Table 2.

Performance score on the 200 draws for cohort 2 with XGBoost and DecisionTree algorithms

SubtypePrecisionF1 scoreSpecificitySensitivityNo. of cases
XGBoost      
CLL 0.80 0.88 0.74 0.98 223 
FL 0.86 0.84 1.00 0.82 11 
HCL 0.71 0.72 1.00 0.740 
LPL 0.87 0.66 0.98 0.53 79 
MCL 0.92 0.93 0.99 0.95 44 
MZL 0.90 0.75 0.99 0.64 71 
Mean global accuracy, 0.83 (SD, 0.004); mean weighted accuracy, 0.78 (SD, 0.017) 
DecisionTree      
CLL 0.95 0.95 0.95 0.94 223 
FL 0.92 0.96 1.00 11 
HCL 1.00 0.89 1.00 0.80 
LPL 0.75 0.73 0.95 0.71 79 
MCL 0.82 0.86 0.98 0.91 44 
MZL 0.80 0.82 0.96 0.83 71 
Mean global accuracy, 0.88 (SD, 0.002); mean weighted accuracy, 0.78 (SD, 0.017) 
SubtypePrecisionF1 scoreSpecificitySensitivityNo. of cases
XGBoost      
CLL 0.80 0.88 0.74 0.98 223 
FL 0.86 0.84 1.00 0.82 11 
HCL 0.71 0.72 1.00 0.740 
LPL 0.87 0.66 0.98 0.53 79 
MCL 0.92 0.93 0.99 0.95 44 
MZL 0.90 0.75 0.99 0.64 71 
Mean global accuracy, 0.83 (SD, 0.004); mean weighted accuracy, 0.78 (SD, 0.017) 
DecisionTree      
CLL 0.95 0.95 0.95 0.94 223 
FL 0.92 0.96 1.00 11 
HCL 1.00 0.89 1.00 0.80 
LPL 0.75 0.73 0.95 0.71 79 
MCL 0.82 0.86 0.98 0.91 44 
MZL 0.80 0.82 0.96 0.83 71 
Mean global accuracy, 0.88 (SD, 0.002); mean weighted accuracy, 0.78 (SD, 0.017) 

Results are expressed as mean. F1 score represents the harmonic mean of PPV and sensitivity: 2×(PPV×sensitivity)(PPV+sensitivity). Mean: macro average (mean of performance for each type). Weighted: weighted average (average weighted by the number of data points for each type).

or Create an Account

Close Modal
Close Modal