Dimensional Images Research Articles

Optimizing Groundwater Exploration Strategies in Huarong, China with 2-Dimensional Resistivity Imaging

Optimizing Groundwater Exploration Strategies in Huarong, China with 2-Dimensional Resistivity Imaging

A Concise and Accurate Method for Evaluating Alveolar Cleft ReconstructionDiet al Concise and Accurate Method for Evaluating.

Currently, there remain unresolved issues in the treatment of alveolar clefts, the resolution of which could greatly benefit many patients with cleft lip and palate. In alveolar cleft treatment research, a reliable tool for pre- and postoperative assessment is crucial. This study presents a concise and accurate method for postoperative evaluations of alveolar treatment, which can rapidly and accurately obtain the shape and volume of the newly formed bone. This study included helical computed tomography (CT) datasets of 20 patients who underwent alveolar bone grafting at our institute. Two observers independently measured the volume of newly formed bone in the patient's CT images one year postoperatively, with each observer performing the measurement twice. To acquire the volume of the newly formed bone at 1 year postoperatively, the model of the newly formed bone must be constructed first. The acquisition of this model involves Boolean operations on registered preoperative and postoperative cranial 3-dimensional (3D) images. The registration of the preoperative and postoperative models is performed in MIMICS software, and the registration results can be directly confirmed layer by layer on the CT images to ensure accuracy. The mean newly formed bone ratio in this study was 39.81%±17.96%, and the mean processing time was 05:11±01:41 minutes. The intraclass correlation coefficient for bone volume measurements between the two observers was 0.999, indicating high consistency and reproducibility. This method enhances accuracy, is time-efficient, and demonstrates high reliability in evaluating postoperative bone formation.

Intramyocardial Sprouting Tip Cells Specify Coronary Arterialization.

The elaborate patterning of coronary arteries critically supports the high metabolic activity of the beating heart. How coronary endothelial cells coordinate hierarchical vascular remodeling and achieve arteriovenous specification remains largely unknown. Understanding the molecular and cellular cues that pattern coronary arteries is crucial to develop innovative therapeutic strategies that restore functional perfusion within the ischemic heart. Single-cell transcriptomics and histological validation were used to delineate heterogeneous transcriptional states of the developing and mature coronary endothelium with a focus on sprouting endothelium and arterial cell specification. Genetic lineage tracing and high-resolution 3-dimensional imaging were used to characterize the origin and mechanisms of coronary angiogenic sprouting, as well as to fate-map selective endothelial lineages. Integration of single-cell transcriptomic data from ischemic adult mouse hearts and human embryonic data served to assess the conservation of transcriptional states across development, disease, and species. We discover that coronary arteries originate from cells that have previously transitioned through a specific tip cell phenotype. We identify nonoverlapping intramyocardial and subepicardial tip cell populations with differential gene expression profiles and regulatory pathways. Esm1-lineage tracing confirmed that intramyocardial tip cells selectively contribute to coronary arteries and endocardial tunnels, but not veins. Notably, prearterial cells are detected from development stages to adulthood, increasingly in response to ischemic injury, and in human embryos, suggesting that tip cell-to-artery specification is a conserved mechanism. A tip cell-to-artery specification mechanism drives arterialization of the intramyocardial plexus and endocardial tunnels throughout life and is reactivated upon ischemic injury. Differential sprouting programs govern the formation and specification of the venous and arterial coronary plexus.

The Intra-articular Tendinous Graft Diameter of 10-mm Bone-Patellar Tendon-Bone Autografts in Adolescent Patients: An MRI Analysis of 100 Patients.

