Within Sites Adjustable Neuroimaging Report

Curated Report for pCASL Chiller Issues at Kansas:

• First I extract all MR1 and MR2 scans which have "chiller" included in their run sheet notes, and make that a binary grouping variable.

## stringr package to parse subjects with "chiller" in scan sheet notes.
MASTER_ANOVA$pCASL.issues<-str_detect(MASTER_ANOVA$ScanNotes, pattern = "chiller")
  
  ## Make that a factor
  MASTER_ANOVA$pCASL.issues<-if_else(MASTER_ANOVA$pCASL.issues==TRUE, "1", "0")
  MASTER_ANOVA$pCASL.issues<-as.factor(MASTER_ANOVA$pCASL.issues)

  ## Create dataframe to present in table
  Chiller.df<-MASTER_ANOVA %>% filter(pCASL.issues==1) %>% select(SubID,Session, Site, Date, ScanNotes,QC3_Notes, QC4_Notes)
  
  ## Clean Up for analyses
  MASTER_ANOVA<-MASTER_ANOVA[,-c(1:2,45:47)]
  
  datatable(Chiller.df, rownames= FALSE,options = list( pageLength = 5 ))

• Next, regressions are run on pCASL outmean and tSNR with 'pCASL chiller issues' (1/0) as the independent variable. These analyses are hashed out below for clear output formatting. Note that dates were not included for the first set of regressions. I wanted to compare scans with pCASL issues versus all other scans completed at Kansas.
  RESULTS:
  • pCASL outmean is significantly impacted by the grouping variable for chiller issues (greater outmean for chiller issue scans).
    • Upon removing outlier #20145_MR2 from the sample, the 'pCASL.issues' grouping variable is NON-significant, suggesting this scan includes a large portion of variance for outmean. This makes sense, as their scan notes state that pCASL was incomplete.
      • Since pCASL outmean depends on the number of volumes collected, fewer volumes==larger outmean. Thus, we could expect to see larger than average outmean parameters for any particpants with incomplete pCASL collection.
  • pCASL tsnr is not directly impacted by pCASL chiller issues, but 'pCASL.issues' moderate the relationship between pCASL tSNR and outmean when controlling for session (mean tSNR is significantly greater for non-affected scans).
    
  • Total SNR is significantly predicted by pCASL issues for Resting State, Rise Encoding, and Rise Recognition. However, upon removing #20390_MR1 & #20145_MR2 from the sample, these effects are non-significant (yet still trending for RISE2 SNR).

## Only Kansas 
   MASTER_KU<-MASTER_ANOVA %>% filter(Site=="Kansas")

## Outmean
  # summary(lm(pCASL_outmean  ~pCASL.issues,data=MASTER_KU))
  # summary(lm(pCASL_outmean  ~Date,data=MASTER_KU))
  # summary(lm(pCASL_outmean  ~pCASL.issues+Date,data=MASTER_KU)) # Date NS
  # summary(lm(pCASL_outmean  ~pCASL.issues*Date,data=MASTER_KU))  #Interaction, so I'll remove outlier and re-run

## tSNR 
  # summary(lm(pCASL_tsnr  ~Session+pCASL.issues,data=MASTER_KU)) # Non significant
  # summary(lm(pCASL_tsnr  ~pCASL_outmean:pCASL.issues+Session,data=MASTER_KU)) 

## Remove outmean outlier and re-run regression
    #MASTER_KU<-MASTER_KU[!(MASTER_KU$SubID==c("20145", "20390" )),] 
    #summary(lm(pCASL_outmean ~pCASL.issues,data=MASTER_KU))  #Non significant this time. Thus, #20145 & #20390 are what's pulling sample variance. Below, I get into more detail
  #summary(lm(pCASL_tsnr ~pCASL_outmean:pCASL.issues+Session,data=MASTER_KU)) # Still non-significant

 #summary(lm(Rest_snr ~pCASL.issues,data=MASTER_KU))  
 #summary(lm(RISE_enc_snr ~pCASL.issues,data=MASTER_KU))  
 #summary(lm(RISE_rec_snr  ~pCASL.issues,data=MASTER_KU))  

Site reports with adjustable date ranges:

• Up until this point, the importing process has been automated. Now, you'll manually establish your report configuration.

