Kev faib cov kab mob raum mob hauv Sonography Siv GLCM Thiab Artificial Neural Network

Abstract: Cov kab mob raum ntev (CKD) tuaj yeem kho tau yog tias kuaj pom ntxov, tab sis raws li tus kab mob loj tuaj, kev rov zoo yuav ua tsis tau.Thaum kawg, kev kho lub raum hloov pauv xws li hloov pauv lossis lim ntshav yog tsim nyog. Ultrasound yog ib txoj kev kuaj uas kuaj mob raum mob cancer, kab mob inflammatory, kab mob nodular, mob raum kab mob, thiab lwm yam. Nws yog siv los txiav txim seb qhov degree ntawm o siv cov ntaub ntawv xws li lub raum loj thiab cov echo yam ntxwv. Cov qib ntawm kev loj hlob ntawm cov kab mob raum ntev nyob rau hauv kev sim tshuaj ntsuam xyuas tam sim no yog nyob ntawm tus nqi ntawm glomerular filtration rate. Txawm li cas los xij, kev hloov pauv ntawm qhov mob thiab kab mob tuaj yeem pom nrog ultrasound. Hauv txoj kev tshawb no, tag nrho ntawm 741 cov duab, 251 lub raum cov duab ib txwm, 328 cov duab CKD me thiab nruab nrab, thiab 162 cov duab CKD hnyav tau raug sim. Txhawm rau kuaj mob CKD hauv kev kho mob, peb ROI tau teeb tsa: lub cortex ntawm lub raum, ciam teb ntawm cortex thiab medulla, thiab medulla, uas yog cov chaw kuaj xyuas kom tau txais cov ntaub ntawv los ntawm cov duab ultrasound. Cov kev ntsuas tau muab rho tawm los ntawm txhua ROI siv GLCM algorithm, uas yog siv dav hauv kev tsom xam cov duab ultrasound. Thaum txhua qhov ntsuas tau muab rho tawm los ntawm peb thaj chaw, tag nrho ntawm 57 GLCM tsis tau raug rho tawm. Thaum kawg, tag nrho ntawm 58 qhov ntsuas tau siv los ntawm kev ntxiv cov ntaub ntawv hais txog qhov loj ntawm lub raum, uas yog qhov tseem ceeb rau kev kuaj mob raum mob. Artificial neural network (ANN) yog tsim los ntawm 58 input parameters, 10 cov txheej txheem zais, thiab 3 cov txheej txheem tso tawm (ib txwm, me me thiab nruab nrab CKD, thiab hnyav CKD). Siv tus qauv ANN, qhov kev faib tawm zaum kawg yog 95.4 feem pua, lub sijhawm xav tau kev cob qhia yog 38 zaug, thiab qhov kev faib tawm tsis raug yog 4.6 feem pua.

Cistanche deserticola kev ua tau zoo - Tonifying raum

Ntsiab lus: raum ultrasound; grey-theem kev sib koom ua ke matrix (GLCM); artificial neural network; kev faib tawm; mob raum mob (CKD)

1. Taw qhia

Cov kab mob raum ntev (CKD) nce zuj zus nrog cov neeg laus thiab cov kab mob ntev zuj zus ntxiv. Hauv ntau lub tebchaws, nws yog ib qho teeb meem kev noj qab haus huv uas ua rau muaj kev pheej hmoo siab thiab mob stroke, kab mob plawv, thiab muaj teeb meem xws li ntshav qab zib thiab kab mob, nrog rau kev nce nqi kho mob.[1,2]. Cov kab mob raum mob ntev yog siv los ua ib lo lus dav dav uas hais txog kev puas tsuaj rau lub raum rau ntau tshaj 3 lub hlis lossis kev ua haujlwm tsis tu ncua ntawm lub raum tsis hais txog qhov ua rau. Nws yog ib yam kab mob uas ua rau muaj kev pheej hmoo ntawm cov kab mob plawv thiab nrog rau lub raum tsis ua haujlwm thiab ntau yam teeb meem [3,4]. Cov lus txhais hauv qab no ntawm cov kab mob raum tau siv dav los ntawm National Kidney Foundation (NKF): "yog tias lub raum puas, xws li proteinuria, hematuria, lossis pathological abnormality, lossis glomerular filtration rate tsawg dua 60 mL / min / 1.73 m2 thiab kav ntev tshaj 3 lub hlis" [5]. Cov kab mob hauv lub raum muaj ntau yam ua rau thiab kev tshawb pom pathological, tab sis hauv kev kho mob, nws tuaj yeem lees paub tsuas yog los ntawm kev kuaj zis kom pom cov proteinuria thiab kuaj ntshav los kwv yees tus nqi glomerular pom. Hauv kev kho mob, theem ntawm cov kab mob raum mob tau muab faib ua tsib theem ntawm theem 1 txog theem 5 raws li qhov kwv yees glomerular filtration rate rau kev kuaj mob thiab kho cov neeg mob. Cov kab mob raum ntev tuaj yeem hais tau tias yog ib hom kab mob sib kis, tab sis piv nrog rau lwm cov kab mob ntev xws li ntshav siab thiab ntshav qab zib, cov pej xeem paub txog tsis siab [6]. Vim tias nws tuaj yeem kuaj tau yooj yim los ntawm kev sim yooj yim, kev kuaj mob ntxov tuaj yeem ua rau qeeb ntawm tus kab mob los ntawm kev kho thaum ntxov. Txawm li cas los xij, vim tias cov tsos mob tsis meej, tus kab mob feem ntau raug kuaj pom tom qab lub sijhawm ntev, ua rau lub raum tsis ua haujlwm lossis txawm tias tuag vim muaj teeb meem xws li kab mob plawv. Feem ntau, cov kab mob raum ntev mus ntxiv thaum nws tshwm sim. Txawm li cas los xij, kev kuaj mob ntxov tuaj yeem ua rau txo qis hauv lub raum ua haujlwm thiab rov qab los ntawm cov kab mob raum ntev [7].

