Skip to main content

Advertisement

Table 4 Comparison of performance for heuristic and machine learning models with additional consideration for pregnant patients and children less than 11 years old

From: Using artificial intelligence to reduce diagnostic workload without compromising detection of urinary tract infections

Model Name AUC Score Accuracy (%) p-value*** PPV NPV Sensitivity (%) Specificity (%) Relative Workload Reduction (%)
Removal of pregnant patients and children (< 11 yrs)*
 Heuristic mode (30 WBC/μl or 150 bacteria/μl)   58·40 NA 39.14 [± 0.73] 96.29 [± 0.17] 95·4 [± 0·14] 44·60 [± 0·34] 33·74 [± 0·39]
 Random Forest (Class weight - 1:8) 0·920 77·09 <  0.001 53.25 [± 0.50] 97.46 [± 0.26] 95·2 [± 0·26] 68·79 [± 0·58] 38·92 [± 0·42]
Combined XGBoost**
 Pregnant patients 0·828 26·94     94·6 [± 0·56] 26·84 [± 1·88] 25·29 [± 0·92]
 Children (< 11 yrs) 0·913 62·00     94·8 [± 0·88] 55·00 [± 2·12] 46·24 [± 1·48]
 Pregnant patients 0·894 71·65     95·3 [± 0·24] 60·93 [± 0·65] 43·38 [± 0·41]
 Combined performance 0.749 65·65 <  0.001 47.64 [± 0.51] 97.14 [± 0.28] 95·2 [± 0·22] 60·93 [± 0·60] 41·18 [± 0·39]
  1. [95% Confidence Interval]
  2. *Pregnant patients and children (< 11 yrs) are not included in the classification process. It is assumed that all patients in these populations will receive culture and this is reflected in the reported relative workload reduction
  3. ** Independent classification algorithms trained and tested on stratified patient populations
  4. *** p-values obtained by comparison to heuristic model by McNemar test