Method and system for dose determination of radiation therapy
AssigneeScientificRT GmbH
InventorBenjamin Sobotta
Life of this patent
Estimating the term…
Abstract
Methods for performing dose determination and cost function gradients in a radiation therapy are disclosed, which include: discretizing a volume-of-interest (VOI) into a set of voxels; identifying a set of beamlets which deposit dose contributions of radiation to the VOI, and each beamlet has a weight factor; transforming the dose contributions into a first domain, and transforming the weight factors into a second domain orthogonal to the first domain; calculate the local derivatives of a cost function of dose and cost function gradients with respect to the weights of the beamlets.
Claims25
A method for performing dose determination, comprising:
- creating a set of volume elements from a volume of interest in a physical object or system or a virtual model thereof;
- identifying a plurality of beamlets, each of said beamlets depositing a dose contribution of radiation to at least one subset of said volume elements;
- transforming the dose contributions of all beamlets for each of said volume elements into a first domain, yielding a set of transformed dose contributions for each of said volume elements;
- assigning a weight to each of said beamlets to create a set of weights;
- transforming the set of weights into a second domain orthogonal to the first domain, yielding a set of transformed weights;
- and
- determining a dose for each of said volume elements through summing up the elementwise products of the set of transformed dose contributions and the set of transformed weights.
A method for determining a gradient of a cost function of dose, comprising:
- creating a set of volume elements from a volume of interest in a physical object or system, or a virtual model thereof;
- identifying a plurality of beamlets, each of said beamlets depositing a dose contribution of radiation to at least one subset of said volume elements;
- transforming the dose contributions of all beamlets for each of said volume elements into a first domain, yielding a set of transformed dose contributions from each of said beamlets;
- associating a weight with each of said beamlets;
- identifying a dose for each of said volume elements;
- identifying a cost function of said determined dose;
- computing the gradient of the cost function with respect to the dose in each volume element;
- determining the transform of the gradient of the cost function with respect to each of said weights in the first domain through summing up the elementwise products of the set of transformed dose contributions and the cost function gradient with respect to the dose in each volume element;
- and
- applying a transform from the first domain yielding the gradient of the cost function with respect to each of said weights.
A non-transitory computer readable medium having stored thereon instructions that when executed cause one or more processors to perform the steps of:
- creating a set of volume elements from a volume of interest in a physical object or system, or a virtual model thereof;
- identifying a plurality of beamlets, each of said beamlets depositing a dose contribution of radiation to at least one subset of said volume elements;
- transforming the dose contributions of all beamlets for each of said volume elements into a first domain, yielding a set of transformed dose contributions for each of said volume elements;
- assigning a weight to each of said beamlets to create a set of weights;
- transforming the set of weights into a second domain orthogonal to the first domain, yielding a set of transformed weights;
- and
- determining a dose for each of said volume elements through summing up the elementwise products of the set of transformed dose contributions and the set of transformed weights.
A non-transitory computer readable medium having stored thereon instructions that when executed cause one or more processors to perform the steps of:
- creating a set of volume elements from a volume of interest in a physical object or system, or a virtual model thereof;
- identifying a plurality of beamlets, each of said beamlets depositing a dose contribution of radiation to at least a subset of said volume elements;
- transforming the dose contributions of all beamlets for each of said volume elements into a first domain, yielding a set of transformed dose contributions for each of said beamlets;
- associating a weight with each of said beamlets;
- identifying a dose for each of said volume elements;
- identifying a cost function of said determined dose;
- computing the gradient of the cost function with respect to the dose in each volume element;
- determining the transform of the gradient of the cost function with respect to each of said weights in the first domain through summing up the elementwise products of the set of transformed dose contributions and the cost function gradient with respect to the dose in each volume element;
- and
- applying a transform from the first domain yielding the gradient of the cost function with respect to each of said weights.
A method for performing dose determination in a radiation therapy, comprising:
- creating a set of volume elements from a volume of interest in a patient, or a virtual model thereof;
- identifying a plurality of beamlets, each of said beamlets depositing a dose contribution of radiation to at least one subset of said volume elements;
- transforming the dose contributions of all beamlets for each of said volume elements into a first domain, yielding a set of transformed dose contributions for each of said volume elements;
- assigning a weight to each of said beamlets to create a set of weights;
- transforming the set of weights into a second domain orthogonal to the first domain, yielding a set of transformed weights;
- and
- determining a dose for each of said volume elements through summing up elementwise products of the set of transformed dose contributions and the set of transformed weights.
A method for determining a gradient of a cost function of dose in a radiation therapy, comprising:
- creating a set of volume elements from a volume of interest in a patient, or a virtual model thereof;
- identifying a plurality of beamlets, each of said beamlets depositing a dose contribution of radiation to at least one subset of said volume elements;
- transforming the dose contributions of all beamlets for each of said volume elements into a first domain, yielding a set of transformed dose contributions from each of said beamlets;
- assigning a weight to each of said beamlets to create a set of weights;
- transforming the set of weights into a second domain orthogonal to the first domain, yielding a set of transformed weights;
- determining a dose for each of said volume elements through summing up the elementwise products of the set of transformed dose contributions and the set of transformed weights;
- identifying a cost function of said determined dose;
- computing the gradient of the cost function with respect to the dose in each volume element;
- determining the transform of the gradient of the cost function with respect to each of said weights in the first domain through summing up the elementwise products of the set of transformed dose contributions and the cost function gradient with respect to the dose in each volume element;
- and
- applying a transform from the first domain yielding the gradient of the cost function with respect to each of said weights.
A method for performing dose determination, comprising:
- creating a set of volume elements from a volume of interest in a physical object or system or a virtual model thereof;
- identifying a plurality of elementary doses, each of said elementary doses depositing a dose contribution of radiation to at least one subset of said volume elements;
- transforming the dose contributions of all elementary doses for each of said volume elements into a first domain, yielding a set of transformed dose contributions for each of said volume elements;
- assigning a weight to each of said elementary doses to create a set of weights;
- transforming the set of weights into a second domain orthogonal to the first domain, yielding a set of transformed weights;
- and
- determining a dose for each of said volume elements through summing up the elementwise products of the set of transformed dose contributions and the set of transformed weights.
A method for determining a gradient of a cost function of dose, comprising:
- creating a set of volume elements from a volume of interest in a physical object or system, or a virtual model thereof;
- identifying a plurality of elementary doses, each of said elementary doses depositing a dose contribution of radiation to at least one subset of said volume elements;
- transforming the dose contributions of all elementary doses for each of said volume elements into a first domain, yielding a set of transformed dose contributions from each of said elementary doses;
- associating a weight with each of said elementary doses;
- identifying a dose for each of said volume elements;
- identifying a cost function of said determined dose;
- computing the gradient of the cost function with respect to the dose in each volume element;
- determining the transform of the gradient of the cost function with respect to each of said weights in the first domain through summing up the elementwise products of the set of transformed dose contributions and the cost function gradient with respect to the dose in each volume element;
- and
- applying a transform from the first domain yielding the gradient of the cost function with respect to each of said weights.
Dependencies are inferred from the claim text.