Ultrasound, uas yog ib txoj hauv kev rau kev kuaj mob raum mob, tseem ceeb heev rau kev kuaj xyuas yog tias lub raum ua haujlwm poob qis vim yog lwm yam laj thawj, xws li qhov loj thiab morphology ntawm lub raum. [8].

Suab puam nyob cistanche - Tonifying raum

(Cov kev tshawb fawb soj ntsuam tau pom tias Cistanche deserticola extract tuaj yeem tiv thaiv lub raum tubular hlwb, tiv thaiv lub raum interstitial lesions, qeeb tus nqi ntawm lub raum tsis ua haujlwm, thiab tiv thaiv cov kab mob thib ob hauv cov neeg mob lub raum tsis ua haujlwm. mob raum.)

GLCM algorithm, uas tshem tawm cov ntsiab lus tshwj xeeb los ntawm kev hais txog cov yam ntxwv ntawm qhov chaw hauv cov duab ultrasound, yog siv dav hauv kev tsom xam cov duab ultrasound los ntawm kev siv qhov sib txheeb ntawm cov pixel tam sim no thiab qhov ci ci ntawm nws cov pixels nyob sib ze [9].

Kev kuaj pom cov kab mob tsis siv neeg siv cov duab xws li ultrasound duab los ntawm cov txheej txheem suav tau yog qhov tseem ceeb hauv xyoo tas los no, tshwj xeeb tshaj yog tom qab tsim cov tshuab kuaj mob hauv computer (CADs) [10,11].

Artificial neural networks (ANNs) feem ntau yog ua haujlwm los tsim cov qauv kev txheeb cais rau kev tshuaj xyuas computer. Ib qho khoom siv neural network (ANN) lossis ib qho yooj yim neural network (NN) yog cov qauv kev suav raws li kev ua haujlwm ntawm cov paj hlwb. Networks ntawm hom no muaj peev xwm kawm thiab paub cov qauv raws li cov ntaub ntawv nkag thiab siv cov qauv no los kwv yees cov ntaub ntawv yav tom ntej los ntawm kev nthuav tawm cov ntaub ntawv kawm. Cov neural tes hauj lwm no muaj cov txheej txheem me me hu ua neurons txuas nrog ib leeg hauv pawg hu ua txheej; cov neural network feem ntau muaj tsawg kawg yog ib txheej nkag, ib txheej zais, thiab ib txheej tso tawm [12].

Hauv daim ntawv no, ntau yam tsis tau txais los ntawm kev siv GLCM algorithm rau ultrasound dluab cais raws li lub raum dluab, me thiab nruab nrab CKD dluab, thiab loj CKD dluab. Cov kab mob raum mob tau muab cais los ntawm kev siv cov khoom siv neural network siv GLCM tsis thiab lub raum loj. Lub hom phiaj ntawm txoj kev tshawb no yog los faib cov kab mob raum ntev mus rau hauv peb pawg los txiav txim seb qhov kev kuaj mob ntxov ntawm cov kab mob raum puas tuaj yeem ua tau.

Superman tshuaj ntsuab cistanche-Tonifying raum

Nyem qhov no mus saib Cistanche rau raum kab mob khoom

【Ask for more】 Email: xue122522@foxmail.com /  Whats App:  0086 18599088692 /  Wechat:  18599088692

2. Cov ntaub ntawv thiab cov txheej txheem

2.1. Kawm Subject thiab Data Acquisition

Txij Lub Ib Hlis 2015 txog Lub Kaum Ob Hlis 2017, qhov kev sim tau ua tiav siv cov duab ultrasound ntawm cov neeg tuaj yeem tuaj yeem tuaj xyuas lub tsev kho mob rau kev kuaj mob ntawm R Tsev Kho Mob hauv Yangsan, Gyeong-sang-nam-do, thiab cov neeg mob kuaj mob CKD.

