be bumped up higher in priority when the benefits are realized, and so the institution will need to have already agreed to the plan so that they may move more quickly to the next phases.

You will learn a lot by going through a selection process. Form a small selection team, write a Request for Proposal (RFP), send it to vendors, go on some site visits, and get to know the potential vendors and their customers. Ask some really tough questions. If something is unclear or doubts are raised get a commitment in writing from the vendor how they will address that issue and when. You may benefit from hiring a consultant to navigate through the process and to help document your particular workflow eccentricities before writing an RFP or to help in the vendor selection process.

Once a vendor is selected the contract negotiations can be gruelling and cause delays especially at large institutions. This is where your ongoing costs and exposure are contained, however, and where your PACS plan is committed to. Pay close attention here.

The implementation and rollout can be very complex. It is important to prepare the users for the change and to keep the communication channels open with them and their divisions.

What background and qualities does a PACS administrator need to possess?

The best PACS administrators tend to be Jacks of All Trades that have strong skills in networking, server, and storage technologies, have a solid understanding of clinical workflows. They have a strong commitment to provide excellent service, the ability to listen, and the empathy to go the extra mile to solve problems for individuals. Depending on the size of your organization a PACS could require anywhere from two to 20 or even more people to support it properly. Small organizations may have one person trained and savvy with computer skills. The other could be a dedicated Radiology technologist or clinical specialist who intimately understands the workflow and the needs of the users. They must be cross trained, however, to understand the bigger picture and that the focus is on performance and on patient care. Larger organizations tend to have more specialized people that are experts in specific areas, such as systems, networking, storage, user communications, desktops, monitors, applications, interfaces, imaging modalities, information systems, databases, reporting, workflow, user training, hanging protocols, data integrity, quality control, and others.

Where do you see the future of PACS?

The trend is for more features, more integration with other systems, more vendor consolidation, more “ologies” supported, and more organizations under one structure. Secure web enabled viewing and reporting is in high demand now as well. Doctors are demanding to be able to review images and reports and to dictate their results for exams in their offices and at their homes. They no longer wish to be tied down to a particular reading station all day long.

What is your next professional endeavor?

In a joint venture with another department at Johns Hopkins we are building a new archive initially for research customers with scaleable software that has a sophisticated security infrastructure built in. It is capable of managing data for multiple customers in multiple organizations each with individual Patient Identifiers, Service Level Agreements, retention policies, applications, users, and information systems. For the enterprise customers who are looking to go beyond their own firewalls this could be the next big thing.

Doug Ricci: “For the past nine years, Doug Ricci has managed several Picture Archive and Communication Systems (PACS) both for clinical and research customers inside and outside of radiology. Currently, he is looking after more than one million exams replicated in two locations, which corresponds to over 170 terabytes of data stored on spinning disks. Ricci also support Johns Hopkins Second Opinion and Teleradiology Consult services.”

Workstation: “A reading workstation in Johns Hopkins Hospital’s Diagnostic Division. It has a monitor for the Digital Dictation system on the left, two high resolution greyscale monitors for the PACS viewer in the center, and a color monitor for the PACS worklist and the homegrown report viewer on the right. The patient and study context is integrated between all three systems.”

The Magazine

Issue 1

Blog

Sales and the 'Talent Magnet'

White Papers