Development of an efficient algorithm for the detection of macular edema from optical coherence tomography images

Purpose

Detection of eye diseases and their treatment is a key to reduce blindness, which impacts human daily needs like driving, reading, writing, etc. Several methods based on image processing have been used to monitor the presence of macular diseases. Optical coherence tomography (OCT) imaging is the most efficient technique used to observe eye diseases. This paper proposes an efficient algorithm to automatically classify normal as well as disease-affected (macular edema) retinal OCT images by using segmentation of Inner Limiting Membrane and the Choroid Layer.

Methods

In the proposed method, preprocessing of the input image is done to improve the quality and reduce the speckle noise. The layer segmentation is done on the gradient image, and graph theory and dynamic programming algorithm is performed. The feature vectors from segmented image are in terms of thickness profile and cyst fluid parameter, and these features are applied to various classifiers.

Results

The proposed method was tested with the standard dataset collected from the Department of Ophthalmology, Duke University, and achieved a high accuracy rate of 99.4975%, sensitivity of 100%, and specificity of 99% for the SVM classifier.

Conclusions

An efficient algorithm is proposed for macular edema detection from OCT images using segmentation based on graph theory and dynamic programming algorithm. The comparison with alternative methods yielded results that demonstrate the superiority of the proposed algorithm for macular edema detection.

     

Cyclosporin as initial treatment for Crohn's disease

Childhood Crohn's disease may cause significant morbidity. T cell activation is considered to be central to Crohn's disease pathology, and as cyclosporin is a powerful inhibitor of T cell activation, and has been used in adult Crohn's disease with encouraging results, it may offer the prospect of remission if given early in the course of disease. Children with newly diagnosed Crohn's disease or those relapsing off treatment were therefore given cyclosporin or conventional treatment (enteral nutrition or corticosteroids) by random allocation. Evaluation was performed initially and at two months. Twenty four children were studied (10 on cyclosporin and 14 on conventional treatment; one child on cyclosporin withdrew). Significant clinical improvement occurred in the group on conventional treatment, but not in the cyclosporin group. Colonoscopic improvement was noted in 5/9 on cyclosporin and 8/14 on conventional treatment, but neither group produced a significant fall in median colonoscopic index. Histological improvement was seen in 7/8 on cyclosporin and 8/13 on conventional treatment, but cyclosporin was not significantly better. Cyclosporin produced improved clinical and histological appearance without matched improvement in blood disease indices. It was not better than conventional treatment, and simple oral administration is probably not suitable for newly diagnosed patients with Crohn's disease.     

Assessment of future remnant liver regeneration after portal vein embolization using three‐dimensional CT and MR volumetric analyses

The purpose of this study is to portray right portal vein embolization (PVE) as a valuable technique that helps in expanding the volume of the left liver lobe and discuss the relevant published work. We describe our experience with four patients who underwent PVE and analyse the value of CT and MRI in the preoperative evaluation of these patients. Four patients with hepatic malignancy (hepatocellular carcinoma) (n = 2) and metastatic liver disease (n = 2) underwent portal vein occlusion. PVE was carried out in three patients using polyvinyl alcohol and stainless steel coils. Portal vein ligation was carried out in the fourth patient. In patients who were candidates for right hepatectomy, CT volumetric analysis was carried out before the surgery to assess the total liver volume and the future remnant liver, which is the residual left hepatic volume (in cases of right hepatectomy) or left lateral segment volume (in cases of right tri‐segmentectomy). Because the left lobe volumes were insufficient, patients were selected to undergo right PVE. Computed tomography volumetry was carried out 2–4 weeks after embolization to assess left hepatic lobe regeneration. Magnetic resonance volumetric analysis was carried out in two patients before and after embolization. All four patients had significant regeneration of the left lobe and tolerated the surgery with uneventful postoperative recovery.     

Effects of nimodipine,Bay K 8644 and pinacidil on α1- and α2-adrenoceptor-mediated vasoconstriction in human hand veins

The effects of nimodipine, Bay K 8644 and pinacidil, three drugs interfering with transmembrane Ca²⁺ fluxes in different ways, were investigated in isolated human hand veins. Their ability to influence the concentration-response relationship for noradrenaline (NA) was assessed in the absence and presence of prazosin or rauwolscine. The contractile response to NA was almost abolished in Ca²⁺-free medium. Nimodipine and pinacidil depressed the NA concentration-response curve both in the absence and presence of α-adrenoceptor blockers. The NA response was only partially inhibited by nimodipine, indicating that NA may activate nimodipine-insensitive influx pathways, presumably receptor-operated calcium channels. Pinacidil inhibited the contractile response to 124 mM K⁺ and reduced the NA-induced contraction in the presence of nimodipine, suggesting that pinacidil has actions other than the opening of potassium channels and subsequent membrane hyperpolarization. Bay K 8644 increased the NA potency fourfold in the presence of rauwolscine, whereas it had no effect on the NA response in the presence of prazosin and in the absence of α-adrenoceptor blockade. Such an action of Bay K 8644 can be reconciled with α-adrenoceptor activation causing membrane depolarization and opening of potential-operated calcium channels. It may be concluded that both α₁- and α₂-adrenoceptor-mediated contractions in human hand veins are highly dependent on Ca²⁺ influx, although the mechanisms utilized to bring about this influx partly differ between the two receptor subtypes.     

The endothelium mediates a nitric oxide-independent hyperpolarization and relaxation in the rat hepatic artery

The rat hepatic artery responds to acetylcholine (ACh) with an endothelium-dependent relaxation, which is unaffected by nitric oxide (NO) synthase and cyclooxygenase inhibition. The purpose of this study was to investigate whether the NO-independent relaxation is caused by hyperpolarization of the smooth muscle cells. In vessels with endothelium ACh induced a hyperpolarization in the presence of 0.3 mM Nw-nitro-l -arginine (l -NOARG) and 10μm indomethacin. The hyperpolarization, which slowly decayed after an initial maximum, generally lasted for at least 20 min. ACh in contrast to levcromakalim failed to hyperpolarize the smooth muscle cells in endothelium-denuded vessels. In vessels contracted by phenylephrine (PhE) ACh caused a concentration-dependent hyperpolarization and relaxation, and both events occurred over the same concentration interval. Curve fitting using the Hill equation showed a close correlation between the hyperpolarization and the relaxation. Exposure to a 30 mM K⁺ solution abolished the hyperpolarization and suppressed the relaxation induced by ACh. Nimodipine did not affect the ACh-induced hyperpolarization, whereas the relaxation induced by ACh and levcromakalim, but not that evoked by the NO donor 3-morpholino-sydnonimin, were attenuated. Glibenclamide had no effect on the ACh-induced hyperpolarization and relaxation, but abolished the corresponding responses to levcromakalim. The results demonstrate a NO-independent hyperpolarization and relaxation in the rat hepatic artery. The hyperpolarization and relaxation were endothelium-dependent, and apparently causally related to each other, since interference with the hyperpolarization or the subsequent effector pathway inhibited the relaxation.     

Regional differences in endothelium-dependent relaxation in the rat: contribution of nitric oxide and nitric oxide-independent mechanisms

Relaxant effects of acetylcholine (ACh), histamine, calcitonin gene-related peptide (CGRP) and the calcium ionophore A23187 were examined in rat femoral (Ø ? 0.2 mm), mesenteric (0.2 mm), intrarenal (0.2 mm) and hepatic (0.3 mm) arteries, and aorta (2 mm). Acetylcholine elicited an endothelium-dependent relaxation in all arteries. Histamine induced an endothelium-dependent relaxation in aorta, and mesenteric and intrarenal arteries, whereas a partly endothelium-dependent and mainly endothelium-independent relaxation was observed in hepatic and femoral arteries, respectively. In hepatic, mesenteric and intrarenal arteries, CGRP induced an endothelium-independent relaxation, whereas either small or no relaxation was obtained in aorta and femoral arteries respectively. A23187 induced an endothelium-dependent relaxation in the aorta and hepatic artery, whereas A23187 had no relaxant effect in femoral, mesenteric and intrarenal arteries. Nω-nitro-l -arginine (l -NOARG, 0.3 mM) reduced the maximum ACh-induced relaxation (in the presence of 10 μM indomethacin) by 66% in the aorta, and abolished the relaxation in femoral and intrarenal arteries. A marked l -NOARG/indomethacin-resistant relaxation was obtained in mesenteric and hepatic arteries. Levcromakalim induced a concentration-dependent and almost complete relaxation in all arteries. When contracted by a 60 mM K⁺ solution, all arteries responded to ACh with a relaxation that was abolished by l -NOARG. These results demonstrate marked regional differences with regard to the vascular effects of ACh, histamine, CGRP and A23187. Whereas nitric oxide appears to mediate endothelium- dependent relaxation regardless of the vascular region, an l -NOARG/indomethacin-resistant relaxation, presumably mediated by an endothelium-derived hyperpolarizing factor, was observed only in mesenteric and hepatic arteries, and aorta.     

The survival motor neuron protein in spinal muscular atrophy

The 38 kDa survival motor neuron (SMN) protein is encoded by two ubiquitously expressed genes: telomeric SMN (SMN(T)) and centromeric SMN (SMN(C)). Mutations in SMN(T), but not SMN(C), cause proximal spinal muscular atrophy (SMA), an autosomal recessive disorder that results in loss of motor neurons. SMN is found in the cytoplasm and nucleus. The nuclear form is located in structures termed gems. Using a panel of anti-SMN antibodies, we demonstrate that the SMN protein is expressed from both the SMN(T) and SMN(C) genes. Western blot analysis of fibroblasts from SMA patients with various clinical severities of SMA showed a moderate reduction in the amount of SMN protein, particularly in type I (most severe) patients. Immunocytochemical analysis of SMA patient fibroblasts indicates a significant reduction in the number of gems in type I SMA patients and a correlation of the number of gems with clinical severity. This correlation to phenotype using primary fibroblasts may serve as a useful diagnostic tool in an easily accessible tissue. SMN is expressed at high levels in brain, kidney and liver, moderate levels in skeletal and cardiac muscle, and low levels in fibroblasts and lymphocytes. In SMA patients, the SMN level was moderately reduced in muscle and lymphoblasts. In contrast, SMN was expressed at high levels in spinal cord from normals and non- SMA disease controls, but was reduced 100-fold in spinal cord from type I patients. The marked reduction of SMN in type I SMA spinal cords is consistent with the features of this motor neuron disease. We suggest that disruption of SMN(T) in type I patients results in loss of SMN from motor neurons, resulting in the degeneration of these neurons.      

Spinal expression of mRNA for immediate early genes in a model of chronic pain

We have used a partial nerve ligation model of chronic pain to investigate if there are changes in the expression of mRNA for several immediate early genes (IEG) that correlate in time with the initial adaptive behavioural changes and with development of allodynia in this model. The animals were inspected for typical changes in posture, and mechanical allodynia was evaluated using von Frey filaments. Expression of three of the immediate early genes examined, c-fos, NGFI-A and jun B, was transiently increased in the ipsilateral dorsal horn of the spinal cord following the partial ligation of the sciatic nerve. The time course and extent of these changes were similar to those reported for acute noxious stimuli. c-jun mRNA expression was significantly enhanced, after a delay of more than 12 h, and then remained elevated over the entire studied period of 4 weeks. These changes occurred only in the ventral horn, particularly in lamina IX. Except for c-jun mRNA, all changes were transient despite behavioural evidence for continuing allodynia. These results from the partial nerve ligation model, when compared with results obtained using other models of acute or chronic nerve injury, suggest that the immediate early genes we have examined are not sufficient to explain the transition to chronic pain states. The results also show that in this model of chronic pain there are prolonged adaptive changes in motor neurons and that these changes are temporally associated with the development of chronic pain and allodynia.