Bone-patellar tendon-bone (BPTB) autograft size may be one modifiable predictor of anterior cruciate ligament (ACL) reconstruction postoperative success, as smaller graft diameter has been associated with higher rates of rupture requiring revision. However, measuring the true intra-articular tendinous graft diameter of the soft tissue portion of a BPTB graft with standard intraoperative methods is difficult while keeping the graft intact. The purpose of the study was to use 3-dimensional magnetic resonance imaging (MRI) measurements to determine the cross-sectional area of the soft tissue, tendinous portion of a standard BPTB autograft with 10-mm diameter bone plugs, and, by calculation, the collagen graft size (ie, graft diameter), as would typically be reported in ACL reconstruction studies that consider soft tissue graft size. It was hypothesized that the calculated collagen graft diameter of 10-mm BPTB autografts would be significantly smaller than 10 mm. Cross-sectional study; Level of evidence, 3. A total of 100 patients (10 girls and 10 boys at each age from 13 to 17 years) who underwent a knee MRI at a single academic orthopaedic center without documented extensor mechanism pathology were identified. The central 10-mm width of the patellar tendon that would be harvested for BPTB autograft was measured. The region of interest area tool was then used to measure the cross-sectional area of a 10-mm BPTB graft with subsequent soft tissue autograft diameter calculation. The mean calculated tendinous graft diameter of a 10 mm-wide BPTB graft was 6.3 ± 0.5 mm and was significantly smaller than a 10-mm reference (P≤ .001). There was no significant association between age and cross-sectional area or graft diameter. Modern 3-dimensional imaging-based measurement techniques demonstrated that the true intra-articular tendinous soft tissue portion of 10-mm BPTB autografts shows substantial variation and is significantly smaller in diameter than the tunnels typically reamed to accommodate the bone plug portions of these grafts. Moreover, as graft size is a predictor of rupture rate, preoperative MRI-based evaluation may be an important tool when considering BPTB autograft for ACL reconstruction. Future comparative clinical research utilizing graft size as a study variable should consider quantifying and utilizing the diameter of the soft tissue component of BPTB autografts.

Arrangement into layers and mechanobiology of multi-cell co-culture models of the uterine wall.

Can a co-culture of three cell types mimic the in vivo layers of the uterine wall? Three protocols tested for co-culture of endometrial epithelial cells (EEC), endometrial stromal cells (ESC), and myometrial smooth muscle cells (MSMC) led to formation of the distinct layers that are characteristic of the structure of the uterine wall in vivo. We previously showed that a layer-by-layer co-culture of EEC and MSMC responded to peristaltic wall shear stresses (WSS) by increasing the polymerization of F-actin in both layers. Other studies showed that WSS induced significant cellular alterations in epithelial and endothelial cells. Human EEC and ESC cell lines and primary MSMC were co-cultured on a collagen-coated synthetic membrane in custom-designed wells. The co-culture model, created by seeding a mixture of all cells at once, was exposed to steady WSS of 0.5 dyne/cm2 for 10 and 30 min. The co-culture of the three different cells was seeded either layer-by-layer or as a mixture of all cells at once. Validation of the models was by specific immunofluorescence staining and confocal microscopy. Alterations of the cytoskeletal F-actin in response to WSS were analyzed from the 2-dimensional confocal images through the Z-stacks following a previously published algorithm. We generated three multi-cell in vitro models of the uterine wall with distinct layers of EEC, ESC, and MSMC that mimic the in vivo morphology. Exposure of the mixed seeding model to WSS induced increased polymerization of F-actin in all the three layers relative to the unexposed controls. Moreover, the increased polymerization of F-actin was higher (P-value < 0.05) when the length of exposure was increased from 10 to 30 min. Furthermore, the inner layers of ESC and MSMC, which are not in direct contact with the applied shearing fluid, also increased their F-actin polymerization. N/A. The mixed seeding co-culture model was exposed to steady WSS of one magnitude, whereas the uterus is a dynamic organ with intra-uterine peristaltic fluid motions that vary in vivo with different time-dependent magnitude. Further in vitro studies may explore the response to peristaltic WSS or other physical and/or hormonal perturbations that may mimic the spectrum of pathophysiological aspects. Numerous in vitro models were developed in order to mimic the human endometrium and endometrium-myometrium interface (EMI) region. The present co-culture models seem to be the first constructed from EEC, ESC, and MSMC on a collagen-coated synthetic membrane. These multi-cell in vitro models better represent the complex in vivo anatomy of the EMI region. The mixed seeding multi-cell in vitro model may easily be implemented in controlled studies of uterine function in reproduction and the pathogenesis of diseases. This study was supported in part by Tel Aviv University funds. All authors declare no conflict of interest.

