Chained Throat Image Model and Classifier for Rapid Infection Diagnosis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for diagnosing viral and bacterial infections, such as COVID-19 and streptococcal infections, are slow, inaccurate, and pose risks to healthcare providers, particularly in children, necessitating a more reliable and fast detection process.
Innovation Solution
A system that combines image data from a subject's throat with clinical factors using a chained model comprising an image data model and a classifier to predict disease states, utilizing machine learning techniques like convolutional neural networks and supervised learning to analyze images and clinical data for accurate infection detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional culturing methods are used to diagnose infections, then accuracy can be maintained, but the diagnosis process becomes slow and complex
Solution Approach 1:
The patent replaces the mechanical biological culturing process with an optical imaging and machine learning system. The throat image capture device captures images of the subject's throat, and a chained model comprising an image data model and classifier automatically analyzes the images to predict disease states, eliminating the need for time-consuming in vitro culturing while maintaining diagnostic accuracy.
Solution Approach 2:
The patent creates a digital copy of the throat examination process through image capture and processing. Instead of physically culturing samples, the system captures visual information through imaging devices and processes these digital images through machine learning models to achieve rapid diagnosis without the temporal delays of traditional methods.
2Reliability
If traditional culturing methods are used to diagnose infections, then diagnostic capability is maintained, but the process becomes difficult to administer properly, especially in children
Solution Approach 1:
The patent replaces the complex mechanical process of sample collection and handling with a simple optical imaging system. The throat image capture device can be easily positioned to capture images of the subject's throat without requiring invasive procedures or complex sample handling, making the test much easier to administer, particularly to children who may be uncooperative.
Solution Approach 2:
The system enables self-service diagnosis capabilities where subjects can undergo examination without requiring skilled technicians to perform complex culturing procedures. The automated image capture and analysis process reduces the need for trained personnel to handle biological samples, simplifying the administration process.
3Reliability
If traditional culturing methods are used to diagnose infections, then diagnostic thoroughness is maintained, but healthcare providers administering the test are put at risk
Solution Approach 1:
The patent replaces direct contact with biological samples through culturing with non-contact optical imaging. The throat image capture device obtains diagnostic information through imaging without requiring healthcare providers to handle potentially infectious materials, thereby eliminating exposure risks while maintaining diagnostic reliability through automated analysis.
Data Source
Figure 1
Figure 2
Figure 3A~3C
AI summary
A method for determining a disease state prediction, relating to a potential disease or medical condition of a subject, includes accessing a set of subject images, the subject images capturing a part of a subject's body, and accessing a set of clinical factors from the subject. The clinical factors are collected by a device or a medical practitioner substantially contemporaneously with the capture of the subject images. The subject images are inputted into an image data model to generate disease metrics for disease prediction for the subject. The disease metrics generated by the image data model and the clinical factors are inputted into a classifier to determine the disease state prediction, and the disease state prediction is returned.