1. Set-up adjustable report parameters:

Adjustable Parameters- Use the chunk below to configure your report.
* StartDate: The date you want your report range to start.
* EndDate: The date you want your report range to start.
* ReportSite: "UPitt", "Kansas", "Northeastern"
* Outliers: (Optional) Include a list of outliers to remove from your report.

##### Enter info you want here ######
StartDate=as.Date("2019-10-30")
EndDate=as.Date("2019-11-30")
ReportSite="Kansas"
#Outliers = c("20145MR2")  # Optional If only running outlier report

2. Report creates data structure:

• Edit the chunk below if running report for a new factor or variable.
• Regression samples will be restricted to reporting date range, but main effects can alternatively be modeled with a different independent variable.

##### Sets-up anova data structure ####
Report<-MASTER_ANOVA %>% filter(Date<=EndDate)  
Report<- Report %>% filter(Site==ReportSite)
Report$Date<-as.Date(Report$Date, format="%Y-%m-%d")
Report$Session<-as.factor(Report$Session)

# Optionally unhash to suit your analyses
#Report<-Report[!(Report$subject_id==Outliers),] 
Report$Specs<-if_else(Report$Date <= StartDate, "Prior","Report" )
Report$Specs<-as.factor(Report$Specs)


#Report$Specs<-if_else(Report$pCASL.issues=="1" , "2","1" )
#Report$Specs<-as.factor(Report$Specs)

3. Run Automated Reports -

• This will only generate mean comparisons (Report vs. Prior) for ANOVA's in which IQM * ReportSpecs is sigificant.

##### Run ANOVAS for significance, and use emmeans to calculate group differences #####
###### By Scanning Aquistion
MPRAGE.Report<-Report %>%
  select(SubID, Session, Date, Specs, starts_with("MPRAGE"))
mlt.MASTER.Report<-melt(MPRAGE.Report, id.vars =c("SubID","Session", "Date" , "Specs"))
colnames(mlt.MASTER.Report)<-c("SubID",  "Session", "Date","Specs", "IQMs","value" )
mlt.MASTER.Report$Specs<-as.factor(mlt.MASTER.Report$Specs)
mlt.MASTER.Report$IQMs<-as.factor(mlt.MASTER.Report$IQMs)
mlt.MASTER.Report$Session<-as.factor(mlt.MASTER.Report$Session)
lm0 <- aov(value ~ Specs *IQMs, data = mlt.MASTER.Report)
emeans <- emmeans(lm0, "Specs", by="IQMs")
model.lm0<-summary(lm0)
sum_test = unlist(model.lm0)
Interaction.p<-sum_test["Pr(>F)3"]
if (Interaction.p<0.05) {
  print( "MPRAGE IQM * Report Range Significant! Mean Comparisions...")
  pairs(emeans)
} else {print( "NS")}

## [1] "NS"

NBACK.Report<-Report %>% select(SubID, Session, Date, Specs, starts_with("Nback"))
mlt.MASTER.Report<-melt(NBACK.Report, id.vars =c("SubID","Session", "Date" , "Specs"))
colnames(mlt.MASTER.Report)<-c("SubID",  "Session", "Date","Specs", "IQMs","value" )
mlt.MASTER.Report$Specs<-as.factor(mlt.MASTER.Report$Specs)
mlt.MASTER.Report$IQMs<-as.factor(mlt.MASTER.Report$IQMs)
mlt.MASTER.Report$Session<-as.factor(mlt.MASTER.Report$Session)
lm0 <- aov(value ~ Specs *IQMs, data = mlt.MASTER.Report)
emeans <- emmeans(lm0, "Specs", by="IQMs")
model.lm0<-summary(lm0)
sum_test = unlist(model.lm0)
Interaction.p<-sum_test["Pr(>F)3"]
if (Interaction.p<0.05) {
  print( "N-Back IQM * Report Range Significant! Mean Comparisions...")
  pairs(emeans)
} else {print(  "NS")}