Characterization of similar Marshak waves observed at the LMJ

We detail results of two experiments performed at the Laser Mégajoule (LMJ) facility aimed at studying similar supersonic Marshak waves propagating in a low-density SiO2 aerogel enclosed in metallic tubes. Similar means here that these two experiments, driven by the same input radiation temperature history, use purposely very different tubes in terms of length (L = 1200 or 2000 μm), diameter (2R = 1000 or 2000 μm), nature of the wall (gold or copper), and aerogel densities (ρ = 30 or 20 mg/cm3), yet the transit time and the radiation temperature of the fronts at the tube exit are the same for both shots. Marshak waves are characterized at the exit using simultaneously for the first time to our knowledge, a one dimensional soft x-ray imager from which the radiation front transit time and curvature are measured and also a broadband x-ray spectrometer to infer its temperature history. These constraining results are then successfully compared to those from simple analytical models [Cohen et al., Phys. Rev. Res. 2, 023007 (2020) and Hurricane et al., Phys. Plasmas 13, 113303 (2006)] and from the three dimensional Lagrangian radiation-hydrodynamics code TROLL to get information on x-ray energy losses. Controlled compensation effects between the length, diameter, and nature of the tubes (governing these losses) are such that the radiation temperature drop along the tubes is eventually the same for these two similar shots.

Deep Learning–Based Prediction of Hepatic Decompensation in Patients With Primary Sclerosing Cholangitis With Computed Tomography

ObjectiveTo investigate a deep learning model for predicting hepatic decompensation using computed tomography (CT) imaging in patients with primary sclerosing cholangitis (PSC). Patients and MethodsRetrospective cohort study involving 277 adult patients with large-duct PSC who underwent an abdominal CT scan. The portal venous phase CT images were used as input to a 3D-DenseNet121 model, which was trained using 5-fold crossvalidation to classify hepatic decompensation. To further investigate the role of each anatomic region in the model’s decision-making process, we trained the model on different sections of 3-dimensional CT images. This included training on the right, left, anterior, posterior, inferior, and superior halves of the image data set. For each half, as well as for the entire scan, we performed area under the receiving operating curve (AUROC) analysis. ResultsHepatic decompensation occurred in 128 individuals after a median (interquartile range) of 1.5 years (142-1318 days) after the CT scan. The deep learning model exhibited promising results, with a mean ± SD AUROC of 0.89±0.04 for the baseline model. The mean ± SD AUROC for left, right, anterior, posterior, superior, and inferior halves were 0.83±0.03, 0.83±0.03, 0.82±0.09, 0.79±0.02, 0.78±0.02, and 0.76±0.04, respectively. ConclusionThe study illustrates the potential of examining CT imaging using 3D-DenseNet121 deep learning model to predict hepatic decompensation in patients with PSC.

An Analytical Study on the Utility of RGB and Multispectral Imagery with Band Selection for Automated Tumor Grading.

The implementation of tumor grading tasks with image processing and machine learning techniques has progressed immensely over the past several years. Multispectral imaging enabled us to capture the sample as a set of image bands corresponding to different wavelengths in the visible and infrared spectrums. The higher dimensional image data can be well exploited to deliver a range of discriminative features to support the tumor grading application. This paper compares the classification accuracy of RGB and multispectral images, using a case study on colorectal tumor grading with the QU-Al Ahli Dataset (dataset I). Rotation-invariant local phase quantization (LPQ) features with an SVM classifier resulted in 80% accuracy for the RGB images compared to 86% accuracy with the multispectral images in dataset I. However, the higher dimensionality elevates the processing time. We propose a band-selection strategy using mutual information between image bands. This process eliminates redundant bands and increases classification accuracy. The results show that our band-selection method provides better results than normal RGB and multispectral methods. The band-selection algorithm was also tested on another colorectal tumor dataset, the Texas University Dataset (dataset II), to further validate the results. The proposed method demonstrates an accuracy of more than 94% with 10 bands, compared to using the whole set of 16 multispectral bands. Our research emphasizes the advantages of multispectral imaging over the RGB imaging approach and proposes a band-selection method to address the higher computational demands of multispectral imaging.

Application of 3-dimensional visualization and image fusion technology in liver cancer with portal vein tumor thrombus surgery.

Liver cancer with portal vein tumor thrombus (PVTT) is a frequent finding and is related to poor prognosis. Surgical resection provides a more promising prognosis in selected patients. The purpose of this study was to explore the application of 3D (3-dimensional) visualization and image fusion technology in liver cancer with PVTT surgery. 12 patients were treated with surgery between March 2019 and August 2022. The preoperative standard liver volume (SLV), estimated future liver remnant (FLR), FLR/SLV, 3D visualization models, PVTT classification, operation programs, surgical results, and prognosis were collected and analyzed. Twelve patients who had complete data of 3D visualization and underwent hemihepatectomy combined with portal vein tumor thrombectomy. The operation plan was formulated by 3D visualization and was highly consistent with the actual surgery. The SLV was 1208.33 ± 63.22 mL, FLR was 734.00 mL and FLR/SLV was 61.62 ± 19.38%. The accuracy of classification of PVTT by 3D visualization was 100%, Cheng type Ⅱa (4 cases), Ⅱb (2 cases), Ⅲa (4 cases), and Ⅲb (2 cases). The 3D visualization model was a perfect fusion with the intraoperative live scene and precise guidance for hepatectomy. No patient was suffering from postoperative liver failure and without procedure‑associated death. 6 patients died of tumor recurrence, and 2 patients died of other reasons. The 12-month cumulative survival rate was 25.9%. 3D visualization and image fusion technology could be used for precise assessment of FLR, classification of PVTT, surgery navigation, and which was helpful in improving the safety of hepatectomy.

Optical coherence tomography and its recent applications for three-dimensional imaging of organoids

An organoid is a self-organizing, miniaturized model of an organ that recapitulates the architecture and physiology of the tissue of origin. With their high resemblance to the structure and function of real organs, organoids have shown enormous potential for studying organ development and growth, as well as modeling diseases in vitro. In organoid studies, optical coherence tomography (OCT), a coherence-based imaging technology, has garnered significant attention as an emerging tool for 3-dimensional organoid imaging. This modality enables the direct observation and monitoring of organoid structure and intercellular events at micrometer-scale resolution and millimeter-scale imaging depth, encompassing the entire body of the organoid. In this paper, we introduce the applications of OCT to recent organoid research, outlining several studies on OCT for the structural and functional imaging of different types of organoids. This review aims to inform bioengineers involved in biomedicine about OCT technologies that could be beneficial for their organoid research.

Typical Development of the Secondary Ossification Centers of the Acetabulum and Their Effects on Acetabular Coverage of the Femoral Head.

We investigated the normal development of the secondary ossification centers of the acetabulum, focusing on their location and the amount of acetabular coverage increased by them. We enrolled 132 patients who were 7 to 16 years of age and had no pelvic deformity but did have ≥1 os ischium, os ilium, and/or os pubis on abdominal or pelvic computed tomographic (CT) scans. The locations of the ossification centers were evaluated by adopting an orientation using 0° for the superior acetabulum, 90° for the anterior acetabulum, 180° for the inferior acetabulum, and 270° for the posterior acetabulum, on a reconstructed 3-dimensional (3D) CT image. The acetabular coverage increase by the os ischium, os ilium, or os pubis was defined as the difference in the posterior acetabular sector angle (ΔPASA), posterosuperior acetabular sector angle (ΔPSASA), superior acetabular sector angle (ΔSASA), anterosuperior acetabular sector angle (ΔASASA), or anterior acetabular sector angle (ΔAASA) measured with and without each secondary ossification center. Patients were grouped into 3 age ranges: late childhood, preadolescence, and early adolescence. The location of each ossification center and the increase in acetabular coverage were compared between these groups. In the late-childhood group, the median start-to-end positions in right hips were 269° to 316° for the os ischium, 345° to 356° for the os ilium, and 81° to 99° for the os pubis. These positions tended to be wider in the early-adolescence group at 252° to 328° for the os ischium (p < 0.001), 338° to 39° for the os ilium (p = 0.005), and 73° to 107° for the os pubis (p = 0.049) in right hips. In right hips in the late-childhood group, the median values were 8.1° for ΔPASA, 14.0° for ΔPSASA, 9.9° for ΔSASA, 11.1° for ΔASASA, and 3.9° for ΔAASA; and in the early-adolescence group, the median values in right hips were 10.7° for ΔPASA, 12.9° for ΔPSASA, 8.4° for ΔSASA, 7.4° for ΔASASA, and 5.6° for ΔAASA. Only the median ΔPASA was larger in the early-adolescence group than in the late-childhood group (p = 0.026). Similar results were observed in left hips. In early adolescence, the secondary ossification centers appeared at more extended areas along the acetabular rim, and the increase in acetabular coverage by the secondary ossification centers tended to be larger in the posterior area but not in the anterior or superior area. Diagnostic Level III. See Instructions for Authors for a complete description of levels of evidence.

