Machine Learning Techniques for the Diagnosis of Schizophrenia Based on Event-Related Potentials.

Antecedent: The event-related potential (ERP) components P300 and mismatch negativity (MMN) have been linked to cognitive deficits in patients with schizophrenia. The diagnosis of schizophrenia could be improved by applying machine learning procedures to these objective neurophysiological biomarkers. Several studies have attempted to achieve this goal, but no study has examined Multiple Kernel Learning (MKL) classifiers. This algorithm finds optimally a combination of kernel functions, integrating them in a meaningful manner, and thus could improve diagnosis. Objective: This study aimed to examine the efficacy of the MKL classifier and the Boruta feature selection method for schizophrenia patients (SZ) and healthy controls (HC) single-subject classification. Methods: A cohort of 54 SZ and 54 HC participants were studied. Three sets of features related to ERP signals were calculated as follows: peak related features, peak to peak related features, and signal related features. The Boruta algorithm was used to evaluate the impact of feature selection on classification performance. An MKL algorithm was applied to address schizophrenia detection. Results: A classification accuracy of 83% using the whole dataset, and 86% after applying Boruta feature selection was obtained. The variables that contributed most to the classification were mainly related to the latency and amplitude of the auditory P300 paradigm. Conclusion: This study showed that MKL can be useful in distinguishing between schizophrenic patients and controls when using ERP measures. Moreover, the use of the Boruta algorithm provides an improvement in classification accuracy and computational cost.

Role of epidermal growth factor and growth hormone-releasing peptide-6 in acceleration of renal tissue repair after kanamycin overdosing in rats.

INTRODUCTION: Aminoglycosides nephrotoxicity limits their use in clinical practice. Growth hormone-releasing peptide-6 (GHRP6) and epidermal growth factor (EGF) have proven cytoprotective effects in various tissues, including the kidney. This study aimed to determine the cytoprotective effect of EGF and GHRP6 on glomerular, proximal tubular, and interstitial morphology in rats treated with an overdose of kanamycin. MATERIALS AND METHODS: Forty-four male Wistar adults rats were submitted to treatment for 20 days with sodium phosphate saline buffer (control group), kanamycin (kanamycin group), kanamycin and EGF (EGF group), kanamycin and GHRP6 (GHRP6 group), kanamycin, EGF, and GHRP6 (EGF-GHRP6 group). The kidneys were studied both during acute kidney injury (n = 19) and recovery phases (n = 25). The percentages of glomerular damage, tubular damage (reversible and irreversible changes), and interstitial damage were quantified in 10 histological fields per kidney using paraffin-embedded sections. RESULTS: The damage in the glomeruli, proximal tubules, and interstitium was less in the groups treated with the cytoprotective treatments than in kanamycin group during acute kidney injury. During the recovery phase, normal structure of several glomeruli and the interstitium was appreciated in the EGF and GHRP6 groups, although tissue repair was not as complete as it in the EGF-GHRP6 group. In the recovery phase, cytoprotective treatments accelerated the recovery of tubular damage and reversible tubular changes prevailed. CONCLUSIONS: These results confirm the cytoprotective properties of EGF and GHRP6 alone and in combination and suggest the possibility of using these agents to accelerate kidney tissue repair after aminoglycoside-induced renal damage.

Cribado auditivo neonatal con potenciales evocados auditivos de estado estable a múltiples frecuencias / Newborn hearing screening test with multiple auditory steady-state responses

