X-Ray Speech Imaging With Automatic Deviation Frame Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing X-ray imaging systems require radiologists to manually sift through a large number of images to identify relevant frames for speech analysis, increasing their workload, particularly in examinations involving articulation impairment or other speech impairments.
Innovation Solution
An X-ray imaging apparatus that synchronously collects X-ray images and voice data during speech production, associates the data with time stamps, and identifies frames where speech deviates from a predetermined sentence, automatically highlighting these frames for display.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a high imaging rate of 30 fps is used to capture speech movement, then the temporal resolution of speech examination is improved, but the number of X-ray images increases to 300-400 images, making manual review more time-consuming
Solution Approach 1:
The patent extracts only the essential information from the large number of X-ray images by using voice information as a filter. The system identifies and extracts only the images corresponding to abnormal speech moments, separating them from the redundant normal images, thereby reducing the review workload while maintaining examination precision
Solution Approach 2:
The patent introduces voice information as an intermediary between the X-ray images and the radiologist's review process. The voice information serves as a mediator that helps identify which images are relevant, allowing radiologists to quickly locate abnormal speech moments without manually reviewing all images
2Reliability
If X-ray imaging is performed multiple times to examine speech, then the comprehensiveness of speech examination is improved, but the total time required to find necessary X-ray images increases with the number of imaging times
Solution Approach 1:
The patent performs preliminary action by collecting and storing voice information during the X-ray imaging process. This preliminary collection of voice data enables later automatic identification of abnormal speech moments, so that when radiologists review the images, they already have the temporal markers needed to quickly locate relevant frames without needing to search through all images from multiple imaging sessions
3Loss of information
If a large number of X-ray images are acquired during speech examination, then the amount of information available for analysis is improved, but the burden on radiologists to find and review necessary images increases
Solution Approach 1:
The patent extracts only the essential information by using voice information as a filter to identify and select only the X-ray images that correspond to abnormal speech moments. This extraction process maintains the necessary diagnostic information while eliminating the burden of reviewing all images, as radiologists are directed only to the relevant frames
Solution Approach 2:
The patent implements feedback by using voice information to automatically identify and highlight abnormal speech moments in the X-ray images. This feedback mechanism provides radiologists with immediate guidance on which images are most relevant, making the review process easier and more efficient without losing any critical diagnostic information
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Automatically identifies and displays relevant X-ray images where speech deviates from a standard sentence, reducing the radiologist's workload by eliminating the need to manually search through numerous images.
Implementation Method 1
an X-ray irradiator for irradiating a patient with X-rays
Implementation Method 2
an X-ray detector for detecting the X-rays with which the patient is irradiated by the X-ray irradiator
Data Source
AI summary
An X-ray imaging apparatus according to this invention includes an X-ray-image-and-voice-information collector for collecting a plurality of X-ray images based on detection results of X-rays with which a patient is irradiated, and for collecting voice information on the patient's voice input from a voice information input part; an X-ray image preserver for preserving the plurality of X-ray images collected together with time information; a voice information preserver for preserving the voice information collected together with time information; a timing identifier for identifying a timing in which the speech of the patient is different from the examination sentence based on the voice information preserved; an X-ray image identifier for identifying a corresponding X-ray image that corresponds to the timing in the plurality of X-ray images preserved; and a display controller for directing a display to display the corresponding X-ray image.