Computer-assisted tumor response assessment and evaluation of the vascular tumor burden
Inventors
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-10413266-B2
Publication Date
2019-09-17
Expiration Date
Abstract
A computer-implemented method for determining and evaluating an objective tumor response to an anti-cancer therapy using cross-sectional images can include receiving cross-sectional images of digital medical image data and identifying target lesions within the cross-sectional images. For each of the target lesions, a target lesion type and anatomical location is identified, a segmenting tool is activated for segmenting the target lesions into regions of interest, lesion metrics are automatically extracted from the regions of interest according to tumor response criteria, and conformity of target lesion identification is monitored using rules associated with the tumor response criteria, prompting a user to address any nonconforming target lesion. The method also includes receiving a presence/absence of metastases, determining changes in lesions metrics, and deriving an objective tumor response based on the tumor response criteria.
Core Innovation
The invention provides a computer system and computer-implemented method for determining an objective tumor response to an anti-cancer therapy in a patient using cross-sectional images of digital medical image data. The system receives the cross-sectional images, identifies one or more target lesions, and identifies a target lesion type comprising a mass, metastasis, or lymph node as well as a target lesion anatomical location. The system segments the one or more target lesions into one or more regions of interest and automatically extracts one or more lesion metrics from the regions of interest based on the target lesion type and tumor response criteria.
The invention monitors conformity of target lesion identification and selection with one or more rules associated with one or more tumor response criteria. Monitoring conformity of target lesion selection comprises determining that the number of selected target lesions does not exceed a maximum total number and does not exceed a maximum number per organ, and determining that the size of each selected target lesion is greater than a minimum size based on the target lesion type. When a violation of one or more rules is identified, the system prompts a user to modify or delete any nonconforming target lesion failing to satisfy the rules.
For response determination, the system tracks the one or more lesion metrics for each target lesion at two or more time points and derives a total tumor burden or a summary of the lesion metrics at each time point using confirmed target lesions and corresponding lesion metrics based on target lesion type and tumor response criteria. The system automatically determines target lesion response at each time point based on a calculated change relative to baseline or nadir, identifies responses of non-target lesions, identifies a presence or absence of new sites of disease and/or new metastases, and derives a final objective tumor response at each time point based on one or more of the target lesion response, non-target lesion response, new metastases status, and the tumor response criteria rules.
The system updates a treatment regimen for the patient based on the derived objective tumor response indicating progressive disease. In additional aspects, the invention generates a summary display that includes objective tumor response and target lesion metrics for identified target lesions across time points to enable more rapid and accurate assessment of a patient's response to anti-cancer therapy, including indications of the presence or absence of one or more new metastases.
Claims Coverage
The partial content provides four independent claims. Across these independent claims, eight main inventive features are explicitly required: cross-sectional image receipt, target lesion identification with lesion type and anatomical location, segmentation into regions of interest, lesion-metric extraction with measurement-type selection by lesion type, monitoring conformity of target lesion selection with tumor-response-criteria rules and user prompt to modify or delete nonconforming lesions, time-point metric tracking and response derivation, incorporation of non-target lesion findings and new metastases status, and regimen update or summary-display generation.
Objective tumor response from cross-sectional images with regimen update
Automatically determining an objective tumor response to an anti-cancer therapy using cross-sectional images, deriving target lesion response across two or more time points relative to baseline or nadir, deriving a final objective tumor response using target lesion response, non-target lesion response and/or new sites of disease/new metastases, and updating a treatment regimen based on progressive disease.
Target lesion classification by type and anatomical location
Identifying a presence or absence of one or more target lesions from the cross-sectional images, identifying a target lesion type comprising a mass, metastasis, or lymph node, and identifying a target lesion anatomical location for each target lesion.
Region-of-interest segmentation and automatic lesion-metric extraction
Segmenting target lesions into regions of interest and automatically extracting one or more lesion metrics from the regions of interest.
Measurement-type rule by target lesion type for lesion metrics
Providing lesion metrics comprising a long axis measurement if target lesions comprise the mass or metastasis, and providing lesion metrics comprising a short axis measurement if the target lesions comprise the lymph node.
Conformity monitoring for target lesion identification and selection
Monitoring conformity of target lesion identification and selection with rules associated with one or more tumor response criteria by determining maximum total number, maximum number per organ, and minimum size based on target lesion type, and prompting a user to modify or delete any nonconforming target lesion failing to satisfy any of the rules.
Longitudinal metric tracking and tumor-burden or summary derivation
Tracking the one or more lesion metrics for each target lesion at each of two or more time points, deriving a total tumor burden or a summary of the one or more lesion metrics at each time point using confirmed target lesions and corresponding lesion metrics, and determining target lesion response at each time point based on calculated change relative to baseline or nadir.
Non-target lesion response and new metastases identification
Identifying presence or absence of one or more non-target lesions, receiving an input for a response for each non-target lesion, receiving a presence or absence of one or more new sites of disease, and deriving a final objective tumor response at each time point based on one or more of target lesion response, non-target lesion response, and new sites of disease or new metastases.
Summary display generation for more rapid and accurate assessment
Generating a summary display comprising the identified one or more target lesions at a first timepoint and the second timepoint, the one or more lesion metrics, and an indication of the presence or absence of one or more new metastases, where the summary display enables a rapid and accurate assessment of a patient's response.
Across the independent claims provided, the document covers systems and methods that identify and type target lesions, segment regions of interest, extract lesion metrics with measurement-type selection by lesion category, enforce tumor-response-criteria conformity rules for target lesion selection with user prompts, compute longitudinal changes and objective tumor response using target and non-target lesion information and new metastases status, and either update the treatment regimen based on derived objective tumor response or generate a summary display to enable more rapid and accurate assessment.
Stated Advantages
Enable a more rapid and accurate assessment of a patient's response to anti-cancer therapy.
Improvements to the treatment regimen for the patient.
Reduce user/reader error and variability.
Enable automated computation of vascular tumor burden (VTB) and optional necrotic tumor burden.
Provide validation results with stronger predictive power for progression-free survival than other criteria.
Documented Applications
Objective tumor response determination to anti-cancer therapy in a patient based on cross-sectional medical images, including updating a treatment regimen when progressive disease is indicated.
Supporting more rapid and accurate assessment of a patient's response to anti-cancer therapy by generating a summary display across two time points.
Anti-angiogenic therapy assessment using a vascular tumor burden derived from restricted pixel-intensity ranges on CT with radiocontrast compensation [procedural detail omitted for safety].
Validation against progression-free survival using vascular tumor burden criteria compared to other criteria such as RECIST, Choi, and MASS, including inter-observer performance and faster assessment time versus manual methods.
Interested in licensing this patent?