Owkin. has filed a patent for a method and apparatus to determine regions of interest in histology images. The method involves tiling images, extracting features using a trained feature extractor, clustering features to assign tiles to regions of interest, and outputting the results. The claim details a method for training the feature extractor using contrastive loss. GlobalData’s report on Owkin gives a 360-degree view of the company including its patenting strategy. Buy the report here.

Smarter leaders trust GlobalData

Report-cover

Data Insights Owkin Inc - Company Profile

Buy the Report

Data Insights

The gold standard of business intelligence.

Find out more

According to GlobalData’s company profile on Owkin, AI assisted radiology was a key innovation area identified from patents. Owkin's grant share as of January 2024 was 14%. Grant share is based on the ratio of number of grants to total number of patents.

Training feature extractor for histology image analysis

Source: United States Patent and Trademark Office (USPTO). Credit: Owkin Inc

The patent application (Publication Number: US20240020958A1) describes a method and system for training a feature extractor using histology images. The method involves receiving a training set of annotation-free histology images, tiling them into smaller tiles, performing data augmentation, extracting features from these tiles, and training the feature extractor using a contrastive loss to bring matching pairs of tiles closer and different pairs further apart. This approach aims to improve the accuracy and efficiency of feature extraction in histology images for various applications.

Furthermore, the patent application details a method for training a weakly-supervised machine learning model using the trained feature extractor. By applying the feature extractor to histology images with global labels and training the machine learning model using the extracted features, the system can enhance the performance of the model without the need for extensive manual annotations. The system also includes components such as an image processor, feature extractor, and clustering module to facilitate the training process and improve the identification of regions of interest within histology images. Overall, the methods and systems outlined in the patent application offer innovative solutions for optimizing feature extraction and machine learning training in the field of histology image analysis.

To know more about GlobalData’s detailed insights on Owkin, buy the report here.

Data Insights

From

The gold standard of business intelligence.

Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.

GlobalData

GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.

GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies