Search

In our day-to-day life, synchronizing vision and language aspects plays a crucial role in solving various real-time challenges. Image captioning is one of them, and it aims to recognise objects, activities, and their relationships in order to provide a syntactically and semantically correct visual description. There are existing works of image captioning in various directions, such as news, fashion, art, and medical domains. The core architectural idea of image captioning is based on merging CNN, RNN, and transformer models. In practice, there are many conceivable combinations, and brute forcing all of them would take a long time. As we know, there is no work on interpreting image captioning models across various usecases. In this research article, we examine and analyze different image captioning models used across various domains, and multiple insights are extracted to determine the best combinational architecture for a new application without ignoring contextual semantics. We examined numerous designs and determined that LSTM is best for image captioning across several domains.

Effective accident management acts as a vital part of emergency and traffic control systems. In such systems, accident data can be collected from different sources (unmanned aerial vehicles, surveillance cameras, on-site people, etc.) and images are considered a major source. Accident site photos and measurements are the most important evidence. Attackers will steal data and breach personal privacy, causing untold costs. The massive number of images commonly employed poses a significant challenge to privacy preservation, and image encryption can be used to accomplish cloud storage and secure image transmission. Automated severity estimation using deep-learning (DL) models becomes essential for effective accident management. Therefore, this article presents a novel Privacy Preserving Image Encryption with Optimal Deep-Learning-based Accident Severity Classification (PPIE-ODLASC) method. The primary objective of the PPIE-ODLASC algorithm is to securely transmit the accident images and classify accident severity into different levels. In the presented PPIE-ODLASC technique, two major processes are involved, namely encryption and severity classification (i.e., high, medium, low, and normal). For accident image encryption, the multi-key homomorphic encryption (MKHE) technique with lion swarm optimization (LSO)-based optimal key generation procedure is involved. In addition, the PPIE-ODLASC approach involves YOLO-v5 object detector to identify the region of interest (ROI) in the accident images. Moreover, the accident severity classification module encompasses Xception feature extractor, bidirectional gated recurrent unit (BiGRU) classification, and Bayesian optimization (BO)-based hyperparameter tuning. The experimental validation of the proposed PPIE-ODLASC algorithm is tested utilizing accident images and the outcomes are examined in terms of many measures. The comparative examination revealed that the PPIE-ODLASC technique showed an enhanced performance of 57.68 dB over other existing models.

Thermal cameras, as opposed to RBG cameras, work effectively in extremely low illumination situations and can record data outside of the human visual spectrum. For surveillance and security applications, thermal images have several benefits. However, due to the little visual information in thermal images and intrinsic similarity of facial heat maps, completing face identification tasks in the thermal realm is particularly difficult. It can be difficult to attempt identification across modalities, such as when trying to identify a face in thermal images using the ground truth database for the matching visible light domain or vice versa. We proposed a method for detecting objects and actions on thermal human face images, based on the classification of five different features (hat, glasses, rotation, normal, and hat with glasses) in this paper. This model is presented in five steps. To improve the results of feature extraction during the pre-processing step, initially, we resize the images and then convert them to grayscale level using a median filter. In addition, features are extracted from pre-processed images using principle component analysis (PCA). Furthermore, the horse herd optimization algorithm (HOA) is employed for feature selection. Then, to detect the human face in thermal images, the LeNet-5 method is used. It is utilized to detect objects and actions in face areas. Finally, we classify the objects and actions on faces using the ANU-Net approach with the Monarch butterfly optimization (MBO) algorithm to achieve higher classification accuracy. According to experiments using the Terravic Facial Infrared Database, the proposed method outperforms “state-of-the-art” methods for face recognition in thermal images. Additionally, the results for several facial recognition tasks demonstrate good precision.

