Character information identification method for medical image and device thereof
A technology of medical imaging and text information, applied in character and pattern recognition, instruments, computer components, etc., can solve problems such as errors, increased workload of medical staff, and delay of time for patients to take films, so as to improve recognition efficiency and improve The effect of accuracy
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Example Embodiment
[0021] Example one
[0022] Such as figure 1 As shown, this embodiment is a schematic flow chart of the method for recognizing text information in medical images. The method includes the following steps 110 to 140.
[0023] Positioning step 110: receiving an input of an operator selecting a text area on a medical image, and determining a text area to be recognized for text information according to the input.
[0024] In step 110, it is possible to detect the area selected by the operator on the medical image, and then use the detected area as the text area to be recognized for text information. For example, provide the operator with a frame selection tool such as a mouse or touch Carry out the frame selection of the area. Or, by providing example input text for the operator to input the information of the medical image that he observes, combining the information to determine the text area to be recognized by the text information, for example, the information corresponding to the upp...
Example Embodiment
[0032] Example two
[0033] This embodiment is an improvement made on the basis of the first embodiment, such as figure 2 As shown, in addition to the same steps 110 to 150 as in the first embodiment, it also includes a learning step 170 and / or a receiving step 160.
[0034] As described in the first embodiment, in addition to the recognized text that has passed the verification, the verification result can also be an output that prompts the operator that the recognition is wrong. Regarding the former (that is, the recognized text that has passed the verification), this embodiment receives the operator’s feedback on the output. The feedback includes the operator’s positive feedback and representation recognition based on the recognized text and the actual text on the image. Wrong negative feedback, according to the positive feedback and negative feedback, determine the positive sample and negative sample of the recognized text that passed the verification, and then use the known c...
Example Embodiment
[0036] Example three
[0037] This embodiment is an improvement made on the basis of the first embodiment or the second embodiment. Here, the improvement on the first embodiment is taken as an example for description, and the improvement can also be implemented in a similar position of the second embodiment. Such as image 3 As shown, in addition to the same steps 110 to 150 as in the first embodiment, this embodiment also includes a re-identification step 180 (this re-identification step 180 can also be placed between the identification step 130 and the correction step 140 in the second embodiment) . In the re-recognition step, the operator’s feedback on the recognition of the initially recognized characters output from the recognition step is received. If the feedback is lower than the preset minimum recognition threshold condition, the operator is prompted to re-execute the positioning step. The actual text on the image gives feedback that characterizes the recognition accur...
PUM

Abstract
Description
Claims
Application Information

- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.
© 2023 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap