Monitoring Marks for Image-Based Assay Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Low-quality imaging systems used in image-based assays introduce imperfections that significantly affect accuracy, requiring long training times and large sample sets for machine learning models, and often result in artifacts.
Innovation Solution
Incorporating monitoring marks with predetermined geometric and optical properties on sample holders, allowing for image correction and training of machine learning models using both low-quality and high-quality imaging systems to enhance assay image quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a low-quality imaging system is used to reduce cost and improve portability, then device complexity and cost are reduced, but measurement precision and reliability deteriorate due to imperfections in optical, mechanical, and electrical systems
Solution Approach 1:
The patent introduces monitoring marks as an intermediary element between the low-quality imaging system and the assay sample. These marks serve as reference objects that mediate the correction of imaging imperfections, allowing the system to compensate for its own deficiencies without requiring hardware upgrades
Solution Approach 2:
The patent performs preliminary correction of imaging imperfections by training a machine learning model beforehand using monitoring marks with known geometric and optical properties. This pre-training enables the system to automatically correct distortions in subsequent assay images without real-time intervention
2Measurement precision
If machine learning methods are applied to correct imaging imperfections, then measurement precision improves, but training time and computational resources increase significantly
Solution Approach 1:
The patent enables the machine learning model to self-adjust and correct for imaging system imperfections by learning from monitoring marks with known properties. The system serves itself by automatically adapting to its own deficiencies without requiring extensive external calibration or manual intervention
Solution Approach 2:
The patent changes the training approach by using monitoring marks with predetermined geometric and optical parameters as reference points. This parameter-based training method reduces the complexity of training data requirements and accelerates model convergence compared to traditional approaches
3Device complexity
If traditional machine learning training is used without monitoring marks, then device simplicity is maintained, but measurement precision deteriorates due to artifacts and variable imperfections in imaging systems
Solution Approach 1:
The monitoring marks act as an intermediary reference that bridges the gap between the simple low-quality imaging system and the requirement for high measurement precision. These marks provide known geometric and optical properties that serve as a reference framework for correcting imaging artifacts
Solution Approach 2:
The patent uses monitoring marks with predetermined and known geometric and optical properties as reference copies. By comparing assay images against these known reference patterns, the system can identify and correct distortions without requiring complex hardware modifications
Data Source
AI summary
The present disclosure relates to devices, apparatus and methods of improving the accuracy of image-based assay, that uses imaging system having uncertainties or deviations (imperfection) compared with an ideal imaging system. One aspect of the present invention is to add the monitoring marks on the sample holder, with at least one of their geometric and/optical properties of the monitoring marks under predetermined and known, and taking images of the sample with the monitoring marks, and train a machine learning model using the images with the monitoring mark.