Background and Objective: Imaging methodologies such as, computed tomography (CT) aid in three-dimensional (3D) reconstruction of patient-specific aneurysms. The radiological data is useful in understanding their location, shape, size, and disease progression. However, there are serious impediments in discerning the blood vessel wall thickness due to limitations in the current imaging modalities. This further restricts the ability to perform high-fidelity fluid structure interaction (FSI) studies for an accurate assessment of rupture risk. FSI studies would require the arterial wall mesh to be generated to determine realistic maximum allowable wall stresses by performing coupled calculations for the hemodynamic forces with the arterial walls., Methods: In the present study, a novel methodology is developed to geometrically model variable vessel wall thickness for the lumen isosurface extracted from CT scan slices of patient-specific aneurysms based on clinical and histopathological inputs. FSI simulations are carried out with the reconstructed models to assess the importance of near realistic wall thickness model on rupture risk predictions., Results: During surgery, clinicians often observe translucent vessel walls, indicating the presence of thin regions. The need to generate variable vessel wall thickness model, that embodies the wall thickness gradation, is closer to such clinical observations. Hence, corresponding FSI simulations performed can improve clinical outcomes. Considerable differences in the magnitude of instantaneous wall shear stresses and von Mises stresses in the walls of the aneurysm was observed between a uniform wall thickness and a variable wall thickness model., Conclusion: In the present study, a variable vessel wall thickness generation algorithm is implemented. It was shown that, a realistic wall thickness modeling is necessary for an accurate prediction of the shear stresses on the wall as well as von Mises stresses in the wall. FSI simulations are performed to demonstrate the utility of variable wall thickness modeling., Competing Interests: Declaration of Competing Interest The authors declare no potential conflict of interest., (Copyright © 2024 IPEM. Published by Elsevier Ltd. All rights reserved.)

Background and Objective: Understanding the factors that influence the rupture of aneurysms is of primary concern to the clinicians, who are grappled with patient management. It is important to know how the relation between morphological features of the cerebral aneurysm, and the mechanical stresses on the containing arterial walls are influenced by the hemodynamic forces. Present study investigates three different shapes, which have been identified correspondingly in patient-specific scenarios as well. The primary objective is to categorize the bifurcation aneurysms into standard shapes such as, spherical, beehive and pear-shaped, based on patient-specific clinical studies and further compare and contrast the model aneurysms with the patient specific configurations, for their hemodynamic factors as well as the attendant stresses on the wall. MethodsComputational fluid dynamic simulations are performed accounting for the fluid-structure interaction (FSI) effects between the flowing fluid and the containing vessel wall. Blood is assumed to be Newtonian, while the arterial walls are assumed to be linearly elastic. A commercial solver is used for performing detailed calculations. Hemodynamic and bio-mechanical rupture predictions are carried out for the three different shapes. Observations derived from the idealised simulations are compared and contrasted against their patient-specific counterparts. ResultsFrom detailed numerical simulations, it was observed that pear-shaped aneurysms exhibit large re-circulation bubble and flow stagnation zone, with higher residence time for the particles, which may lead to atherosclerotic lesions. Beehive shape allows for maximum flow into the aneurysmal sac with concentrated jet impinging on the dome, leading to high values of maximum WSS (MWSS) resulting in great propensity to form a secondary bleb. However, flow field inside a spherical aneurysm is found to be stable with fewer vortices, and nearly uniform distribution of wall stresses are observed though-out the sac, which perhaps signifies hemodynamically and bio-mechanically stable condition. ConclusionCategorizing patient-specific intracranial aneurysms into standard shapes viz, spherical, beehive and pear could generalize the process of prediction of hemodynamic and bio-mechanical rupture indicators. Comparative assessment of the flow field and stresses reported from the simulations on idealised models, with corresponding patient-specific simulations reveal that, these studies could aid in understanding the generalised shape dependence of hemodynamic and bio-mechanical behaviour of aneurysms., Competing Interests: Declaration of Competing Interest We are glad to submit our manuscript entitled “Hemodynamics and bio-mechanics of morphologically distinct saccular intracranial aneurysms at bifurcations: Idealised vs Patient-specific geometries” for consideration by the editorial team of Computer Methods and Programs in Bio-medicine. This manuscript has neither been published nor it is under consideration for publication anywhere else. Furthermore, on behalf of all the authors, we would like to declare that, there are no known sources of conflict of interest., (Copyright © 2022 Elsevier B.V. All rights reserved.)