Question
What is the population of the study? All I was given was this:
MetroP oo M 11 20% 8:25 PM 1-s2.0-S1 566... A : ELSEVIER Full Length Article Fusing texture, shape and deep model-learned information at decision level for automated classification of lung nodules on chest CT Xie Yutong, Zhang Jianpeng.b. Xia Yongb, Michael Fulham4* Zhang Yanning ARTICLEINFO ABSTRACT nils aganthm empuws level co-occurrence matrix (GLCM-based texture descripooc, Fowler shape dsciptor to characeriehe heeogehy t nodules and a deep couroluional neural nerwock (DCNN) o automatically lean! tbe teaDurerepesentation of nodules ca slice-by. sike bols. Ir alus an AdaBoosted back propazation neural neo BPSN Ugcch eature type ad tuses the decisioas madeb three caliers to ditcrentdate nodules. Weevaluned this igouh aiainthree approaches oa the LIDCIDRI danse. When the nodules witha mposite margnancy Lane 3 were discu ded, regarded as be Ign 0rLegauded as mallgrant, our Computer aided diagrosis (CAD). however, avuids many of these issues and is incrasingly being investigated as an alterve and ibe 2015 gobal cancer statistics showed that there are appruxi clementary appruath to conventional reading [11. Many automated mately 11.1 million new cancer cases each year. Lang cancer has anlung nodule classiicatico approaches have been propoeed in the lit incidence of 13% and death rate of 19.5%, which are the highest rates erature and must of tem consist of mage preprocessing, nodule de acr all cancers []. Early diagnosis and tresatment are the most ef ctn, nodule segmentation, feature extraction and dsiicatiun lective as to impove survval al lung cer patienta, the 5-yeaArhem, leature exaerionis a eritical srep. The fatures uad lor survival lorhase with nearly dingnosis is apprnxietely 54 'wgnoduleelassificationcabe dividk inta hano-raed canures and mmlar rd 'n> 4% il rhentingnasis is Inte when the poliat has *tagery leannes lesrnd b deep 'Kurl networks (nNNs) Hand-efafreí lea disease 21. A n,, an rhe ling, deteeted b eltest 4onpiired tomo tures iidide texture, and Ahape dese tors, sinoe rliere is sa hi ! enr ling nodile and may he henign áf malignant. The Natinal ranne val"es and shape 151· Orke hard-erared tcatu'ns are extraesed, a Screening Trial * nawod rliat serorning with CT will 'esult ill a 20% variety or elassinention terhniques car' he itiod including the sugyar' reduction ng caner deatl, hy detgcarly discaa13.Rwlor machineM67, disione 8,91, K-neast neighbor ologists globally typocally visually analyze chest CI scans a liceby NN) 10], back propagation eura network (BPNN) L.12].random rest (RF) [I3]ad Adabecst [11,15] slice basis which is time consuming expensive and prone to reade bias and requires high degree of sill and coocentration. he most commonly used asual texture descriptor is based on th
Explanation / Answer
This study is based on cancer patients and particularly lung cancer, the population of the study is lung cancer patients. The diagnostic tests which are done are also done on cancer patients, so the population seems to be lung cancer patients.