Table 1 qhia cov yam ntxwv kho mob ntawm cov kev kawm. Tag nrho ntawm 741 cov duab ultrasound tau siv, suav nrog 251 cov duab ib txwm, 328 cov duab CKD me thiab nruab nrab, thiab 162 cov duab CKD hnyav, cais los ntawm kws kho mob. Cov kws tshawb fawb hluav taws xob tau faib cov duab sim rau cov neeg mob CKD me thiab nruab nrab thiab cov neeg mob CKD hnyav los ntawm kev nyeem ntawv, thiab cov neeg mob raum mob cancer, lub raum hlwv, thiab cov kab mob hauv lub raum tau raug cais tawm ntawm qhov kev sim [13].

Hais txog cov khoom siv sim, siv RS80A (Samsung Medison, Seoul, Kauslim) cov khoom siv ultrasound, thiab cov duab cov xwm txheej tau txais nrog qhov nce ntawm 50, dynamic ntau ntawm 56, ncej nruab nrab ntawm 8, lub zog ntawm 95, thiab ntau zaus. Txhua qhov kev xeem tau ua raws li kev yeem koom tes thiab tau txais kev pom zoo los ntawm Pawg Saib Xyuas Kev Tshawb Fawb (IRB) ntawm Catholic University of Busan (CUPIRB-2017-023).

Table 1 qhia cov ntaub ntawv ntawm hnub nyoog, poj niam txiv neej, eGFR, ntshav siab, thiab ntshav qab zib ntawm cov kev xeem.

Table 1. Cov yam ntxwv kho mob.

2.2. Txoj kev sim

Lub cheeb tsam ntawm kev txaus siab (ROI) tau teeb tsa hauv daim duab ultrasound ntawm lub raum, thiab cov txiaj ntsig tau suav nrog MATLAB 2016a (MathWorks Inc., Natick, MA, USA). ROI raug xaiv los ntawm kev teeb tsa thaj tsam cortical, thaj tsam ciam teb ntawm cortex thiab medulla, thiab cheeb tsam medulla. Nyob rau hauv daim duab ultrasound, cov yam ntxwv ntawm lub raum ib txwm yog nyob rau hauv cov duab ntawm ib tug oval muaj li ntawm lub raum cortex, uas qhia tau hais tias qis ncha piv rau lub siab, thiab lub raum sinus, uas qhia tau hais tias siab ncha. Lub ciam teb ntawm lub cortex thiab lub raum sinus yog qhov tseeb, thiab lub suab nrov nyob hauv nruab nrab yog pom. Ntawm qhov tod tes, nyob rau hauv lub raum uas tsis tu ncua puas, lub ultrasonic ncha ntawm lub raum cortex nce vim fibrosis. Yog li ntawd, ciam teb ntawm lub raum cortex brightened thiab lub raum cortex tsis pom meej. Tsis tas li ntawd, vim qhov poob ntawm kev ua haujlwm, qhov loj me me, thiab lub raum atrophy tau pom [14]. Vim tias cov yam ntxwv no tau siv rau hauv kev kuaj mob raum mob ntev, ROI cheeb tsam tau teeb tsa rau 50 × 50 thiab teem rau 3 qhov chaw. Cov duab hauv qab no qhia txog qhov qub, nruab nrab, thiab cov duab ultrasound loj ntawm 872 × 1280 daws teeb meem siv hauv txoj kev tshawb no. Daim duab 1 qhia tau hais tias ib txwm, me me thiab nruab nrab, thiab hnyav CKD ultrasound duab.

Daim duab 1. Cov duab qub.

Daim duab 2 qhia tag nrho daim duab thaiv ntawm qhov kev sim no.

Histogram sib npaug thiab ntau qhov lim dej ua ntej tau ua tiav ntawm thawj daim duab ultrasound. Histogram sib npaug tuaj yeem txhim kho qhov zoo ntawm cov duab los ntawm kev faib cov teeb thiab tsaus nti hauv cov duab sib xws [15]. Cov lim dej ntau tuaj yeem ua rau lub suab nrov txo los ntawm kev qhia txog ciam teb ntawm daim duab [16]. Tom qab ua ntej ua tiav, cov yam ntxwv tsis tau muab rho tawm los ntawm kev siv GLCM algorithm. Tag nrho ntawm 19 qhov ntsuas tau muab rho tawm los ntawm txhua qhov ntawm peb cheeb tsam ntawm daim duab 1. Cov duab qub. Daim duab 2 qhia tag nrho daim duab thaiv ntawm qhov kev sim no. Histogram sib npaug thiab ntau qhov lim dej ua ntej tau ua tiav ntawm thawj daim duab ultrasound. Histogram sib npaug tuaj yeem txhim kho qhov zoo ntawm cov duab los ntawm kev faib cov teeb thiab tsaus nti hauv cov duab sib xws [15]. Cov lim dej ntau tuaj yeem ua rau lub suab nrov txo los ntawm kev qhia txog ciam teb ntawm daim duab [16]. Tom qab ua ntej ua tiav, cov yam ntxwv tsis tau muab rho tawm los ntawm kev siv GLCM algorithm. Tag nrho ntawm 19 qhov ntsuas tau raug muab rho tawm los ntawm txhua qhov ntawm peb thaj tsam ntawm lub cortex, ciam teb ntawm cortex thiab medulla, thiab medulla, thiab, thaum kawg, cov ntaub ntawv loj tau muab tso ua ke los txiav txim siab tag nrho ntawm 58 qhov kev nkag siab.

