Computational and Mathematical Methods in Medicine / 2015 / Article / Tab 6

Research Article

Binary Matrix Shuffling Filter for Feature Selection in Neuronal Morphology Classification

Table 6

Ability to distinguish one single cell type from others and the obtained private feature subsets by BMSF-SVC.

Positive versus negative cell typeAccuracy (%)MCC (%)Recall (%)Private feature subsets

{motoneuron, sensory, tripolar,
bipolar, multipolar, Purkinje}
99.10 ± 0.1297.05 ± 0.4099.76 ± 0.10, Lpd, NW, Co, Pk2, HT, Bar, Su, PDR, Ta2
, Lpd, NW, Bal, Pk2, Vo, Bar, HT, PDR, Ta2
, Lpd, NW, Bal, Pk2, Vo, Bar, HT, Pc, Ta2
, Lpd, Bal, NW, HT, Pc, Vo, PDR, Ta2
, Lpd, Co, NW, HT, PDR, Vo, Bar, Ta2

{sensory, tripolar, bipolar,
multipolar, Purkinje}
97.26 ± 1.4494.3 ± 3.0294.50 ± 5.21SS
SS, NH, , Ta1, SA, Ta2, HT, Su, NW, Vo
SS, NH, , HT, Vo, NW, Lpd, Dp, Su, SA
SS, NH, , Lpd, Ta1, Vo, HT, Le, Ta2, SA
SS, NH, , HT, SA, Ta1, Vo, ND, Le, Dp

{tripolar, bipolar, multipolar,
90.15 ± 1.2480.62 ± 2.4697.85 ± 1.38Pa
Pa, SS, SA, Ta1, ND, Pk2, Btr, NW, Pk, Btl
Pa, SS, SA, Ta1, ND, Ty, Co, Di, Btr
Pa, SS, SA, Ta1, ND, Ty, Di
Pa, SS, SA, ND, Ta1, Ty, Btr, NW, Lpd, Pk

{bipolar, multipolar, Purkinje}
99.16 ± 0.5698.32 ± 1.1299.17 ± 1.41NW
NW, SS, He, Pa, ND,
NW, SS, He
NW, SS, He
NW, SS, He, Pa, ND

{multipolar, Purkinje}
96.95 ± 3.0793.86 ± 6.2495.83 ± 2.95
, Vo, He, Ty, Su
, Vo, Su, Ty, He, Ta2, NW, Btor, Pk
, Vo, Su, NW, Ta2, Di, Pc, He, Btor
, Vo, Su, He, Di, Ta2, Btor, Pk

100.00 ± 0.00100.00 ± 0.00100.00 ± 0.00DR