## [1] "NS"

REST.Report<-Report %>%
  select(SubID, Session,Date, Specs, starts_with("Rest_"))
mlt.MASTER.Report<-melt(REST.Report, id.vars =c("SubID","Session", "Date" , "Specs"))
colnames(mlt.MASTER.Report)<-c("SubID",  "Session", "Date","Specs", "IQMs","value" )
mlt.MASTER.Report$Specs<-as.factor(mlt.MASTER.Report$Specs)
mlt.MASTER.Report$IQMs<-as.factor(mlt.MASTER.Report$IQMs)
mlt.MASTER.Report$Session<-as.factor(mlt.MASTER.Report$Session)
lm0<- aov(value ~ Specs *IQMs, data = mlt.MASTER.Report)
emeans <- emmeans(lm0, "Specs", by="IQMs")
model.lm0<-summary(lm0)
sum_test = unlist(model.lm0)
Interaction.p<-sum_test["Pr(>F)3"]

if (Interaction.p<0.05) {
  print( "Resting State IQM * Report Range Significant! Mean Comparisions...")
  pairs(emeans)
} else {print( "NS")}

## [1] "Resting State IQM * Report Range Significant! Mean Comparisions..."

## IQMs = Rest_dvars_nstd:
##  contrast        estimate   SE   df t.ratio p.value
##  Prior - Report -5.797020 1.34 1206 -4.328  <.0001 
## 
## IQMs = Rest_fd_mean:
##  contrast        estimate   SE   df t.ratio p.value
##  Prior - Report -0.074084 1.34 1206 -0.055  0.9559 
## 
## IQMs = Rest_fwhm_avg:
##  contrast        estimate   SE   df t.ratio p.value
##  Prior - Report -0.008719 1.34 1206 -0.007  0.9948 
## 
## IQMs = Rest_gcor:
##  contrast        estimate   SE   df t.ratio p.value
##  Prior - Report -0.000463 1.34 1206  0.000  0.9997 
## 
## IQMs = Rest_snr:
##  contrast        estimate   SE   df t.ratio p.value
##  Prior - Report -0.031495 1.34 1206 -0.024  0.9812 
## 
## IQMs = Rest_tsnr:
##  contrast        estimate   SE   df t.ratio p.value
##  Prior - Report  1.274516 1.34 1206  0.951  0.3416

RISE1.Report<-Report %>%
  select(SubID, Session, Date, Specs, starts_with("RISE_enc"))
mlt.MASTER.Report<-melt(RISE1.Report, id.vars =c("SubID","Session", "Date" , "Specs"))
colnames(mlt.MASTER.Report)<-c("SubID",  "Session", "Date","Specs", "IQMs","value" )
mlt.MASTER.Report$Specs<-as.factor(mlt.MASTER.Report$Specs)
mlt.MASTER.Report$IQMs<-as.factor(mlt.MASTER.Report$IQMs)
mlt.MASTER.Report$Session<-as.factor(mlt.MASTER.Report$Session)
lm0 <- aov(value ~ Specs *IQMs, data = mlt.MASTER.Report)
emeans <- emmeans(lm0, "Specs", by="IQMs")
model.lm0<-summary(lm0)
sum_test = unlist(model.lm0)
Interaction.p<-sum_test["Pr(>F)3"]

if (Interaction.p<0.05) {
  print( "RISE1 IQM * Report Range Significant! Mean Comparisions...")
  pairs(emeans)
} else {print( "NS")}

## [1] "RISE1 IQM * Report Range Significant! Mean Comparisions..."

## IQMs = RISE_enc_dvars_nstd:
##  contrast        estimate   SE   df t.ratio p.value
##  Prior - Report -6.440955 1.79 1212 -3.598  0.0003 
## 
## IQMs = RISE_enc_fd_mean:
##  contrast        estimate   SE   df t.ratio p.value
##  Prior - Report -0.105586 1.79 1212 -0.059  0.9530 
## 
## IQMs = RISE_enc_fwhm_avg:
##  contrast        estimate   SE   df t.ratio p.value
##  Prior - Report -0.008269 1.79 1212 -0.005  0.9963 
## 
## IQMs = RISE_enc_gcor:
##  contrast        estimate   SE   df t.ratio p.value
##  Prior - Report  0.000086 1.79 1212  0.000  1.0000 
## 
## IQMs = RISE_enc_snr:
##  contrast        estimate   SE   df t.ratio p.value
##  Prior - Report -0.032799 1.79 1212 -0.018  0.9854 
## 
## IQMs = RISE_enc_tsnr:
##  contrast        estimate   SE   df t.ratio p.value
##  Prior - Report  1.442871 1.79 1212  0.806  0.4204