Ib qho khoom siv neural network nrog 10 cov txheej txheem zais tau tsim, thiab cov zis tau tsim los muab faib ua peb hom: ib txwm, me me thiab nruab nrab CKD, thiab hnyav CKD.

Daim duab 2. Daim duab thaiv tag nrho ntawm kev ua cov duab.

2.3. Grey-Level Co-Occurrence Matrix (GLCM)

GLCM [17], tus qauv piav qhia kev ntxhib los mos, yog siv los xam qhov kev txiav txim thib ob los ntawm cov duab qub thiab CKD. Xav txog cov duab kuv nrog qhov loj me ntawm M × N thiab Ng tus naj npawb ntawm cov qib grey sib txawv, qhov kev hloov pauv ntawm kev ntxhib los mos yog xam los ntawm kev siv lub suab grey spatial dependence matrix p (i, j), qhov twg cov pixels sib cais nrog qhov deb d ntawm qib i-th thiab j-th grey. Hauv kev ua haujlwm tam sim no, plaub lub kaum sab xis (0◦, 45◦, 90◦, thiab 135◦ ) nrog rau pixel nrug ntawm 1 tau txiav txim siab. Cov yam ntxwv thib ob (autocorrelation, contrast, correlation, cluster prominence, cluster ntxoov, dissimilarity, zog, entropy, homogeneity, siab tshaj plaws, sum ntawm squares variance, sum nruab nrab, sum variance, sum entropy, txawv variance, txawv entropy, Cov ntaub ntawv ntsuas kev sib raug zoo 1, cov ntaub ntawv ntsuas kev sib raug zoo 2, thiab qhov sib txawv ntawm qhov sib txawv (INV) tau suav nrog GLCM. Daim duab 3a qhia cov txiaj ntsig tau siv MATLAB kom tau txais GLCM tsis. Yog tias koj xaiv ROI cheeb tsam thiab khiav, nws tau txais kev cawmdim raws li cov ntaub ntawv Excel, raws li pom hauv daim duab 3b

Daim duab 3. Kev siv qhov program.

Table 3 qhia txog qhov tsis muaj peev xwm muab rho tawm thaum xam GLCM siv daim duab ROI. Table 1 qhia txog qhov sib txawv thiab cov ntawv sau siv los suav cov qauv kev ntxhib los mos uas yog cov yam ntxwv ntawm GLCM.

Table 2. Cov khoom sib txawv thiab cov ntawv sau siv los suav cov qauv kev ntxhib los mos.

2.4. Artificial Neural Network (ANN).

ANN yog lub hauv paus ntawm kev kawm tshuab uas tam sim no tau dhau los ua qhov tseem ceeb hauv kev tshawb fawb thiab kev loj hlob. Lub tswv yim ntawm kev kawm tshuab yog lub peev xwm ntawm lub khoos phis tawj kom nkag siab txog cov qauv ntawm cov ntaub ntawv uas siv cov qauv lej lossis cov lej lej. Lub hauv paus ntawm ANN yog tsim los ntawm ib txheej ntawm cov tswv yim, txheej txheem, thiab cov khoom tso tawm. Raws li qhov tshwm sim, los ntawm lub tswv yim tseem ceeb ntawm kev ua cov ntaub ntawv lub voj voog, ANN tom qab ntawd ua cov qauv lej nyuaj los tsim cov txiaj ntsig zoo rau cov ntaub ntawv lossis cov teeb meem ntu [19].

Daim duab 4 qhia txog daim duab thaiv ntawm neural network siv hauv qhov kev sim no. ROI raug xaiv los ntawm 3 qhov chaw rau kev tshawb nrhiav pom hauv cov duab qub.

ROI raug xaiv los ntawm cortex, ciam teb ntawm lub cortex thiab medulla, thiab medulla. Qhov luaj li cas ntawm lub raum, uas yog ib qho tseem ceeb heev nyob rau hauv kev kuaj mob ntawm lub raum kab mob, raug xaiv raws li ib tug feature parameter. Raws li qhov tshwm sim ntawm kev suav GLCM, 19 qhov ntsuas tau raug muab rho tawm los ntawm txhua cheeb tsam cortex; uas yog, ciam teb ntawm lub cortex thiab medulla thiab medulla (Table 3). Tsis tas li ntawd, los ntawm kev ntxiv qhov loj ntawm lub raum raws li qhov ntsuas, cov txheej txheem nkag tau muaj tag nrho ntawm 58 nodes. Tom qab dhau los ntawm 10 cov txheej txheem zais, cov txheej txheem tso tawm tau muab faib ua peb hom: ib txwm, me me, thiab nruab nrab CKD, thiab CKD hnyav.

