Image analysis – Applications – Biomedical applications
Reexamination Certificate
1999-03-16
2002-04-02
Mancuso, Joseph (Department: 2621)
Image analysis
Applications
Biomedical applications
C128S922000
Reexamination Certificate
active
06366683
ABSTRACT:
FIELD OF THE INVENTION
This invention relates to data entry systems, and more particularly to systems for generating reports of image analyses.
BACKGROUND AND SUMMARY OF THE INVENTION
The process of medical image interpretation has been fundamentally unchanged for decades. The radiologist or other medical imaging professional sits or stands while visually examining images displayed on a stationary light box or a film alternator (a motor driven machine that allows the display of 50 or more sets of filmed images). While studying and interpreting the images, the radiologist typically dictates findings and conclusions from the examination into a recording device. Smaller practices often use a tape recorder and an in-house medical typist; larger practices may use telephone lines to transmit dictations to a contracted off-site transcription company. After transcription, the report is made available, either on paper or in electronic form, to the interpreting radiologist for editing and subsequent signature. This signed report serves as the primary work product of radiologists and other imaging specialists.
Although medical image interpretation comprises a large part of many medical specialties, including cardiology, pathology, gastroenterology, and surgery, radiology is used to herein illustrate the prototypical medical interpretation process. The practice of radiology is the most image-intensive medical specialty—most radiology practices generate income almost exclusively from the production and interpretation of medical images.
Referring physicians who request that the radiologist perform an interpretation of the image, health care payors, and courts view the imaging report as an official document that is the final product of an imaging examination. Because the dictated report is the formal means of communication between the radiologist and the referring physician and because its findings and conclusions are often the basis for vital medical treatment decisions, its accuracy and timeliness are essential. Many health care organizations, including the Joint Commission on Accreditation of Hospital Organizations (JCAHO), require that a final written report of each imaging examination be available in the patient's permanent record in a timely fashion (usually 24 hours). Some health care payors have considered implementing this requirement as a condition of reimbursement for professional imaging services.
Many radiology practices routinely fail to achieve this benchmark for report finalization. As a result, shortening the report dictation-transcription-signature cycle has been the subject of substantial concern and study. Despite this heightened level of concern, report finalization times have been resistant to change.
There are several important limitations of conventional reporting and transcription. Transcription services are costly, contributing substantially to the overhead of radiology practice revenues. The purchase and maintenance of recording hardware, electronic communication lines, modems, and information systems components add to this cost.
Transcribed preliminary reports often are not available in a timely fashion. Recorded dictations are subject to accidental erasure or loss. Electronic or phone-based systems sometimes lose data, or capture dictations with poor data integrity, resulting in inaccurate or unreported imaging examinations. These problems delay the final report, thereby frustrating the referring physician, delaying or preventing reimbursement for the examination, and degrading practice efficiency. Some phone-based systems allow a referring physician to hear an unedited preliminary dictation created by a trainee radiologist. But these services are used infrequently because the report is not final, and must be accessed in linear fashion, requiring the referring physician to listen to the entire report from the beginning, with limited ability to rewind or fast-forward.
Transcriptionists have a significant error rate, with a significant fraction of these errors being substantive, such as errors of missing or incorrect information that would have led to unnecessary treatment or testing, or that could have caused risk of complications or morbidity for the patient.
Report signature causes additional delays. Even when preliminary (unedited) reports are available on a radiology information system, they are often of limited utility to the referring physician. Transcription errors are frequent. At teaching institutions, these preliminary reports are usually dictated by trainees, and have not yet been reviewed, edited, and signed by the senior physician legally responsible for the report.
The text report often does not meet the needs of referring physicians. Reports sometimes do not in the opinion of referring physicians address the clinical question, or may be confusing. A multitude of different terms may be used to describe a single common abnormal finding leading to further confusion.
The text report is frequently vague, incomplete, or inaccurate. Although the American College of Radiology has developed standards for the diagnostic radiology report and encourages “precise anatomic and radiological terminology to describe the findings accurately,” no adequate terminology or lexicon has been developed to capture the semantic content of a diagnostic imaging report. For example, one study found that three existing lexicons, the Unified Medical Language System (UMLS) Meta-thesaurus, the Systematic Nomenclature for Medicine (SNOMED), and the American College of Radiology (ACR) codes, contained only 14%, 23%, and 19%, respectively, of the terms needed to capture the meaning of a specific type of imaging report.
A text report has minimal utility for subsequent practice management and data mining. Even in the best of circumstances, when the transcribed and signed report is available on a radiology information system, only an unstructured text format is available. Text-based searches of those systems can be time consuming, and have low accuracy in retrieving a desired subset of reports. Because the semantic content is not stored or indexed in a semantically-coherent fashion, expensive and inaccurate post-processing or re-coding of the data is required if additional analyses are required. The lack of structure and consistency of a free-form dictated or written report also leads to high transmission and storage costs, and the inconsistencies may lead to vague and unclear reports.
Computer software systems have been created to provide a substitute for dictation, but these have been found to suffer limitations. For instance, U.S. Pat. No. 5,267,155 to Buchanan et at describes a system in which a document template includes “boiler plate” text, with “holes” to be filled in by a physician based on an examination. While useful for certain simple needs, this is insufficiently flexible and expressive to meet the standards for certain more complex radiology reports, for instance, as well as numerous other image analysis reports. These more complex reports may involve multiple images with multiple findings, arbitrary ordering of sentences, as well as complex causal linkages and degrees of belief or confidence, which existing systems can not convey. The multitude of possible terms and findings for such applications would require an intractably large number of templates, with locating an appropriate template taking far more resources than simply generating a manual report.
The embodiment disclosed herein overcomes these disadvantages by providing a system and method of generating an image analysis report relating to an image of a subject. The method includes querying a user for an image description providing locational information regarding the image, querying the user for a physical description providing locational information regarding the subject, and querying the user for a finding description providing selected characteristics of the image. A description set including the user's responses to the queries is stored and processed to generate the report. Any of the querying steps ma
Bennet K. Langlotz PC
Langlotz Bennett K.
Mancuso Joseph
Tabatabai Abolfazl
LandOfFree
Apparatus and method for recording image analysis information does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Apparatus and method for recording image analysis information, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Apparatus and method for recording image analysis information will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2902640