A lightweight fire detection algorithm for small targets based on YOLOv5s.

In response to the current challenges fire detection algorithms encounter, including low detection accuracy and limited recognition rates for small fire targets in complex environments, we present a lightweight fire detection algorithm based on an improved YOLOv5s. The introduction of the CoT (Contextual Transformer) structure into the backbone neural network, along with the creation of the novel CSP1_CoT (Cross stage partial 1_contextual transformer) module, has effectively reduced the model's parameter count while simultaneously enhancing the feature extraction and fusion capabilities of the backbone network; The network's Neck architecture has been extended by introducing a dedicated detection layer tailored for small targets and incorporating the SE (Squeeze-and-Excitation) attention mechanism. This augmentation, while minimizing parameter proliferation, has significantly bolstered the interaction of multi-feature information, resulting in an enhanced small target detection capability; The substitution of the original loss function with the Focal-EIoU (Focal-Efficient IoU) loss function has yielded a further improvement in the model's convergence speed and precision; The experimental results indicate that the modified model achieves an mAP@.5 of 96% and an accuracy of 94.8%, marking improvements of 8.8% and 8.9%, respectively, over the original model. Furthermore, the model's parameter count has been reduced by 1.1%, resulting in a compact model size of only 14.6MB. Additionally, the detection speed has reached 85 FPS (Frames Per Second), thus satisfying real-time detection requirements. This enhancement in precision and accuracy, while simultaneously meeting real-time and lightweight constraints, effectively caters to the demands of fire detection.

miR-222-5p promotes dysfunction of human vascular smooth muscle cells by targeting RB1.

BACKGROUND: Coronary atherosclerosis (AS) is characterized by the formation of plaque in the vessel wall. The structural and functional changes of vascular smooth muscle cells (VSMCs) can promote plaque formation and induce plaque instability. OBJECTIVE: To investigate the functions and mechanism of miR-222-5p in VSMCs under the treatment of oxidized low-density lipoprotein (ox-LDL). METHODS: miR-222-5p expression in ox-LDL-treated VSMCs and the serum of Apolipoprotein E (ApoE) knockout mice was detected by reverse transcription quantitative polymerase chain reaction. The viability and migration of VSMCs were detected by Cell Counting Kit-8 and Transwell assays. Protein levels of proliferation and migration-related factors were evaluated by western blotting. Luciferase reporter assays were performed to explore the binding between miR-222-5p and retinoblastoma susceptibility protein (RB1) gene in VSMCs. ApoE-knockout mice were infected with the lentivirus inhibiting miR-222-5p expression to explore the effect of miR-222-5p on pathological changes. Hematoxylin and eosin (H&E) staining, trichrome staining, and Oil Red O staining were conducted to determine the necrotic core area and atherosclerotic lesion size in the ascending aorta of ApoE-knockout mice. RESULTS: With the accumulation of ox-LDL concentration and treatment time, miR-222-5p expression was gradually upregulated in VSMCs. Similarly, miR-222-5p expression was increased in the serum of ApoE-knockout mice. miR-222-5p knockdown inhibited the proliferative and migratory abilities of ox-LDL-treated VSMCs, and the inhibitory effect on cellular behaviors was then significantly reversed by co-knockdown of RB1. RB1 is a downstream target gene of miR-222-5p, and miR-222-5p bound with 3'-untranslated region of RB1 in VSMCs. We further confirmed that miR-222-5p knockdown alleviated pathological changes and inhibited lipid deposition in the serum of ApoE-knockout mice in vivo. CONCLUSION: miR-222-5p accelerates the dysfunction of VSMCs and promotes pathological changes and lipid deposition in ApoE-knockout mice by targeting RB1. The study may provide novel therapeutic targets for AS.

A robo-pigeon based on an innovative multi-mode telestimulation system.

In this paper, we describe a new multi-mode telestimulation system for brain-microstimulation for the navigation of a robo-pigeon, a new type of bio-robot based on Brain-Computer Interface (BCI) techniques. The multi-mode telestimulation system overcomes neuron adaptation that was a key shortcoming of the previous single-mode stimulation by the use of non-steady TTL biphasic pulses accomplished by randomly alternating pulse modes. To improve efficiency, a new behavior model ("virtual fear") is proposed and applied to the robo-pigeon. Unlike the previous "virtual reward" model, the "virtual fear" behavior model does not require special training. The performance and effectiveness of the system to alleviate the adaptation of neurons was verified by a robo-pigeon navigation test, simultaneously confirming the practicality of the "virtual fear" behavioral model.