Daim duab 4. Kev faib tawm raws li tus qauv ANN.

Table 3. Parameters ntawm GLCM

3. Cov txiaj ntsig

Kev siv cov duab ua ntej, GLCM algorithm, thiab cov khoom siv neural network tau ua tiav siv MATLAB R2016a. Cov cuab yeej no muab cov neeg siv-phooj ywg interface thiab muaj ntau lub zog inbuilt, yog li nws yooj yim los siv algorithms hauv nws. Windows 10 (64-ntsis) nrog 3.60 GHz Intel i9 processor thiab 64 GB ntawm RAM tau siv rau txoj kev tshawb no.

GLCM algorithm. Cov ntsiab lus ntawm cov yam ntxwv tau txais los ntawm GLCM

Daim duab 5. Kev xaiv ntawm ROI.

Daim duab 6 qhia cov txiaj ntsig ntawm kev siv histogram sib npaug rau thaj tsam ROI thaum lub sijhawm rov qab los.

Daim duab 6. Cov txiaj ntsig ntawm preprocessing: (a) histogram sib npaug; (b) ntau yam filter.

Tom qab ua ntej ntawm US dluab, kev ntxhib los mos statistical nta tau muab rho tawm siv GLCM algorithm. Cov ntsiab lus ntawm cov yam ntxwv tau txais siv GLCM txheeb ze rau plaub qhov kev taw qhia sib txawv uas tau suav. Cov lus 4–7 qhia txog GLCM cov txiaj ntsig ntawm cortex, ciam teb ntawm cortex thiab medulla, thiab medulla, feem. Table 7 qhia qhov loj ntawm lub raum uas zoo li qub thiab nrog me me thiab nruab nrab CKD thiab hnyav CKD.

Table 4. Qhov tseem ceeb ntawm GLCM nta ntawm lub cortex (txhais tau tias ± STD).

Table 5. Qhov tseem ceeb ntawm GLCM nta ntawm ciam teb ntawm cortex thiab medulla (txhais tau tias ± STD).

Table 6. Qhov tseem ceeb ntawm GLCM nta ntawm medulla (txhais tau tias ± STD).

Table 7. Qhov luaj li cas ntawm lub raum ib txwm thiab nrog me me thiab nruab nrab CKD thiab hnyav CKD (txhais tau tias ± STD).

Daim duab 7 qhia txog qhov tshwm sim ntawm ANN suav nrog 64 cov khoom siv, 10 cov txheej txheem zais, thiab 3 qhov tso tawm. Hauv daim duab 7a, cov txiaj ntsig kev cob qhia yog 95.6 feem pua, cov txiaj ntsig tau txais txiaj ntsig yog 97.3 feem pua, thiab qhov kev xeem tau yog 85.7 feem pua. Kev ua qauv ua qauv muaj qhov yuam kev ntawm 0.030511 raws li qhia hauv daim duab 7b. Siv tus qauv ANN no, qhov kev faib tawm zaum kawg yog 95.4 feem pua. Daim duab 8 qhia txog ROC nkhaus rau cov txiaj ntsig.

Daim duab 7. Cov txiaj ntsig ntawm ANN.

Daim duab 8. ROC nkhaus ntawm cov txiaj ntsig.

4. Kev sib tham

Thoob plaws ntiaj teb, tus naj npawb ntawm cov neeg mob uas muaj kab mob raum ntev tau nce ntawm tus nqi zoo heev. Cov kab mob hauv lub raum tsis zoo yog qhov tshwj xeeb tshaj yog pom nyob rau hauv kev sib txuas nrog ntshav qab zib, ntshav siab, thiab hnub nyoog laus, tab sis Kaus Lim Kauslim tau dhau los ua neeg laus, thiab tus naj npawb ntawm cov neeg mob ntshav siab thiab ntshav qab zib tau nce ntxiv vim kev ua neej nyob sab hnub poob; Tsis tas li ntawd, kwv yees li 10 feem pua ​​​​ntawm cov neeg laus muaj mob raum mob thiab muaj hnub nyoog 60 xyoo lossis laus dua. Qhov tshwm sim ntawm tus kab mob raum ntev yog nce sai heev. Qhov tshwm sim no txhais tau hais tias tus naj npawb ntawm cov neeg mob uas xav tau kev lim ntshav lossis hloov lub raum vim muaj kab mob raum ntev kuj nce ntxiv, thiab thaum tus naj npawb ntawm cov neeg mob uas lub raum tsis ua haujlwm kawg nce, qhov kev poob loj yog qhov tsis tuaj yeem tshwm sim ntawm ob qho tib si hauv zej zog. thiab lub teb chaws cov ntsiab lus nrog rau cov neeg mob thiab tsev neeg [2].

