17523937. METHOD TO IMPROVE MODEL PERFORMANCE BY ARTIFICIAL BLENDING OF HEALTHY TISSUE simplified abstract (International Business Machines Corporation)
METHOD TO IMPROVE MODEL PERFORMANCE BY ARTIFICIAL BLENDING OF HEALTHY TISSUE
Organization Name
International Business Machines Corporation
Inventor(s)
METHOD TO IMPROVE MODEL PERFORMANCE BY ARTIFICIAL BLENDING OF HEALTHY TISSUE - A simplified explanation of the abstract
This abstract first appeared for US patent application 17523937 titled 'METHOD TO IMPROVE MODEL PERFORMANCE BY ARTIFICIAL BLENDING OF HEALTHY TISSUE
Simplified Explanation
The patent application describes a method to improve the performance of machine learning algorithms in analyzing human tissue images. The method involves combining two constituent images of human tissue to create an augmented image, and then using this augmented image to train a model.
- The method involves receiving two constituent images of human tissue.
- A portion of the second image is overlapped on a portion of the first image to create an augmented image.
- The augmented image is used as part of a dataset to train a machine learning model.
Potential Applications
This technology has potential applications in various fields, including:
- Medical Imaging: The method can be used to enhance the accuracy of machine learning algorithms in analyzing medical images, such as X-rays, CT scans, or MRI scans.
- Cancer Diagnosis: By improving the performance of machine learning algorithms in analyzing tissue images, this method can aid in the early detection and diagnosis of cancer.
- Drug Discovery: The augmented images can be used to train machine learning models for identifying potential drug targets or predicting the efficacy of drugs on specific tissues.
Problems Solved
The method addresses the following problems:
- Limited Dataset: By creating augmented images from existing constituent images, the method expands the dataset available for training machine learning models, overcoming the limitation of a small dataset.
- Performance Improvement: The method aims to enhance the performance of machine learning algorithms by providing them with more diverse and representative training data.
Benefits
The use of this technology offers several benefits:
- Improved Accuracy: By training machine learning models with augmented images, the accuracy of analyzing human tissue images can be enhanced.
- Enhanced Generalization: The augmented images provide a more diverse representation of human tissue, allowing the machine learning models to generalize better to unseen data.
- Cost and Time Efficiency: Instead of collecting a large number of new images, the method utilizes existing constituent images to create augmented images, saving both time and resources.
Original Abstract Submitted
A method for improving machine learning algorithm performance is described. The method may comprise receiving a first constituent image of human tissue; receiving a second constituent image of human tissue; overlapping a portion of the second constituent image on a portion of the first constituent image to create an augmented image; and training a model using a dataset comprising at least the augmented image.