4-Dimensional Right Ventricular Outflow Tract Segmentation to Evaluate Candidacy for Transcatheter Pulmonary Valve Replacement.

Candidacy for transcatheter pulmonary valve replacement utilizing newer valve platforms is aided by industry-driven fit analyses. We report augmentation of this decision-making process by visualization of 4-dimensional cine imaging of patient-specific right ventricular outflow tract anatomy and a virtual valve model within a virtual reality platform.

Three-dimensional imaging of upper tract urothelial carcinoma improves diagnostic yield and accuracy.

Upper tract urothelial carcinoma (UTUC) is a rare form of urothelial cancer with a high incidence of recurrence and a low survival rate. Almost two-thirds of UTUCs are invasive at the time of diagnosis; therefore, improving diagnostic methods is key to increasing survival rates. Histopathological analysis of UTUC is essential for diagnosis and typically requires endoscopy biopsy, tissue sectioning, and labeling. However, endoscopy biopsies are minute, and it is challenging to cut into thin sections for conventional histopathology; this complicates diagnosis. Here, we used volumetric 3-dimensional (3D) imaging to explore the inner landscape of clinical UTUC biopsies, without sectioning, revealing that 3D analysis of phosphorylated ribosomal protein S6 (pS6) could predict tumor grade and prognosis with improved accuracy. By visualizing the tumor vasculature, we discovered that pS6+ cells were localized near blood vessels at significantly higher levels in high-grade tumors than in low-grade tumors. Furthermore, the clustering of pS6+ cells was associated with shorter relapse-free survival. Our results demonstrate that 3D volume imaging of the structural niches of pS6 cells deep inside the UTUC samples improved diagnostic yield, grading, and prognosis prediction.

Quantum inspired approach for denoising with application to medical imaging.

Background noise in many fields such as medical imaging poses significant challenges for accurate diagnosis, prompting the development of denoising algorithms. Traditional methodologies, however, often struggle to address the complexities of noisy environments in high dimensional imaging systems. This paper introduces a novel quantum-inspired approach for image denoising, drawing upon principles of quantum and condensed matter physics. Our approach views medical images as amorphous structures akin to those found in condensed matter physics and we propose an algorithm that incorporates the concept of mode resolved localization directly into the denoising process. Notably, our approach eliminates the need for hyperparameter tuning. The proposed method is a standalone algorithm with minimal manual intervention, demonstrating its potential to use quantum-based techniques in classical signal denoising. Through numerical validation, we showcase the effectiveness of our approach in addressing noise-related challenges in imaging and especially medical imaging, underscoring its relevance for possible quantum computing applications.

Enhanced Semantic Visual Cryptography with AI-driven error reduction for improved two-dimensional image quality and security