Suav tshuaj ntsuab cistanche - Tonifying raum

(Cov kev tshawb fawb soj ntsuam tau pom tias ntau yam ntawm Cistanche deserticola tuaj yeem tswj hwm thiab rov ua kom lub raum ntxiv, nrog rau lub raum kom muaj zog uas ncaj qha txhim kho kev ua haujlwm ntawm lub cev lub zog ntawm lub Hoobkas mitochondria, tsis tu ncua tsim lub zog, ua kom lub cev zoo siab, txhim kho nws. txias kam rau ua, thiab txo qhov qaug zog.)

Hauv txoj kev tshawb no, lub cortex ntawm lub raum, ciam teb ntawm cortex thiab medulla, thiab medulla tau teeb tsa raws li ROIs rau kev kuaj mob raum mob ntawm cov duab ultrasound. Cov kev ntsuas tau muab rho tawm los ntawm txhua cheeb tsam ROI siv GLCM algorithm, uas yog siv dav hauv kev tsom xam cov duab ultrasound. Cov tsis yog autocorrelation, contrast, correlation, cluster prominence, pawg ntxoov ntxoo, dissimilarity, zog, entropy, homogeneity, siab tshaj plaws probability, cov sum ntawm squares variance, sum nruab nrab, sum variance, sum entropy, txawv variance, txawv entropy, cov ntaub ntawv ntsuas ntawm Kev sib raug zoo 1, cov ntaub ntawv ntsuas kev sib raug zoo 2, thiab qhov sib txawv ntawm qhov sib txawv (INV). Thaum txhua qhov ntsuas tau muab rho tawm los ntawm peb thaj chaw, tag nrho ntawm 57 GLCM tsis tau raug rho tawm. Thaum kawg, tag nrho ntawm 58 qhov ntsuas tau siv los ntawm kev ntxiv cov ntaub ntawv hais txog qhov loj ntawm lub raum, uas yog qhov tseem ceeb rau kev kuaj mob raum mob. Tag nrho ntawm 58 input parameters tau sim los ntawm kev tsim lub ANN, uas yog ib txoj kev kawm tshuab. Input parameters tau teeb tsa rau 58, thiab cov txheej txheem zais tau teeb tsa rau 10. Vim tias 10 lossis ntau dua cov txheej zais tsis pom qhov cuam tshuam rau qhov kev faib tawm, qhov kev sim tau ua nrog 10 cov txheej txheem zais. Peb qhov txiaj ntsig tau muab cais tawm yog qhov qub, me me thiab nruab nrab CKD, thiab CKD hnyav. Nws tau xav tias qhov kev faib tawm ntawm cov txiaj ntsig tau muab faib tau nce ntxiv los ntawm kev siv tag nrho 58 qhov ntsuas qhov ntsuas. Qhov kev faib tawm qhov tseeb yog 95.4 feem pua, lub sijhawm xav tau kev cob qhia yog 38 zaug, thiab qhov kev faib tsis raug yog 4.6 feem pua. Hauv qhov kev sim no, peb hom kev mob tau muab cais tawm, uas yog, ib txwm, me me, nruab nrab CKD, thiab CKD hnyav, tab sis yuav tsum muaj kev qhia ntxaws ntxiv ntawm cov kab mob. Tsis tas li ntawd, qhov kev sim tau ua los ntawm kev tau txais 741 cov ntaub ntawv cov khoom, uas yog vim cov ntaub ntawv tsis loj; Yog li, txoj kev kawm tshuab tau raug xaiv. Thaum tau txais cov ntaub ntawv ntau, peb npaj yuav siv txoj kev kawm tob tom qab tau txais cov ntaub ntawv ntau ntxiv. Txhawm rau siv nws hauv kev kho mob, nws yog ib qho tsim nyog yuav tsum tau ua kom muaj ntau hom kev faib tawm. Tom qab ua kom tiav cov txiaj ntsig muaj nyob rau hauv cov khoom siv ultrasonic, nws yuav tsum tau kho dua tshiab los ntawm kev tawm tswv yim los ntawm tus sonographer.

5. Cov lus xaus

Mob raum mob tuaj yeem kho tau yog tias kuaj pom ntxov,Cistanche deserticola yog ib txoj hauv kev zoo rau tonify lub raum. Cistanche deserticola tuaj yeem ua ke nrog goji berries thiab Cynomorium solarium rau brewing thiab haus, uas muaj txiaj ntsig zoo rau tonifying lub raum. Cistanche deserticola ma yog tsuas yog siv los kho qhov mob thiab tsis muaj zog ntawm lub duav thiab lub hauv caug los ntawm lub raum yang deficiency, uas muaj cov nyhuv tonifying lub raum thiab muaj qee yam cuam tshuam rau cov kab mob raum thaum ntxov.