Introducción: Las técnicas más utilizadas en los programas de cribado auditivo presentan limitaciones reconocidas en la detección de las pérdidas auditivas. Se presenta una nueva alternativa de cribado auditivo basada en el uso de potenciales evocados de estado estable. Objetivo: Evaluar la factibilidad de una prueba semiautomática de cribado auditivo en una población de recién nacidos sanos. Métodos: Se estudiaron 50 neonatos (7-18 días de nacidos) sin factores clínicos de riesgo y con respuesta detectable bilateral a 40 db nHL con potenciales evocados auditivos de tronco cerebral. A cada uno se le realizó una prueba con potenciales evocados auditivos de estado estable a múltiples frecuencias (PEAee-MF) (500 y 2.000Hz) y un registro sin estímulo para simulación de sordera. Resultados: Los umbrales auditivos se encontraron entre 25 y 50dB HL para ambas frecuencias (42,5±7dB HL para 500Hz y 35,5±6dB HL para 2.000Hz). Los tiempos medios de registro fueron de 2,6±1,6min por cada oído y la duración total del procedimiento (incluyendo preparación y colocación de electrodos) fue de 17,8±3,7min. Con la prueba de cribado se obtuvo una adecuada eficiencia diagnóstica (sensibilidad, 100%, y especificidad, 96%). Conclusiones: Los resultados descritos sugieren que los PEAee-MF (estimulación simultánea con múltiples frecuencias) pudieran ser utilizados como prueba de cribado auditivo en una población de recién nacidos sanos. Para esto, es necesario seguir perfeccionando la técnica combinando una metodología válida y sencilla con determinadas facilidades de automatización (AU)

Introduction: The techniques most frequently used within a screening context (otoacoustic emissions and click auditory brainstem response) have well-known limitations in hearing loss detection. Objective: This study examines the feasibility of a semi-automated multiple auditory steady-state responses (MSSR) system designed for neonatal hearing screening. Methods: A sample of 50 newborns without risk factors (well-babies) was tested within two weeks of birth. All had detectable auditory brainstem responses to clicks down to 40dB nHL in both ears. Two amplitude modulated carrier tones of 500 and 2,000Hz were mixed together and presented simultaneously. Each infant (and ear) was screened with the MSSR system; to simulate a hearing loss, a recording without stimulation was also obtained. Results: Mean auditory thresholds were 42.5±7dB HL at 500Hz and 35.5±6dB HL at 2,000Hz. The average duration of the MSSR recording was 2.6±1.6minutes for each tested ear and the overall duration of the screening procedure (including electrode fitting and infant preparation) was 17.8±3.7minutes. The diagnostic sensibility and the positive predictive values of the MSSR semi-automatic screening system was 100% and 96% respectively, with specificity of 96% and negative predictive values of 100%. Conclusions: Although the diagnostic efficiency of the semi-automated MSSR system was found adequate, further technological improvements are still necessary to facilitate its use in the context of universal newborn hearing screening program (AU)

[Newborn hearing screening test with multiple auditory steady-state responses]. / Cribado auditivo neonatal con potenciales evocados auditivos de estado estable a múltiples frecuencias.

INTRODUCTION: The techniques most frequently used within a screening context (otoacoustic emissions and click auditory brainstem response) have well-known limitations in hearing loss detection. OBJECTIVE: This study examines the feasibility of a semi-automated multiple auditory steady-state responses (MSSR) system designed for neonatal hearing screening. METHODS: A sample of 50 newborns without risk factors (well-babies) was tested within two weeks of birth. All had detectable auditory brainstem responses to clicks down to 40dB nHL in both ears. Two amplitude modulated carrier tones of 500 and 2,000Hz were mixed together and presented simultaneously. Each infant (and ear) was screened with the MSSR system; to simulate a hearing loss, a recording without stimulation was also obtained. RESULTS: Mean auditory thresholds were 42.5±7dB HL at 500Hz and 35.5±6dB HL at 2,000Hz. The average duration of the MSSR recording was 2.6±1.6 minutes for each tested ear and the overall duration of the screening procedure (including electrode fitting and infant preparation) was 17.8±3.7 minutes. The diagnostic sensibility and the positive predictive values of the MSSR semi-automatic screening system was 100% and 96% respectively, with specificity of 96% and negative predictive values of 100%. CONCLUSIONS: Although the diagnostic efficiency of the semi-automated MSSR system was found adequate, further technological improvements are still necessary to facilitate its use in the context of universal newborn hearing screening program.