Automatic Speech Recognition with Clinical Decision Support
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing methods for transcribing human speech into textual documents are time-consuming and prone to errors, especially in medical and legal contexts, where quick and accurate production of reports is critical for timely decision-making.
Innovation Solution
An automated system that uses an automatic speech recognizer to produce a draft transcript, which is then verified and processed by a decision support system to generate alerts and outputs, ultimately producing a formatted document without encoding for review and action by healthcare professionals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If automatic speech recognition is used to produce transcripts, then productivity is improved, but manufacturing precision deteriorates due to errors in the transcript
Solution Approach 1:
The system implements feedback by having the decision support system verify the coded information against medical knowledge bases and clinical guidelines, automatically identifying and correcting errors in the speech recognition output before finalizing the transcript
Solution Approach 2:
The patent introduces a decision support system as an intermediary between speech recognition and final transcript production. This intermediary layer codes the speech into structured data, validates it against medical knowledge, and only then produces the final transcript, ensuring both speed and accuracy
2Manufacturing precision
If human transcriptionists are used to ensure accuracy, then manufacturing precision is improved, but productivity deteriorates due to limited typing speed
Solution Approach 1:
The patent replaces the mechanical typing process of human transcriptionists with an automated system that uses speech recognition, coding, and decision support validation. This substitution eliminates the speed limitation of human typing while maintaining accuracy through automated verification against medical knowledge bases
3Manufacturing precision
If additional formatting and editing processes are applied to meet specific format requirements, then manufacturing precision is improved, but productivity deteriorates due to additional delays
Solution Approach 1:
The decision support system performs preliminary coding and validation of the speech content into structured data formats that inherently meet medical reporting requirements. By pre-structuring the data according to medical knowledge bases and clinical guidelines before final transcript generation, the system eliminates subsequent formatting delays
4Manufacturing precision
If comprehensive error correction is applied to the transcript, then manufacturing precision is improved, but productivity deteriorates due to additional processing time
Solution Approach 1:
The system performs error correction preliminarily by validating the coded speech data against medical knowledge bases, drug interaction databases, and clinical guidelines before the transcript is finalized. This preliminary validation prevents errors from reaching the final transcript rather than correcting them afterward
Solution Approach 2:
The decision support system provides continuous feedback during the coding process, automatically identifying inconsistencies with medical knowledge and triggering real-time corrections, ensuring accuracy without requiring separate post-processing error correction steps
Data Source
AI summary
Speech is transcribed to produce a transcript. At least some of the text in the transcript is encoded as data. These codings may be verified for accuracy and corrected if inaccurate. The resulting transcript is provided to a decision support system to perform functions such as checking for drug-drug, drug-allergy, and drug-procedure interactions, and checking against clinical performance measures (such as recommended treatments). Alerts and other information output by the decision support system are associated with the transcript. The transcript and associated decision support output are provided to a physician to assist the physician in reviewing the transcript and in taking any appropriate action in response to the transcript.