Cistanche deserticola kev ua tau zoo-tonifying raum

Tab sis raws li tus kab mob no zuj zus, kev rov qab los ua tsis tau. Thaum kawg, kev kho lub raum hloov pauv xws li hloov pauv lossis lim ntshav yuav tsum tau siv. Hauv lwm lo lus, nws yog ib qho tseem ceeb los kuaj xyuas thiab kho cov kab mob raum ntev nyob rau theem pib. Ultrasound yog ib qho kev sim rau kev kuaj mob raum mob cancer, kab mob inflammatory, kab mob nodular, kab mob raum, thiab lwm yam, thiab yog siv los xyuas cov ntaub ntawv ntawm cov qib ntawm o uas siv cov ntaub ntawv xws li lub raum loj thiab echo yam ntxwv.

Hauv txoj kev tshawb no, cov duab ultrasound, suav nrog 251 lub raum cov duab ib txwm, 328 cov duab mob raum mob me thiab nruab nrab, thiab 162 lub raum raum mob hnyav, tau siv hauv 741 tus neeg mob. Txhawm rau kuaj mob raum tsis zoo hauv kev kho mob, peb ROI tau teeb tsa, uas yog, lub raum ntawm lub raum, ciam teb ntawm cortex thiab medulla, thiab medulla, uas yog cov chaw kuaj xyuas kom tau txais cov ntaub ntawv los ntawm cov duab ultrasound. Cov kev ntsuas tau muab rho tawm los ntawm txhua ROI siv GLCM algorithm, uas yog siv dav hauv kev tsom xam cov duab ultrasound. Thaum txhua qhov ntsuas tau muab rho tawm los ntawm peb thaj chaw, tag nrho ntawm 57 GLCM tsis tau raug rho tawm. Thaum kawg, tag nrho ntawm 58 qhov ntsuas tau siv los ntawm kev ntxiv cov ntaub ntawv hais txog qhov loj ntawm lub raum, uas yog qhov tseem ceeb rau kev kuaj mob raum mob. ANN tau tsim los ntawm 58 qhov kev nkag tsis tau, 10 cov txheej txheem zais, thiab 3 cov txheej txheem tso tawm (ib txwm, me me thiab nruab nrab CKD, thiab CKD hnyav). Siv tus qauv ANN, qhov kev faib tawm zaum kawg yog 95.4 feem pua, lub sijhawm xav tau kev cob qhia yog 38 zaug, thiab qhov kev faib tawm tsis raug yog 4.6 feem pua. Nws ntseeg tau tias qhov kev sim no tuaj yeem siv los ua lub hauv paus rau kev siv qhov kev kuaj mob tsis siv neeg hauv thaj tsam ntawm kev kuaj mob raum mob siv cov duab ultrasound. Tsis tas li ntawd, kev siv cov txiaj ntsig kev sim tau xav tias yuav ua lub luag haujlwm tseem ceeb hauv kev txiav txim siab kho mob, suav nrog kev kuaj mob ntxov thiab kev kho mob raum mob ntev.

Cov ntaub ntawv

1. Irina, L.; Flavio, R. Idalina, IB; Rau, A.; Luis, B.; Alice, SS New Potential Biomarkers for Chronic Kidney Disease Management—A Review of the Literature. Int. J. Mol. Sci. 2021, 22, 43 ib.

2. Couser, WG; Remuzzi, G.; Mendis, S.; Tonelli, M. Kev koom tes ntawm cov kab mob raum ntev mus rau lub ntiaj teb lub nra ntawm cov kab mob loj. Raum Int. 2011, 80, 1258–1270. [CrossRef]

3. Cañadas-Garre, M.; Anderson, K.; McGoldrick, J.; Maxwell, AP; McKnight, AJ Proteomic thiab metabolomic mus kom ze hauv kev tshawb nrhiav biomarkers hauv kab mob raum. J. Proteom. 2019, 193, 93–122. [CrossRef]

4. Lovey, AS; Eckardt, KU; Tsukamoto, Y.; Levin, UA; Coresh, J.; Rossert, J.; de Zeeuw, D.; Hostetter, TH; Lameire, N.; Eknoyan, G. Kev txhais thiab kev faib tawm ntawm cov kab mob hauv lub raum: Ib txoj hauj lwm nqe lus los ntawm Kab Mob Raum: Txhim Kho Ntiaj Teb Cov txiaj ntsig (KDIGO). Raum Int. 2005, 67, 2089–2100. [CrossRef]

5. Komenda, P.; Rigatto, C.; Tangri, N. Kev tshuaj xyuas cov tswv yim rau cov tsis paub CKD. Clin. J. Am. Soc. Nephrol. 2016, 11, 925–927. [CrossRef] [PubMed]

6. Lee, HS; Jang, HB; Yog, MG; Park, SI; Lee, HJ Amino Acid Metabolite Koom Tes Nrog Kab Mob Raum Hniav: Ib Yim Xyoo Ua Raws Li Kev Kawm Kaus Lim Kab Mob. Biomedicines 2020, 8, 222. [CrossRef]

