Application and comparison of classification algorithms for recognition of Alzheimer's disease in electrical brain activity (EEG)

The early detection of subjects with probable Alzheimer's disease (AD) is crucial for effective appliance of treatment strategies. Here we explored the ability of a multitude of linear and non-linear classification algorithms to discriminate between the electroencephalograms (EEGs) of patients with varying degree of AD and their age-matched control subjects. Absolute and relative spectral power, distribution of spectral power, and measures of spatial synchronization were calculated from recordings of resting eyes-closed continuous EEGs of 45 healthy controls, 116 patients with mild AD and 81 patients with moderate AD, recruited in two different centers (Stockholm, New York). The applied classification algorithms were: principal component linear discriminant analysis (PC LDA), partial least squares LDA (PLS LDA), principal component logistic regression (PC LR), partial least squares logistic regression (PLS LR), bagging, random forest, support vector machines (SVM) and feed-forward neural network. Based on 10-fold cross-validation runs it could be demonstrated that even tough modern computer-intensive classification algorithms such as random forests, SVM and neural networks show a slight superiority, more classical classification algorithms performed nearly equally well. Using random forests classification a considerable sensitivity of up to 85% and a specificity of 78%, respectively for the test of even only mild AD patients has been reached, whereas for the comparison of moderate AD vs. controls, using SVM and neural networks, values of 89% and 88% for sensitivity and specificity were achieved. Such a remarkable performance proves the value of these classification algorithms for clinical diagnostics.

Candida species distribution and antifungal susceptibility testing according to European Committee on Antimicrobial Susceptibility Testing and new vs. old Clinical and Laboratory Standards Institute clinical breakpoints: a 6-year prospective candidaemia survey from the fungal infection network of Switzerland

We analyzed the species distribution of Candida blood isolates (CBIs), prospectively collected between 2004 and 2009 within FUNGINOS, and compared their antifungal susceptibility according to clinical breakpoints defined by the European Committee on Antimicrobial Susceptibility Testing (EUCAST) in 2013, and the Clinical and Laboratory Standards Institute (CLSI) in 2008 (old CLSI breakpoints) and 2012 (new CLSI breakpoints). CBIs were tested for susceptiblity to fluconazole, voriconazole and caspofungin by microtitre broth dilution (Sensititre® YeastOne™ test panel). Of 1090 CBIs, 675 (61.9%) were C. albicans, 191 (17.5%) C. glabrata, 64 (5.9%) C. tropicalis, 59 (5.4%) C. parapsilosis, 33 (3%) C. dubliniensis, 22 (2%) C. krusei and 46 (4.2%) rare Candida species. Independently of the breakpoints applied, C. albicans was almost uniformly (>98%) susceptible to all three antifungal agents. In contrast, the proportions of fluconazole- and voriconazole-susceptible C. tropicalis and F-susceptible C. parapsilosis were lower according to EUCAST/new CLSI breakpoints than to the old CLSI breakpoints. For caspofungin, non-susceptibility occurred mainly in C. krusei (63.3%) and C. glabrata (9.4%). Nine isolates (five C. tropicalis, three C. albicans and one C. parapsilosis) were cross-resistant to azoles according to EUCAST breakpoints, compared with three isolates (two C. albicans and one C. tropicalis) according to new and two (2 C. albicans) according to old CLSI breakpoints. Four species (C. albicans, C. glabrata, C. tropicalis and C. parapsilosis) represented >90% of all CBIs. In vitro resistance to fluconazole, voriconazole and caspofungin was rare among C. albicans, but an increase of non-susceptibile isolates was observed among C. tropicalis/C. parapsilosis for the azoles and C. glabrata/C. krusei for caspofungin according to EUCAST and new CLSI breakpoints compared with old CLSI breakpoints.