A multi-layer feature attention enhanced network is proposed to further improve the diagnostic accuracy of the severity of diabetic retinopathy.To address the inconsistent expression of global and local features when processing diabetic retinopathy images,a dual-branch parallel model combining ResNet-50 and DeiT-S is employed as the backbone architecture,and a feature fusion module is designed at the end of the network.Concurrently,a multi-scale location awareness enhancement module is developed to extract multi-scale information through dilated convolution with positional attention mechanism for enhancing the feature representation of lesions in fundus images,and a local feature enhancement module is constructed to strengthen the model's capability in extracting local information,thus improving model's capability to identify small lesions and minor changes.The experimental results show that the proposed multi-layer feature attention enhanced network achieves an accuracy of 87.61%,exhibiting excellent classification performance.This advancement provides a strong support for further development of diabetic retinopathy detection technology.

A multi-layer feature attention enhanced network is proposed to further improve the diagnostic accuracy of the severity of diabetic retinopathy.To address the inconsistent expression of global and local features when processing diabetic retinopathy images,a dual-branch parallel model combining ResNet-50 and DeiT-S is employed as the backbone architecture,and a feature fusion module is designed at the end of the network.Concurrently,a multi-scale location awareness enhancement module is developed to extract multi-scale information through dilated convolution with positional attention mechanism for enhancing the feature representation of lesions in fundus images,and a local feature enhancement module is constructed to strengthen the model's capability in extracting local information,thus improving model's capability to identify small lesions and minor changes.The experimental results show that the proposed multi-layer feature attention enhanced network achieves an accuracy of 87.61%,exhibiting excellent classification performance.This advancement provides a strong support for further development of diabetic retinopathy detection technology.

Brain glucose metabolism disorder is one of the pathophysiological features of Alzheimer's disease (AD), in which aerobic glycolysis (Warburg effect) metabolic pathway abnormality is one of the causes of early cognitive impairment in AD, and improving brain energy metabolism has become an important strategy to prevent AD. AMP-activated protein kinase (AMPK) is a central hub regulating glucose metabolism, the most sensitive molecular compound for sensing fluctuations of energy levels in the body. Activation of AMPK can affect the Warburg effect and its key rate-limiting enzyme activity, regulate brain glucose metabolism involved in the pathogenesis of AD, to achieve the purpose of delaying AD progression and improving cognitive function in early clinical stage. In this review, we will discuss the pathogenesis of AD and targeting of AMPK from the perspective of Warburg effect in glucose metabolism.

To observe the effects pf propofol combined with fentanyl on metabolic rate of energy cxpen diture in anesthesia. Method: Thirty-one elective neurosurgical adult patients, ASA class Ⅰ-Ⅱ, received tolal intravenous anesthesia with propofol combined with fentanyh Oxygen consumption (VO_2), carbon dioxide production (VCO_2), respiratory quotient(RQ), metabolic rate(MR)were measured. Resuh: VO_2 and MR increased when patients' posture changed or there existed strong operative stimulation. VO_2 and MR during post induction were 91.09% and 91.29% of the level before anesthesia respectively(P