The 87th Annual Scientific Assembly and Annual Meeting of the Radiological Society of North America (RSNA 2001)
11/25 Ė 11/30/01, Chicago
By Jingkun Hu, Philips Research USA
The RSNA is the annual event for the medical imaging industry mainly and also imaging related medical information systems such as Hospital Information Systems (HIS), Clinical Information Systems (CIS), Radiology Information Systems (RIS), and Picture Archiving and Communication Systems (PACS). All major product innovations from variety of vendors all over the world are presented here. Also there are scientific sessions in diagnostic radiology for radiologists, integrating the healthcare enterprise (IHE) for medical information technology (IT) professionals. The attendance is huge, over 25,000.
Medical imaging including image processing and enhancing in various modalities such as CT, MRI, and US is the major show as usual. One of the trends is using 3D technology to perform computer-aided-diagnosis. Iíll not address imaging and radiology diagnosis part in this report rather focusing on emerging information technology application in medical IT domain.
As you may not know, medical IT is moving slower than general IT and E-commerce. E.g., it can be seen that a strong trend in distributing patient medical records through Web and about half of the medical information system products are Web supported using ASP, JSP, or other technologies. The red-hot technology Ė XML, is not seen in exhibits yet, but it was presented in several scientific posters and papers. Some of the information systems use voice recognition technology in driving workflow, interpreting radiologistís dictation, but I havenít seen any applications to deliver medical records or test results in voice (or voice XML). I didnít see any pen computing technology applications in the field. Probably there are such applications but they are not shown in this exhibit.
There is a technical workshop on PDAs/Handhelds in viewing medical records including images through wireless connection to central storage systems. Such devices can be used for physicians to view their current or previous medical records while on duty. This is much more efficient and accurate for physicians to make decisions and all in all provides the quality of healthcare.
There is a big demonstration in IHE session illustrating the IHE technical framework of connecting multi-vendor modalities and systems, distributing and managing patient records in heterogeneous environment. The whole demonstration is workflow driven. The main objective is to ease the choice of imaging modalities and management systems without worrying the mismatch among devices or systems from different vendors. The main issues here are actually data integration and interoperability among systems.
This strongly requires medical IT related standardization activities. This is not only a technical battle but also a political battle from different vendors. I believe, eventually an agreement will be made if the concept of open standard format such as XML gradually gets accepted. For example, most of the medical IT vendors are using XML technology now, but the question is how to use it. Even every vendor uses XML data format, it can be still proprietary XML format. So, it could be a very significant research area to propose a standard data representation for medical information.
Multi-modality exams are usually performed on a patient within a radiology department. Or sometimes physicians may need to access patientís previous exam results remotely. Assume physicians know where the medical records are stored (e.g., URL), one problem here is that how to dynamically arrange those records from different locations in one window so that physicians can jump around and compare records. Microsoft has proposed a solution to it by using something called Ďdigital dashboardí. What it does is to divide one big window into sub windows. And each sub window links to a clinical context object. Even for a local RIS, different users have different focuses in viewing patient records. So it will be very efficient to have profile/preference based adaptive user interface.
Speech Recognition (SR) as usual is a very hot research area. One of the main applications in medical IT is to process radiologist/physicianís dictation using SR and then to create a report from it. Such a system was in the show from Medquist, a Philips Medical company. They claimed that the accuracy of this system is close to 100%. As I said, I didnít see any voice XML applications in this show. This is a good research area in medical IT. One example is to distribute medical reports to either physicians or patients using voice XML applications.
There is a strong trend of using PDAs/Handheld devices to retrieving/viewing patient records within hospitals or physicians offices. Several research areas need to be explored such as scalable user interface and wireless networking.
I didnít see any pen computing applications in this show. But there is a strong need in medical IT. One scenario is nurses/physicians taking notes, capturing observations, and recording vital signs. The quality and efficiency of the care will be greatly improved if pen-computing technology is matured and applied in such environment.
Decision support systems in medical IT is very complicated. Currently, decision support in medical IT focuses on the computer-aided-detection.
Information security is also a major concern in medical IT especially when HEALTH INSURANCE PORTABILITY AND ACCOUNTABILITY ACT (HIPAA) is taken into account. Several issues exist in healthcare information systems such as access control, clinical impact of HIPAA privacy, HIPAA compliance, and economics of HIPAA compliance.
1. RSNA site, http://www.rsna.org
2. Digital Dashboard on Microsoft Business, http://www.microsoft.com/business/dd/default.asp
3. A medical transcription information management system, http://www.medquist.com
4. dictating the voice for the future, http://www.dvi.com