Headandnecksquamouscarcinomasaffectallethnicitiesandagegroups, accountingforsignificant mortality and therapyrelated sideeffects. Over 50,000 new cases are diagnosed each year in the UnitedStates,leadingtolarge,richrepositoriesofpatientdata.Foreachofthesecases,oncologists needtoanticipatesurvival,oncologic,andsideeffectoutcomesassociatedwithtreatmentstrategiesin ordertoselectatreatmentwhichbalancesefficacyandtoxicity.However,despitethewealthofdata available,riskpredictionalgorithmsforcancersarerudimentaryandincorporateonlyahandfulof patientcharacteristics,largelyduetoalackofcomputationalmethodologies. We propose to develop a computational methodology that supports the construction of a validated precision model of patientspecific outcomes for head and neck cancers, based on demographics,toxicity,andcompleximagingdata.Ourensemblemethodologywillberevolutionaryin that it will be the first in the field to include both imaging and nonimaging data, while taking into accountlargescalebiologicalandclinicalcorrelates.Theapproachisinnovativethroughitsleverageof big data repositories and through its unique blend of computational modeling principles from bioengineering,statistics,andcomputerscience.Thesemethodsallowustoincorporatediversedata types,handlemissingdata,andmodelmultipleoutcomes. Fromaclinicalperspective,thisintegrativeapproachisnovelinthefieldofcancertherapy.The resultingmodelandwebbasedenvironmentwillmarkasignificantadvanceinbiomedicalcomputing becauseitwillbeabletoidentify,forthefirsttime,specificsubgroupswhoareatriskfordistinct oncologic,toxicity,andsurvivalprofiles.Furthermore,the?currentclinicalpracticeparadigmforhead andnecktreatmentisstagedriven.Withvalidation,thisnovelapproachcanidentifypatientsbasednot on clinical staging, but on precision modeling using cohort data and similar sets of patients. The methodologyisapplicabletoclinicalpracticeinvolvinghumansubjectsandhasthepotentialtochange thestandardofcareandoutcomesoftreatmentinthefield.
Over50,000newcasesofheadandneckcancersarediagnosedeachyearintheUnitedStates,leading tolarge,richrepositoriesofpatientdata.Thisprojectdevelopsthecomputationalmethodologyneeded to transform these repositories into precise models which can predict patientspecific responses to cancertherapy.Theprojectwillleadtobetterinsightsintothebiological,therapeuticanddemographic factorsrelatedtocancersurvivalandtoxicity,allowingpersonalizationofcareattheindividuallevel, andlargescaleprecisionmedicinebasedonquantitativerisk.
Showing the most recent 10 out of 31 publications