RISE2.Report<-Report %>%
  select(SubID, Session, Date, Specs, starts_with("RISE_rec"))
mlt.MASTER.Report<-melt(RISE2.Report, id.vars =c("SubID","Session", "Date" , "Specs"))
colnames(mlt.MASTER.Report)<-c("SubID",  "Session", "Date","Specs", "IQMs","value" )
mlt.MASTER.Report$Specs<-as.factor(mlt.MASTER.Report$Specs)
mlt.MASTER.Report$IQMs<-as.factor(mlt.MASTER.Report$IQMs)
mlt.MASTER.Report$Session<-as.factor(mlt.MASTER.Report$Session)
lm0<- aov(value ~ Specs *IQMs, data = mlt.MASTER.Report)
emeans<- emmeans(lm0, "Specs", by="IQMs")
model.lm0<-summary(lm0)
sum_test = unlist(model.lm0)
Interaction.p<-sum_test["Pr(>F)3"]

if (Interaction.p<0.05) {
  print( "RISE2 IQM * Report Range Significant! Mean Comparisions...")
  pairs(emeans)
} else {print( "NS")}

## [1] "NS"

DTI.Report<-Report %>%
  select(SubID,Session,  Date, Specs, starts_with("DTI"))
mlt.MASTER.Report<-melt(DTI.Report, id.vars =c("SubID","Session", "Date" , "Specs"))
colnames(mlt.MASTER.Report)<-c("SubID",  "Session", "Date","Specs", "IQMs","value" )
mlt.MASTER.Report$Specs<-as.factor(mlt.MASTER.Report$Specs)
mlt.MASTER.Report$IQMs<-as.factor(mlt.MASTER.Report$IQMs)
mlt.MASTER.Report$Session<-as.factor(mlt.MASTER.Report$Session)
lm0<- aov(value ~ Specs *IQMs, data = mlt.MASTER.Report)
emeans<- emmeans(lm0, "Specs", by="IQMs")
model.lm0<-summary(lm0)
sum_test = unlist(model.lm0)
Interaction.p<-sum_test["Pr(>F)3"]

if (Interaction.p<0.05) {
  pairs(emeans)
} else {print( "NS")}

## [1] "NS"

pCASL.Report<-Report %>%
  select(SubID,Session,  Date, Specs, starts_with("pCASL_"))
mlt.MASTER.Report<-melt(pCASL.Report, id.vars =c("SubID","Session", "Date" , "Specs"))
colnames(mlt.MASTER.Report)<-c("SubID",  "Session", "Date","Specs", "IQMs","value" )
mlt.MASTER.Report$Specs<-as.factor(mlt.MASTER.Report$Specs)
mlt.MASTER.Report$IQMs<-as.factor(mlt.MASTER.Report$IQMs)
mlt.MASTER.Report$Session<-as.factor(mlt.MASTER.Report$Session)
lm0 <- aov(value ~ Specs *IQMs, data = mlt.MASTER.Report)
emeans<- emmeans(lm0, "Specs", by="IQMs")
model.lm0<-summary(lm0)
sum_test = unlist(model.lm0)
Interaction.p<-sum_test["Pr(>F)3"]

if (Interaction.p<0.05) {
  print( "pCASL IQM * Report Range Significant! Mean Comparisions...")
  pairs(emeans)
} else {print( "NS")}

## [1] "NS"

Flair.Report<-Report %>%
  select(SubID,Session,  Date, Specs, starts_with("FLAIR_snr"))
lm0<- aov(FLAIR_snr ~ Specs , data = Flair.Report)
emeans<- emmeans(lm0, "Specs")
model.lm0<-summary(lm0)
sum_test = unlist(model.lm0)
Interaction.p<-sum_test["Pr(>F)1"]

if (Interaction.p<0.05) {
  print( "Flair IQM * Report Range Significant! Mean Comparisions...")
  pairs(emeans)
} else {print( "NS")}

## [1] "NS"

Kansas Chiller Issues Report:

• Report Range: 2019-10-30 - 2019-11-30
  • Here I compare the significance of reporting range (grouping variable) and pCASL.issues (grouping variable), to assess whether pCASL issues significantly contribute to the variance across other sequence parameters. As you'll see, pCASL.issues significantly predicts SNR across several sequences.

