Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 2
Encontrada(s) 1 página(s)

Dipòsit Digital de Documents de la UAB
Publicación Artículo científico (JournalArticle)

Widespread intra-axonal signal fraction abnormalities in bipolar disorder from multicompartment diffusion, Sensitivity to diagnosis, association with clinical features and pharmacologic treatment

  • Canales-Rodriguez, Erick Jorge|||0000-0001-6421-2633
  • Verdolini, Norma|||0000-0001-9488-2881
  • Alonso-Lana, Silvia|||0000-0002-3063-6929
  • Torres, María Llanos
  • Panicalli, Francesco
  • Argila-Plaza, Isabel|||0000-0002-1604-4708
  • Rodriguez-Cano, Elena
  • Montoro, Irene
  • Garcia-Ruiz, Beatriz
  • Jimenez, Esther|||0000-0001-6929-6207
  • Varo, Cristina|||0000-0002-8859-3052
  • Lluch, Anna
  • del Mar Bonnin, Caterina
  • Maluf, Silvana
  • Pujol, Marc
  • Jaurrieta Guarner, Núria
  • Sarró, Salvador|||0000-0003-1835-2189
  • Vieta, Eduard|||0000-0002-0548-0053
  • Vilella, Elisabet|||0000-0002-1887-5919
  • Salvador, Raymond|||0000-0001-5557-1562
  • Pomarol-Clotet, Edith|||0000-0002-8159-8563
Despite diffusion tensor imaging (DTI) evidence for widespread fractional anisotropy (FA) reductions in the brain white matter of patients with bipolar disorder, questions remain regarding the specificity and sensitivity of FA abnormalities as opposed to other diffusion metrics in the disorder. We conducted a whole-brain voxel-based multicompartment diffusion MRI study on 316 participants (i.e., 158 patients and 158 matched healthy controls) employing four diffusion metrics: the mean diffusivity (MD) and FA estimated from DTI, and the intra-axonal signal fraction (IASF) and microscopic axonal parallel diffusivity (Dpar) derived from the spherical mean technique. Our findings provide novel evidence about widespread abnormalities in other diffusion metrics in BD. An extensive overlap between the FA and IASF results suggests that the lower FA in patients may be caused by a reduced intra-axonal volume fraction or a higher macromolecular content in the intra-axonal water. We also found a diffuse alteration in MD involving white and grey matter tissue and more localised changes in Dpar. A Machine Learning analysis revealed that FA, followed by IASF, were the most helpful metric for the automatic diagnosis of BD patients, reaching an accuracy of 72%. Number of mood episodes, age of onset/duration of illness, psychotic symptoms, and current treatment with lithium, antipsychotics, antidepressants, and antiepileptics were all significantly associated with microstructure abnormalities. Lithium treatment was associated with less microstructure abnormality. A whole-brain voxel-based approach combining DTI and SMT diffusion-derived metrics to reveal specific brain tissue microstructure abnormalities in bipolar disorder (BD) was carried out. We also evaluate the sensitivity of various diffusion metrics for the automatic diagnosis of BD through machine learning algorithms and cross-validation. Finally, we look for differences between BD subtypes and the possible effect of duration of illness, age at onset, history of psychosis, and pharmacological treatment with lithium, antidepressant, antipsychotic, and antiepileptic drugs.
Dipòsit Digital de Documents de la UAB
Publicación Artículo científico (JournalArticle)

Evaluation of machine learning algorithms and structural features for optimal MRI-based diagnostic prediction in psychosis

  • Salvador, Raymond|||0000-0001-5557-1562
  • Radua, Joaquim|||0000-0003-1240-5438
  • Canales-Rodriguez, Erick Jorge|||0000-0001-6421-2633
  • Solanes, Aleix|||0000-0002-2491-200X
  • Sarró, Salvador|||0000-0003-1835-2189
  • Goikolea Alberdi, Jose Manuel|||0000-0003-0676-0791
  • Valiente, Alicia
  • Monté, Gemma C.
  • Natividad, María del Carmen
  • Guerrero-Pedraza, Amalia|||0000-0001-7567-9407
  • Moro, Noemí
  • Fernández-Corcuera, Paloma
  • Amann, Benedikt L.|||0000-0002-4407-1519
  • Maristany, Teresa
  • Vieta, Eduard|||0000-0002-0548-0053
  • McKenna, Peter J.
  • Pomarol-Clotet, Edith|||0000-0002-8159-8563
A relatively large number of studies have investigated the power of structural magnetic resonance imaging (sMRI) data to discriminate patients with schizophrenia from healthy controls. However, very few of them have also included patients with bipolar disorder, allowing the clinically relevant discrimination between both psychotic diagnostics. To assess the efficacy of sMRI data for diagnostic prediction in psychosis we objectively evaluated the discriminative power of a wide range of commonly used machine learning algorithms (ridge, lasso, elastic net and L0 norm regularized logistic regressions, a support vector classifier, regularized discriminant analysis, random forests and a Gaussian process classifier) on main sMRI features including grey and white matter voxel-based morphometry (VBM), vertex-based cortical thickness and volume, region of interest volumetric measures and wavelet-based morphometry (WBM) maps. All possible combinations of algorithms and data features were considered in pairwise classifications of matched samples of healthy controls (N = 127), patients with schizophrenia (N = 128) and patients with bipolar disorder (N = 128). Results show that the selection of feature type is important, with grey matter VBM (without data reduction) delivering the best diagnostic prediction rates (averaging over classifiers: schizophrenia vs. healthy 75%, bipolar disorder vs. healthy 63% and schizophrenia vs. bipolar disorder 62%) whereas algorithms usually yielded very similar results. Indeed, those grey matter VBM accuracy rates were not even improved by combining all feature types in a single prediction model. Further multi-class classifications considering the three groups simultaneously made evident a lack of predictive power for the bipolar group, probably due to its intermediate anatomical features, located between those observed in healthy controls and those found in patients with schizophrenia. Finally, we provide MRIPredict (), a free tool for SPM, FSL and R, to easily carry out voxelwise predictions based on VBM images.