Abstract Visual Cryptography (VC) has emerged as a vital technique in the information security domain, with the fundamental purpose of securing 2-Dimensional (2D) image content through encryption and facilitating secure communication. While traditional VC has been instrumental in safeguarding data, it often falls short in maintaining image quality and semantic accuracy upon reconstruction. To address these limitations, this research encompasses the development of an Enhanced Semantic VC (ESVC) model, which aims to refine the encryption process while ensuring the semantic integrity of the images. The ESVC model introduces a new approach that merges VC with artificial intelligence (AI) to enhance 2D image encryption and decryption. The novel aspect of this research lies in the integration of AI-driven reinforcement learning (RL) to increase the quality of the 2D image by measuring the errors between the original secret image and the reconstructed image. This innovative framework is tailored for the secure transmission of 2D grayscale images, ensuring the preservation of semantic integrity while measuring and minimizing image quality loss. By integrating RL algorithms with a measurement of error reduction protocol, the model promises robust encryption capabilities with enhanced resilience against a plethora of cyber threats, thereby elevating the standard for secure image communication. Empirical evaluation of the ESVC model yields promising results, with the Peak Signal-to-Noise Ratio (PSNR) of reconstructed images achieving impressive values between +39 and +42 decibels (dB). These findings underscore the ESVC model’s capability to produce high-fidelity decrypted images, significantly surpassing traditional VC methods in both security and image quality. The research findings illuminate the potential of merging AI with VC to achieve a harmonious balance between computational efficiency and encryption strength, marking a significant advancement in the domain of visual data protection.

Preoperative three-dimensional simulation of the breast appearance after wide local excision or level one oncoplastic techniques for breast-conserving treatment does not set unrealistic expectations for aesthetic outcome: One-year follow-up of a randomised controlled trial

IntroductionSimulation of aesthetic outcomes of wide local excision and level one oncoplastic breast conserving treatment (BCT) using 3-dimensional surface imaging (3D-SI) prepares women for their aesthetic outcome. It remains unknown whether women’s memory of this information at the one-year follow-up matches their perception of reality or affects the quality of life. MethodsWith ethical approval, a prospective 3-arm RCT was conducted and it included 3D-simulation, viewing post-operative 2D photographs of other women and standard care. At one-year post-surgery, the participants completed a visual analogue scale (VAS) for the question “How well do you think the information about how your breasts are likely to look after surgery reflects how they actually look today?” and the BCT BREAST-Q module. The Kruskal–Wallis test was used to examine between-group differences at a 5% significance level. ResultsFrom 2017 to 2019, 117 women completed the primary endpoint of being informed about the aesthetic outcome via verbal description, photographs or simulation. Seventy-eight (74%) of the 106 women who remained eligible attended the one-year follow-up. The standardised preoperative 3D-SI simulation did not affect the patient’s perception of the aesthetic outcome compared to standard care or viewing 2D photographs as measured using the VAS (p = 0.40) or BREAST-Q scores for satisfaction with information (p = 0.76), satisfaction with breasts (p = 0.70), and psychosocial wellbeing domains (p = 0.81). DiscussionViewing their own 3D-SI standardised simulation did not significantly affect how the participants perceived their aesthetic outcome. In addition, it did not alter the patient-reported satisfaction. These results demonstrated that simulation for wide local excision or level one oncoplastic surgery does not set unrealistic expectations of the aesthetic outcome when used in a preoperative setting. SynopsisThe use of a non-bespoke three-dimensional simulation of the aesthetic outcome for breast conserving treatment in the preoperative setting does not over-inflate expectations compared to standard care.

Superselective Angiography of Vasa Vasorum Within Partially Thrombosed Vertebral Aneurysm: A Case Report.

Partially thrombosed vertebral artery aneurysms (PTVAs) are rare, most of which are not easy to treat. Furthermore, endovascular treatment of PTVAs may not have favorable outcomes. The relationship between PTVAs and well-developed vasa vasorum (VV), including the mechanism of aneurysm growth, has been reported, but there are no reports of imaging findings by digital subtraction angiography (DSA). In this case, we successfully performed superselective angiography of well-developed VV and evaluated its imaging characteristics. We present the first DSA report of a well-developed VV of PTVA. A 54-year-old patient presented with a PTVA that exerted a mass effect on the medulla oblongata. The aneurysm had no cavity due to thrombosis. The 3-dimensional DSA images indicated VV. Superselective angiography of the VV indicated staining of the thrombosed aneurysm and draining into the suboccipital cavernous sinus through the venous VV. Thus, VV embolization with n-butyl cyanoacrylate was performed. After 3 months, the contrast effect of the aneurysm on contrast-enhanced magnetic resonance imaging disappeared and aneurysm shrinkage was observed. We successfully identified a VV within PTVA. Superselective VV angiography showed staining of the thrombosed component and venous return draining into the suboccipital cavernous sinus. In this case, the embolization of the VV proved to be an effective endovascular treatment of PTVA, but the safety of this method is a challenge. Further case studies are required to validate this method, and we hope it will evolve into a new treatment of PTVA.