#summary(lm(pCASL_outmean~ Specs,data=Report))
#summary(lm(pCASL_outmean~ pCASL.issues,data=Report))
#summary(lm(pCASL_outmean~ Session,data=Report))

#summary(lm(pCASL_tsnr~ Specs,data=Report))
#summary(lm(pCASL_tsnr~ pCASL.issues,data=Report))
#summary(lm(pCASL_tsnr~ Session,data=Report))

#summary(lm(MPRAGE_snr_total~ pCASL.issues,data=Report)) 
#summary(lm(MPRAGE_snr_total~ Specs,data=Report)) 

summary(lm(Rest_snr~ pCASL.issues,data=Report))   #SIGNIFICANT

## 
## Call:
## lm(formula = Rest_snr ~ pCASL.issues, data = Report)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -0.78196 -0.18003  0.00224  0.15893  0.80804 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)    2.25078    0.02029  110.95   <2e-16 ***
## pCASL.issues1  0.33546    0.16687    2.01   0.0457 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.2869 on 201 degrees of freedom
##   (1 observation deleted due to missingness)
## Multiple R-squared:  0.01971,    Adjusted R-squared:  0.01483 
## F-statistic: 4.041 on 1 and 201 DF,  p-value: 0.04574

summary(lm(Rest_snr~ Specs,data=Report))  #NON-Significant

## 
## Call:
## lm(formula = Rest_snr ~ Specs, data = Report)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -0.78413 -0.18648  0.00546  0.15996  0.80588 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  2.25295    0.02129  105.81   <2e-16 ***
## SpecsReport  0.03149    0.07151    0.44     0.66    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.2896 on 201 degrees of freedom
##   (1 observation deleted due to missingness)
## Multiple R-squared:  0.0009642,  Adjusted R-squared:  -0.004006 
## F-statistic: 0.194 on 1 and 201 DF,  p-value: 0.6601

summary(lm(RISE_enc_snr~pCASL.issues,data=Report)) #SIGNIFICANT

## 
## Call:
## lm(formula = RISE_enc_snr ~ pCASL.issues, data = Report)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -0.77688 -0.17646 -0.00616  0.16629  0.77262 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)     2.2467     0.0198 113.443   <2e-16 ***
## pCASL.issues1   0.3409     0.1633   2.087   0.0381 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.2808 on 202 degrees of freedom
## Multiple R-squared:  0.02111,    Adjusted R-squared:  0.01627 
## F-statistic: 4.357 on 1 and 202 DF,  p-value: 0.03811

summary(lm(RISE_enc_snr~Specs,data=Report)) #NON-Significant

## 
## Call:
## lm(formula = RISE_enc_snr ~ Specs, data = Report)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -0.77900 -0.17858 -0.00426  0.17194  0.77050 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  2.24880    0.02080 108.130   <2e-16 ***
## SpecsReport  0.03280    0.07001   0.468     0.64    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.2836 on 202 degrees of freedom
## Multiple R-squared:  0.001085,   Adjusted R-squared:  -0.00386 
## F-statistic: 0.2195 on 1 and 202 DF,  p-value: 0.64

summary(lm(RISE_rec_snr~ pCASL.issues,data=Report))  #SIGNIFICANT

## 
## Call:
## lm(formula = RISE_rec_snr ~ pCASL.issues, data = Report)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -0.76260 -0.17446  0.00512  0.16543  0.85164 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)    2.23826    0.02002 111.818   <2e-16 ***
## pCASL.issues1  0.34442    0.16506   2.087   0.0382 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.2838 on 202 degrees of freedom
## Multiple R-squared:  0.0211, Adjusted R-squared:  0.01625 
## F-statistic: 4.354 on 1 and 202 DF,  p-value: 0.03818

summary(lm(RISE_rec_snr~ Specs,data=Report)) #NON-Significant

## 
## Call:
## lm(formula = RISE_rec_snr ~ Specs, data = Report)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -0.76411 -0.18481  0.00942  0.17248  0.85013 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  2.23976    0.02101  106.58   <2e-16 ***
## SpecsReport  0.04031    0.07075    0.57    0.569    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.2866 on 202 degrees of freedom
## Multiple R-squared:  0.001605,   Adjusted R-squared:  -0.003338 
## F-statistic: 0.3247 on 1 and 202 DF,  p-value: 0.5694

Removing 2 outliers mitigates 'pCASL.issues' significance.

