While Computerised Tomography (CT) may have been the first imaging tool to study human brain, it has not yet been implemented into clinical decision making process for diagnosis of Alzheimers disease (AD). On the other hand, with the nature of being prevalent, inexpensive and non-invasive, CT does present diagnostic features of AD to a great extent. This study explores the significance and impact on the application of the burgeoning deep learning techniques to the task of classification of CT brain images, in particular utilising convolutional neural network (CNN), aiming at providing supplementary information for the early diagnosis of Alzheimers disease. Towards this end, three categories of CT images (N=285) are clustered into three groups, which are AD, Lesion (e.g. tumour) and Normal ageing. In addition, considering the characteristics of this collection with larger thickness along the direction of depth (z) (∼3-5mm), an advanced CNN architecture is established integrating both 2D and 3D CNN networks. The fusion of the two CNN networks is subsequently coordinated based on the average of Softmax scores obtained from both networks consolidating 2D images along spatial axial directions and 3D segmented blocks respectively. As a result, the classification accuracy rates rendered by this elaborated CNN architecture are 85.2%, 80% and 95.3% for classes of AD, Lesion and Normal respectively with an average of 87.6%. Additionally, this improved CNN network appears to outperform the others when in comparison with 2D version only of CNN network as well as a number of state of the art hand-crafted approaches. As a result, these approaches deliver accuracy rates in percentage of 86.3, 85.6±1.10, 86.3±1.04, 85.2±1.60, 83.1±0.35 for 2D CNN, 2D SIFT, 2D 1 KAZE, 3D SIFT and 3D KAZE respectively. The two major contributions of the paper constitute a new 3-D approach while applying deep learning technique to extract signature information rooted in both 2D slices and 3D blocks of CT images and an elaborated hand-crated approach of 3D KAZE.

No abstract

The ventrolateral periaqueductal gray (vlPAG) is an important brain area, in which 5-HTergic neurons play key roles in descending pain modulation. It has been proposed that opioid peptides within the vlPAG can excite the 5-HTergic neurons by alleviating tonic inhibition from GABAergic neurons, the so-called disinhibitory effect. However, no direct morphological evidence has been observed for the micro-circuitry among the opioid peptide-, GABA-, and 5-HT-immunoreactive (ir) profiles nor for the functional involvement of the opioid peptides in the intrinsic properties of GABAergic and 5-HTergic neurons. In the present study, through microscopic observation of triple-immunofluorescence, we firstly identified the circuitry among the endomorphin-1 (EM1, an endogenous ligand for the μ-opioid receptor)-ir terminals and GABA-ir and 5-HT-ir neurons within the rat vlPAG. The synaptic connections of these neurons were further confirmed by electron microscopy. Through the in vitro whole-cell patch-clamp method, we showed that EM1 has strong inhibitory effects on the spiking of GABAergic neurons. However, although the resting membrane potential was hyperpolarized, EM1 actually increased the firing of 5-HTergic neurons. More interestingly, EM1 strongly inhibited the excitatory input to GABAergic neurons, as well as the inhibitory input to 5-HTergic neurons. Finally, behavioral results showed that pretreatment with a GABA(A) receptor antagonist potentiated the analgesic effect of EM1, while treatment with a GABA(A) receptor agonist blocked its analgesic effect. In summary, by utilizing morphological and functional methods, we found that the analgesic effect of EM1 is largely dependent on its potent inhibition on the inhibitory inputs to 5-HTergic neurons, which overwhelms EM1's direct inhibitory effect on 5-HTergic neurons.

Endomorphin 1 (EM1) and endomorphin 2 (EM2) are endogenous ligands for mu-opioid receptors (MOR). In the central nervous system, EM-immunoreactive (IR) neuronal cell bodies are located mainly in the hypothalamus and the nucleus tractus solitarius (NTS). EM-IR fibers and terminals are found widely distributed in many brain areas, including the different columns of the periaqueductal gray (PAG). The hypothalamus, NTS, and PAG are closely involved in modulation of vocalization, autonomic and neuroendocrine functions, pain, and defensive behavior through endogenous opioid peptides that bind to the MOR in these regions. Projections exist from both the hypothalamus and the NTS to the PAG. In order to examine whether there are EM1- and/or EM2-ergic projections from the hypothalamus and NTS to the PAG, immunofluorescence histochemistry for EM1 and/or EM2 was combined with fluorescent retrograde tracing. In rats that had Fluoro-Gold (FG) injected into different columns of the PAG, some of the EM1- or EM2-IR neurons in the hypothalamus, but none in the NTS, were labeled retrogradely with FG. The majority of the EM1/FG and EM2/FG double-labeled neurons in the hypothalamus were distributed in the dorsomedial nucleus, areas between the dorsomedial and ventromedial nucleus, and arcuate nucleus; a few were also seen in the ventromedial, periventricular, and posterior nucleus. The present results indicate that the EM-IR fibers and terminals in the PAG originate principally from the hypothalamus. They also suggest that EMs released from hypothalamus-PAG projecting neurons might mediate or modulate various functions of the PAG through binding to the MOR.

No abstract