Systems and methods for automated and interactive analysis of bone scan images for detection of metastases
Inventors
Sjöstrand, Karl Vilhelm • Richter, Jens Filip Andreas • Edenbrandt, Lars
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12414748-B2
Publication Date
2025-09-16
Expiration Date
Abstract
Presented herein are systems and methods that provide for improved computer aided display and analysis of nuclear medicine images. In particular, in certain embodiments, the systems and methods described herein provide improvements to several image processing steps used for automated analysis of bone scan images for assessing cancer status of a patient. For example, improved approaches for image segmentation, hotspot detection, automated classification of hotspots as representing metastases, and computation of risk indices such as bone scan index (BSI) values are provided.
Core Innovation
A processor accesses a bone scan image set for a human subject obtained following administration of an agent and automatically segments each image to identify skeletal regions of interest corresponding to particular anatomical regions of a skeleton. The skeletal regions of interest include at least one of a femur region and a humerus region, thereby obtaining an annotated set of images. Atlas-based segmentation is implemented by comparing each member of the bone scan image set with a corresponding atlas image set and registering the atlas image with the image using identified knee region and/or elbow region landmarks so that the atlas identifications are applied to the image set.
The processor automatically detects an initial set of one or more hotspots, where each hotspot corresponds to an area of elevated intensity in the annotated set of images. The detection uses intensities of pixels in the annotated set and region-dependent threshold values, including values associated with the femur region and/or the humerus region that provide enhanced hotspot detection sensitivity to compensate for reduced uptake of the agent therein. For each hotspot in the initial set, the processor extracts hotspot features and calculates a metastasis likelihood value corresponding to a likelihood of the hotspot representing a metastasis based on the hotspot features.
The processor causes rendering of a graphical representation of at least a portion of the initial set of hotspots for display within a graphical user interface (GUI). In a further workflow, the processor receives via the GUI a user selection of a second subset of the initial set of hotspots and calculates one or more risk index values for the human subject based at least in part on a computed fraction of the skeleton occupied by the second subset of hotspots. A system implementation comprises a processor and memory with instructions to perform the access, segmentation, hotspot detection with region-dependent thresholds, metastasis likelihood calculation from extracted features, GUI rendering, user selection reception, and risk index value calculation.
Claims Coverage
The provided dataset includes four independent claims. Across these claims, the core inventive coverage centers on atlas-based segmentation using landmark registration for femur/humerus regions, hotspot detection using region-dependent pixel-intensity threshold values with enhanced sensitivity for femur/humerus, hotspot feature extraction and metastasis likelihood calculation, and GUI rendering with optional user selection to compute risk index values from a fraction of the skeleton occupied by selected hotspots.
Atlas-based skeletal region segmentation using knee and/or elbow landmarks
Automatically segmenting each image by comparing each member of the bone scan image set with a corresponding atlas image, identifying skeletal regions of interest including a femur region comprising at least a portion of a knee region and/or a humerus region comprising at least a portion of an elbow region, and registering the corresponding atlas image with the image using the identified knee region and/or elbow region as landmark(s) to apply the atlas identifications to the image of the bone scan image set.
Region-dependent pixel-intensity threshold hotspot detection with femur/humerus sensitivity compensation
Automatically detecting an initial set of one or more hotspots by identifying hotspots using intensities of pixels in the annotated set of images and one or more region-dependent threshold values, wherein the region dependent threshold values include one or more values associated with the femur region and/or the humerus region that provide enhanced hotspot detection sensitivity in the femur region and/or the humerus region to compensate for reduced uptake of the agent therein.
Hotspot feature extraction and metastasis likelihood calculation per hotspot
For each hotspot in the initial set of hotspots, extracting a set of hotspot features associated with the hotspot and calculating a metastasis likelihood value corresponding to a likelihood of the hotspot representing a metastasis based on the set of hotspot features associated with the hotspot.
GUI rendering and user selection of a second subset of hotspots
Causing rendering of a graphical representation of at least a portion of the initial set of hotspots for display within a graphical user interface (GUI) and receiving via the GUI a user selection of a second subset of the initial set of hotspots.
Risk index calculation based on fraction of skeleton occupied by selected hotspots
Calculating one or more risk index values for the human subject based at least in part on a computed fraction of the skeleton of the human subject occupied by the second subset of hotspots.
The independent claims collectively cover an automated lesion marking and quantitative analysis workflow for nuclear medicine bone scan images that uses atlas-based segmentation with knee and/or elbow landmark registration, region-dependent hotspot detection thresholds with femur/humerus sensitivity compensation, per-hotspot feature extraction leading to metastasis likelihood values, and GUI-based hotspot visualization with optional user selection that enables risk index value calculation from the fraction of skeleton occupied by selected hotspots.
Stated Advantages
Provides enhanced hotspot detection sensitivity in the femur region and/or the humerus region to compensate for reduced uptake of the agent therein.
Improves hotspot detection linearity and hotspot detection performance.
Documented Applications
Automated and interactive analysis of nuclear medicine bone scan images to detect lesion-like hotspots and classify hotspots as metastasis with metastasis-likelihood scoring, including computation of patient-level bone scan index (BSI)/risk index values based on skeletal hotspot burden.
Atlas-based and GUI-enabled workflow integrated with clinical imaging workflows via GUI/PACS and cloud implementation, including generating reports and evidence of BSI performance using simulations and metastatic patient repeat scans.
Metastatic patient repeat scan analysis to evaluate repeat scan reproducibility.
Interested in licensing this patent?