• YET, still significant predictor for REST SNR

### Remove outlliers & re-run
Clean.Report <-Report[!(Report$SubID==c("20145", "20390")),]
summary(lm(Rest_snr~ pCASL.issues,data=Report))  #SIGNIFICANT

## 
## Call:
## lm(formula = Rest_snr ~ pCASL.issues, data = Report)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -0.78196 -0.18003  0.00224  0.15893  0.80804 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)    2.25078    0.02029  110.95   <2e-16 ***
## pCASL.issues1  0.33546    0.16687    2.01   0.0457 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.2869 on 201 degrees of freedom
##   (1 observation deleted due to missingness)
## Multiple R-squared:  0.01971,    Adjusted R-squared:  0.01483 
## F-statistic: 4.041 on 1 and 201 DF,  p-value: 0.04574

summary(lm(Rest_snr~ Specs,data=Clean.Report)) 

## 
## Call:
## lm(formula = Rest_snr ~ Specs, data = Clean.Report)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -0.7833 -0.1824  0.0070  0.1581  0.8067 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  2.25209    0.02129 105.798   <2e-16 ***
## SpecsReport -0.01249    0.07526  -0.166    0.868    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.2887 on 198 degrees of freedom
##   (1 observation deleted due to missingness)
## Multiple R-squared:  0.0001391,  Adjusted R-squared:  -0.004911 
## F-statistic: 0.02755 on 1 and 198 DF,  p-value: 0.8683

summary(lm(RISE_enc_snr~pCASL.issues,data=Clean.Report)) #NS

## 
## Call:
## lm(formula = RISE_enc_snr ~ pCASL.issues, data = Clean.Report)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -0.77605 -0.17602 -0.00496  0.16696  0.77345 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)    2.24585    0.01995 112.560   <2e-16 ***
## pCASL.issues1  0.21453    0.28288   0.758    0.449    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.2822 on 199 degrees of freedom
## Multiple R-squared:  0.002882,   Adjusted R-squared:  -0.002129 
## F-statistic: 0.5752 on 1 and 199 DF,  p-value: 0.4491

summary(lm(RISE_enc_snr~Specs,data=Clean.Report)) #NS

## 
## Call:
## lm(formula = RISE_enc_snr ~ Specs, data = Clean.Report)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -0.77811 -0.17808 -0.00702  0.16552  0.77139 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  2.24791    0.02077  108.21   <2e-16 ***
## SpecsReport -0.01251    0.07363   -0.17    0.865    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.2826 on 199 degrees of freedom
## Multiple R-squared:  0.0001451,  Adjusted R-squared:  -0.004879 
## F-statistic: 0.02888 on 1 and 199 DF,  p-value: 0.8652

summary(lm(RISE_rec_snr~ pCASL.issues,data=Clean.Report)) #NS

## 
## Call:
## lm(formula = RISE_rec_snr ~ pCASL.issues, data = Clean.Report)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -0.7618 -0.1805  0.0000  0.1624  0.8525 
## 
## Coefficients:
##               Estimate Std. Error t value Pr(>|t|)    
## (Intercept)    2.23744    0.02017 110.931   <2e-16 ***
## pCASL.issues1  0.23255    0.28595   0.813    0.417    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.2852 on 199 degrees of freedom
## Multiple R-squared:  0.003312,   Adjusted R-squared:  -0.001696 
## F-statistic: 0.6614 on 1 and 199 DF,  p-value: 0.4171

summary(lm(RISE_rec_snr~ Specs,data=Clean.Report)) # NS

## 
## Call:
## lm(formula = RISE_rec_snr ~ Specs, data = Clean.Report)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -0.76324 -0.18200  0.01049  0.17228  0.85100 
## 
## Coefficients:
##              Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  2.238892   0.021006 106.583   <2e-16 ***
## SpecsReport -0.003684   0.074453  -0.049    0.961    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.2857 on 199 degrees of freedom
## Multiple R-squared:  1.23e-05,   Adjusted R-squared:  -0.005013 
## F-statistic: 0.002448 on 1 and 199 DF,  p-value: 0.9606