7. Hewadikaram, D.; Bandara, M.; Pattividana, AN; Jayaweera, H. Ib qho tshiab ultrasound txheej txheem los kuaj xyuas kab mob raum thaum ntxov. F1000Research 2019, 7, 448. [CrossRef] [PubMed]

8. Jovanovic, D.; Gasic, IB; Pavlovic, S.; Naumovic, R. Kev sib txheeb ntawm lub raum loj nrog rau lub raum ua haujlwm thiab anthropometric tsis nyob rau hauv cov kev noj qab haus huv thiab cov neeg mob uas muaj kab mob raum. Ren. Ua tsis tiav. 2013, 35, 896–900. [CrossRef]

9. Gao, S.; Peng, Y.; Guo, H.; Li, W.; Gao, T.; Xu, Y.; Tang, X. Kev tsom xam kev ntxhib los mos thiab kev faib tawm ntawm ultrasound daim siab daim siab. Biome. Mater. Eng. 2014, 24, 1209–1216. [CrossRef] [PubMed]

10. Saleem, ZR Txoj Kev Siv Hauv Computer-Aided Diagnosis for Breast Cancer Detection using Mammograms: A Review. J. Healthc. Eng. Xyoo 2020, 2020, 9162464.

11. Fujita, H.; koj, J.; Li, Q. Arimura, H. State-of-the-Art of Computer-Aided Detection/Diagnosis (CAD). Hauv Kev Ua Haujlwm ntawm ICMB 2010, Hong Kong, Tuam Tshoj, 28–30 Lub Rau Hli 2010; Phau 6165, p. 296–305.

12. Yaj, JSR; Sun, CT; Mizutani, E. Neuro-Fuzzy thiab Soft Computing—A Computational Approach to Learning and Machine Intelligence. IEEE Trans. Autom. Tswj 1997, 42, 1482–1484. [CrossRef]

13. Abbiss, H.; Tus tsim, GL; Trengove, RD Metabolomics txoj hauv kev rau kev kuaj mob thiab kev nkag siab ntawm cov kab mob raum. Metabolites 2019, 9, 34. [CrossRef]

14. Hansen, KL; Nielsen, MB; Ewertsen, C. Ultrasonography ntawm lub raum: Kev Ntsuam Xyuas Pictorial. Diagnostics 2016, 6, 2. [CrossRef] [PubMed]

15. Singh, P.; Mukundan, R.; De Ryke, R. Feature Enhancement in Medical Ultrasound Videos Siv Contrast-Limited Adaptive Histogram Equalization. J. Tus lej. Cov duab 2020, 33, 273–285. [CrossRef] [PubMed]

16. Jassim, FA Image Denoising Siv Interquartile Range Filter nrog rau qhov nruab nrab hauv zos. Int. J. Soft Computer. Eng. 2013, 2, 424–428.

17. Haralick, RM; Dinstein, ib.; Shanmugamm, K. Textural nta rau kev faib cov duab. IEEE Trans. Syst. Txiv neej Cybern. 1973, 3, 610–621. [CrossRef]

18. Tommy, L.; Patrik, IB; Thomas, A.; Taub, N.; Anders, G. Grey-theem invariant Haralick kev ntxhib los mos nta. PLoS IB 2019, 14, e0212110.

19. Farizawani, AG; Puth, M.; Marina, Y.; Rivaie, A. Kev tshuaj xyuas ntawm cov txheej txheem kev kawm txuj ci neural network tsim los ntawm ntau qhov sib txawv ntawm kev sib txuas ntawm gradient mus kom ze. J. Phys. Conf. Ser. 2020, 1529, 022040. [CrossRef]

20. Igbinedion, BO; Okaka, E. Mob raum Kab Mob: Sonographic/Clinical Findings at University of Benin Teaching Hospital. Ann. Biomed. Sci. 2017, 16, 60–69.

21. Prashanth, KS; Rau, N.; Chandrashekar, HM Sonographic Grading of Renal Parenchymal Changes and its Comparision with Estimated Glomerular Filtration Rate (EGFR) Siv Cov Khoom Noj Modifified hauv Cov Kab Mob Raum. Int. J. Kev. Med. Surg. Radiol. 2020, 5, B13–B16.

22. Priyanka; Kumar, D. Feature Extraction thiab xaiv lub raum Ultrasound Duab Siv GLCM thiab PCA. Procedia Comput. Sci. 2020, 167, 1722–1731. [CrossRef]

23. Kuo, CC; Chang, CM; Liu, KT; Lin, WK; Chiang, HY; Chung, CW; Ho, MR; Sun, PR; Yang, RL; Chen, KT Automation ntawm lub raum kev ua haujlwm kwv yees thiab kev faib tawm los ntawm ultrasound-raws li lub raum duab siv kev kawm tob. NPJ Digit. Med. 2019, 2, 29. [CrossRef] [PubMed]