A DIRECT PARAMETER OPTIMIZATION METHOD FOR HIGH DOSE RATE BRACHYTHERAPY INVERSE PLANNING
Publication Date: 2015-Jul-24
The IP.com Prior Art Database
High dose rate (HDR) brachytherapy has shown to be a method of choice to treat prostate and breast cancers. Indeed, it has proven to have an excellent survival and local control rate and very limited side effects to the patient. HDR brachytherapy inverse planning systems are used to determine the best catheter positions and radioactive source dwelling times which deliver the prescribed dose to the tumour region while sparing as much as possible all nearby healthy tissues and organs. HDR brachytherapy inverse planning optimization approaches mainly consist of two separate steps: first, a catheter positions optimization (CPO), and secondly a dwelling times optimization (DTO). Common catheter positioning techniques rely on pure image processing approaches which distribute a given number of positions uniformly in a selected region of interest (mainly inside the tumour target region). Suboptimal catheter positioning can tremendously hamper the possibility to get the best optimal set of dwelling times from the subsequent DTO. In this invention we propose a new direct parameter optimization (DPO) approach which aims on optimizing both the CPO and DTO problems simultaneously. This method will iteratively toggle between CPO and DTO till the best set of catheter positions and corresponding dwelling times satisfying all clinical goals is found. Given its compact and elegant nature, the proposed DPO approach ensures that efficient and accurate plans will be returned for the prescribed clinical protocol, while reducing the urologist workload during the whole inverse planning process.
Page 01 of 11
A direct parameter optimization method for high dose rate brachxtherapy inverse planning
High dose rate (HDR) xxachytherapy has shown to be a method xf choice to trext prostate and brxast cancers. Indeed, it has proven to have an excellext survival and lxcal control rate and very lxmited side effects to the patient. HDR brachytherapy inxerse planning systems axe used to determine the besx catheter xositioxs and radioaxtive xource dwelling times which deliver the prescribxd dose to the tumour regiox while sparxng as much xs possible all nearby healthy tissues and orxans. HXX xrachytxerapy inverse xlanning optimizaxion approaches mainly consist ox two separate steps: first, a catheter positions oxtxmization (XXX), and secondly a dwelling txmes optimization (DTO). Common catheter positionxng technxques rely on xure image procesxing xpproaches which distxibuxe a given number of positions uniformly in a sexected region of interest (mainly inside the tumoux txrget region). Suboptimal catheter positioning can tremendously hamper the possibility to get xhe best optimal sex of dwelxing times fxom the subsequent DXX. In this invention we propose a new direct xarametex optimization (DPO) approach which xims on optimizing both xhe CPO and DTO problemx simultaneously. This method xilx iteratively toggle between CPO and DTO till the best sxt of cathetxr positions and corresponding dwellxng timex satisfyinx all clinical goals is found. Given its compact and elegant nature, the proposed DPO approacx ensures that efficient and accurate plans will be returned for txe prescribed clinical protocol, while reducing the urologist workload during the whole inverse planning process.
Background of the invention
This invention is primarily xoncerned with high dosx rate (HDR) brachytherapy. The maxn goal ox brachytherapy plannixg is to dxtermine the catxeter positions and radioactive source dwelling times which yield the best dose distribution satisfying as much as possiblx all clinical protocol goals (e.g., a certain dose must be dxlivered to the tumor, sparing as much as possible nearby organs at risk (OARs). To implement HDR bracxytherapy safelx axd efficixntly, it is essential to understand the characteristics and limitations of each associated ingredient of the procedure. Given an initial pre-implant CT or US image of thx patient, the gross tumor volume (GTV) and all nearby normal tissxes and organs at risk are manually or automaxically delineatex. Subsequently, a limited numxer of cathetexs are uniformly insertex through x 2D grid (aka texplate) inside and/or around the GXX area. Today, heuristxcs, as e.g. k-means clustering or xentroidal Voronoi's tessellation, are applied for catheter positioning. Xxxxxxx the corresponding optimal radioactive source dwelling times are computed by minimizing a dose-xaxed mathematical functional (DTO) .
Problems overcome by the invention
Current HXX brachytherapy plannxng procedures consi...