Method and system for the automatic analysis of an image of a biological sample

Simple SummaryContent extracted from patent full text and abstract with AI.

This patent describes a method and system for automatically analyzing images of biological samples (such as tissue or cell images) to detect pathological relevance, for example, the presence of cancer. The system extracts and processes local image features, aggregates them into global descriptors using advanced 'bag of visual words' techniques (inspired by text analysis), applies machine learning models with adaptively tuned similarity measures, and produces both a single confidence score per region and a detailed heatmap showing which pixels most contribute to the pathological classification. This allows for high precision and transparency in automated diagnosis.

Use CasesContent extracted from patent full text and abstract with AI.

• Automated analysis of histological tissue images for cancer detection in pathology labs.
• Assisting radiologists or pathologists in identifying and highlighting suspicious regions in CT, MRI, or PET scans for various diseases.
• Screening of liquid biopsies (such as blood or other body fluids) for abnormal cells or disease markers.
• Second-opinion support systems for medical professionals, increasing consistency in diagnoses.
• Analysis and visualization of neurological scans (e.g. PET, MRI) to assess disorders or abnormalities.
• Digital pathology platforms integrating AI-driven decision support and explainable heatmaps.
• Large-scale screening in population health initiatives or remote diagnostics.

BenefitsContent extracted from patent full text and abstract with AI.

• Enables faster and more consistent analysis of large volumes of medical images, reducing workload for clinicians.
• Improves diagnostic accuracy by highlighting specific regions of interest, aiding clinical decision-making.
• Uses adaptive similarity and classification methods for robust and reliable results across varied sample types and imaging conditions.
• Provides pixel-wise heatmap explanations, increasing trust and interpretability for users and facilitating identification of subtle pathologies.
• Supports a wide range of biological sample types and imaging modalities (tissue, fluids, CT/MRI/PET, etc.).
• Facilitates the development of second-opinion or assistive diagnostic tools, potentially reducing human error.
• Permits easy integration into existing digital pathology and AI workflows due to modular software implementation.

Technical Classifications (CPCs)

Inventors & Applicants

Patent Abstract

Method for the automatic analysis of an image (1, 11, 12, 13) of a biological sample with respect to a pathological relevance, wherein f) local features of the image (1,11,12,13) are aggregated to a global feature of the image (1,11,12,13) using a bag of visual word approach, g) step a) is repeated at least two times using different methods resulting in at least two bag of word feature datasets, , h) computation of at least two similarity measures using the bag of word features obtained from a training image dataset and bag of word features from the image (1, 11, 12, 13) i) the image training dataset comprising a set of visual words, classifier parameters, including kernel weights and bag of word features from the training images, j) the computation of the at least two similarity measures is subject to an adaptive computation of kernel normalization parameters and / or kernel width parameters, f) for each image (1,11,12,13) one score is computed depending on the classifier parameters and kernel weights and the at least two similarity measures, the at least one score being a measure of the certainty of one pathological category compared to the image training dataset, g) for each pixel of the image (1,11,12,13) a pixel-wise score is computed using the classifier parameters, the kernel weights, the at least two similarity measures, the bag of word features of the image (1, 11, 12, 13), all the local features used in the computation of the bag of word features of the image (1, 11, 12, 13) and the pixels used in the computations of the local features, h) the pixel-wise score is stored as a heatmap dataset linking the pixels of the image (1, 11, 12, 13) to the pixel-wise scores.