Image analysis – Applications – Biomedical applications
Reexamination Certificate
2001-05-18
2004-05-11
Johns, Andrew W. (Department: 2621)
Image analysis
Applications
Biomedical applications
C382S305000, C707S793000, C128S923000
Reexamination Certificate
active
06735329
ABSTRACT:
BACKGROUND
The present invention relates to methods of and apparatus for making a history or record, particularly, but not exclusively, of medical treatments or procedures, and to methods of and apparatus for using the history or record. More particularly, it relates to producing and digitizing a number of images to create a digital library of images, producing and providing a digital library of texts corresponding to the images in the library of images, providing an image from outside the digital library, digitizing it and comparing it to the digital images in the digital library, and selecting from the digital library of texts a text which corresponds to the outside image, and using the selected text to produce at least a portion of a record.
In many fields and technologies it is important to create and maintain a record, or history or narrative, of situations, facts, operations or procedures. Such records are important for achieving repeatability and accuracy, for educational or evaluative purposes, and/or for reconstructive purposes. For example, in industrial production, they may be used to detect deviations from specifications and production standards, and/or they may be used to create a “standard operating procedure” (“SOP”). In the legal field, transcripts or records of legal proceedings may be used to revisit or review the propriety of proceedings and decisions. In research or experimental science, laboratory notebooks or journals may reflect and/or evidence results of chemical combinations or hypothesis. In medicine, medical records and transcripts of surgical procedures may be used for diagnosis, to determine subsequent treatments, and/or to determine or assess prognoses. In each of these examples, and in many other instances and fields, records provide a database or library of knowledge, teach, instruct or inform about past practices or products, and/or aid in the detection and elimination of anomalies and inaccuracies.
Making and keeping a record has been, and is, a painstaking process. Typically, it has involved an individual hand writing a description or narrative, which may then be archived or preserved for reference by the individual who created it, or others. Developments in recording technologies, e.g., photography, sound and voice recording and storage, digital storage of information, etc., have somewhat eased the burden of creating a record, but there is room for further improvement.
Turning to the field of medicine, in which the present invention finds particular, but not exclusive, applicability, medical records were typically produced by the care giver writing by hand, for example, in a patient's chart. More recently, a care giver or physician may speak into a sound recording machine, or dictate, a description of a treatment or procedure. The recording or “tape” is then transcribed by others into a written record or chart. Also more recently, photographic records may be used. For example, a surgical procedure may be filmed, and the film may be accompanied by a verbal description dictated or spoken by the surgeon or attending physician while the procedure is taking place. Notwithstanding these advances, problems and inefficiencies remain.
One problem stems from outsourcing dictated medical records, even when such outsourcing is to organizations or people specializing in the transcription of medical records. The dictated item must be communicated to the transcriptionist, and back, and physicians and care givers must review and edit the transcription, leading to inefficiencies in time and handling, and increased costs. Another problem is the time interval or delay between the procedure and the availability of the transcript. This is true even when a physician dictates during a procedure, which itself may interfere with the concentration or performance of the physician. Typically, current medical record procedures still require a physician or surgeon to go to a workstation to dictate a description of a procedure or treatment after it is completed, at least to review a transcript, but in most instances, to dictate a description as well.
There are some attempts to improve the efficiency of producing medical records. For example, there are service providers, e.g., Speech Machines, Inc. and MedQuist, Inc., which specialize in transcribing dictated descriptions of medical treatments. They may use voice or term recognition systems (e.g., Dragon System's “Naturally Speaking,” IBM's “ViaVoice” and Lernout & Hauspie's “VoiceXpress”) wherein a vocal term or phrase is recognized by a computer which then converts it into a word-processing result. U.S. Pat. No. 6,031,526 discloses a system wherein generating electronic and printed medical records provides automatic integration of captured video still images and voice dictated information concerning the image, and wherein a voice recognition module allows the system to respond to voice commands and automatically transcribe the dictated text into a word processing document. Another, generally similar example of such systems is that provided by cMore Medical Solutions, Inc. of Minneapolis, Minn. Although efficiency of making a record may be increased, there are still transmission and handling delays, even when use is made of browser or internet-based systems.
One way to relieve physicians' dictation burden would be to film or photograph a treatment or surgical procedure and use the recorded images to trigger a descriptive text. Such a solution would likely involve a computer or computers. U.S. Pat. No. 4,996,707 (O'Malley et al.) discloses a computer system having the capability to receive and store graphic images which might be useful in such a solution. The system includes software that can digitize images, enabling them to be stored in memory and then accessed and used for various purposes, including identification, printing, or converting the digitized images to speech using a “text-to-speech” module. There is no disclosure or suggestion of digitizing a large number of images to create a digital library of images, providing an image from outside the digital library, digitizing it and comparing it to the digital images in the digital library, and providing for the production of a descriptive text associated with the image from outside the digital library.
Computers and storage and manipulation of data using computers hold promise for improving medical record creation, record keeping and use of stored medical records. For example, U.S. Pat. No. 5,050,220 involves an optical fingerprint correlator wherein a fingerprint is digitized, and then may be compared to a database of fingerprints to try to find a match. Such a correlator could be adapted to identify a patient, pull up the patient's medical record, and compare a current or recent diagnostic image (e.g., an MRI image) to the record. U.S. Pat. No. 5,031,228 discloses another image recognition system and method for identifying a pattern in images, and U.S. Pat. No. 5,668,897 discloses a method and apparatus for imaging and image processing, including digitizing an image and comparing the digitized image against a codebook of stored digital images. None of these patents discloses using an image to select or create a text describing an image.
The use of microprocessors, computers and computer management of data, including images, in the field of medicine is reflected in U.S. Pat. Nos. 5,241,472; 5,261,404; 5,740,802; 5,951,571; 5,961,456 and 6,024,695, the disclosures of which patents are incorporated herein by reference. Typically, the systems and methods disclosed in these patents involve obtaining images, digitizing the images and storing and/or manipulating or using the images, e.g., in U.S. Pat. No. 5,241,472 to create a text file. In U.S. Pat. No. 5,951,571 a computer is used to access a data storage unit containing previously acquired and digitally stored images of a patient. U.S. Pat. No. 5,961,456 is directed to a system and method for using current actual images and computer generated reference images, and the U.S. Pat. Nos.
Johns Andrew W.
Nakhjavan Shervin
LandOfFree
Methods and apparatus for image recognition and dictation does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Methods and apparatus for image recognition and dictation, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Methods and apparatus for image recognition and dictation will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3247623