Dynamic Image Reconstruction with Motion Priors in Application to Three Dimensional Magnetic Particle Imaging

Dynamic Image Reconstruction with Motion Priors in Application to Three Dimensional Magnetic Particle Imaging

A method to estimate prey densityfrom single-camera images: A case study with chinstrap penguins and Antarctic krill.

Estimating the densities of marine prey observed in animal-borne video loggers when encountered by foraging predators represents an important challenge for understanding predator-prey interactions in the marine environment. We used video images collected during the foraging trip of one chinstrap penguin (Pygoscelis antarcticus) from Cape Shirreff, Livingston Island, Antarctica to develop a novel approach for estimating the density of Antarctic krill (Euphausia superba) encountered during foraging activities. Using the open-source Video and Image Analytics for a Marine Environment (VIAME), we trained a neural network model to identify video frames containing krill. Our image classifier has an overall accuracy of 73%, with a positive predictive value of 83% for prediction of frames containing krill. We then developed a method to estimate the volume of water imaged, thus the density (N·m-3) of krill, in the 2-dimensional images. The method is based on the maximum range from the camera where krill remain visibly resolvable and assumes that mean krill length is known, and that the distribution of orientation angles of krill is uniform. From 1,932 images identified as containing krill, we manually identified a subset of 124 images from across the video record that contained resolvable and unresolvable krill necessary to estimate the resolvable range and imaged volume for the video sensor. Krill swarm density encountered by the penguins ranged from 2 to 307 krill·m-3 and mean density of krill was 48 krill·m-3 (sd = 61 krill·m-3). Mean krill biomass density was 25 g·m-3. Our frame-level image classifier model and krill density estimation method provide a new approach to efficiently process video-logger data and estimate krill density from 2D imagery, providing key information on prey aggregations that may affect predator foraging performance. The approach should be directly applicable to other marine predators feeding on aggregations of prey.

AMIKOMNET: Novel Structure for a Deep Learning Model to Enhance COVID-19 Classification Task Performance

Since early 2020, coronavirus has spread extensively throughout the globe. It was first detected in Wuhan, a province in China. Many researchers have proposed various models to solve problems related to COVID-19 detection. As traditional medical approaches take a lot of time to detect the virus and require specific laboratory tests, the adoption of artificial intelligence (AI), including machine learning, might play an important role in handling the problem. A great deal of research has seen the adoption of AI succeed in the early detection of COVID-19 using X-ray images. Unfortunately, the majority of deep learning adoption for COVID-19 detection has the shortcomings of high error detection and high computation costs. In this study, we employed a hybrid model using an auto-encoder (AE) and a convolutional neural network (CNN) (named AMIKOMNET) with a small number of layers and parameters. We implemented an ensemble learning mechanism in the AMIKOMNET model using Adaboost with the aim of reducing error detection in COVID-19 classification tasks. The experimental results for the binary class show that our model achieved high effectiveness, with 96.90% accuracy, 95.06% recall, 94.67% F1-score, and 96.03% precision. The experimental result for the multiclass achieved 95.13% accuracy, 94.93% recall, 95.75% F1-score, and 96.19% precision. The adoption of Adaboost in AMIKOMNET for the binary class increased the effectiveness of the model to 98.45% accuracy, 96.16% recall, 95.70% F1-score, and 96.87% precision. The adoption of Adaboost in AMIKOMNET in the multiclass classification task also saw an increase in performance, with an accuracy of 96.65%, a recall of 94.93%, an F1-score of 95.76%, and a precision of 96.19%. The implementation of AE to handle image feature extraction combined with a CNN used to handle dimensional image feature reduction achieved outstanding performance when compared to previous work using a deep learning platform. Exploiting Adaboost also increased the effectiveness of the AMIKOMNET model in detecting COVID-19.