Visual Estimation Research Articles

Comparative study on real-time pose estimation of vision-based unmanned underwater vehicles

Background: Navigation and localization are key to the successful execution of autonomous unmanned underwater vehicles (UUVs) in marine environmental monitoring, underwater 3D mapping, and ocean resource surveys. The estimation of the position and the orientation of autonomous UUVs are a long-standing challenging and fundamental problem. As one of the underwater sensors, camera has always been the focus of attention due to its advantages of low cost and rich content information in low visibility waters, especially in the fields of visual perception of the underwater environment, target recognition and tracking. At present, the visual real-time pose estimation technology that can be used for UUVs is mainly divided into geometry-based visual positioning algorithms and deep learning-based visual positioning algorithms. Methods: In order to compare the performance of different positioning algorithms and strategies, this paper uses C++ and python, takes the ORB-SLAM3 algorithm and DF-VO algorithm as representatives to conduct a comparative experiment and analysis. Results: The geometry-based algorithm ORB-SLAM3 is less affected by illumination, performs more stably in different underwater environments, and has a shorter calculation time, but its robustness is poor in complex environments. The visual positioning algorithm DF-VO based on deep learning takes longer time to compute, and the positioning accuracy is more easily affected by illumination, especially in dark conditions. However, its robustness is better in unstructured environments such as large-scale image rotation and dynamic object interference. Conclusions: In general, the deep learning-based algorithm is more robust, but multiple deep learning networks make it need more time to compute. The geometry-based method costs less time and is more accurate in low-light and turbid underwater conditions. However, in real underwater situations, these two methods can be connected as binocular vision or methods of multi-sensor combined pose estimation.

The ability of contemporary cardiologists to judge the ischemic impact of a coronary lesion visually

BackgroundLandmark trials showed that invasive pressure measurement (Fractional Flow Reserve, FFR) was a better guide to coronary stenting than visual assessment. However, present-day interventionists have benefited from extensive research and personal experience of mapping anatomy to hemodynamics. AimsTo determine if visual assessment of the angiogram performs as well as invasive measurement of coronary physiology. Methods25 interventional cardiologists independently visually assessed the single vessel coronary disease of 200 randomized participants in The Objective Randomized Blinded Investigation with optimal medical Therapy of Angioplasty in stable angina trial (ORBITA). They gave a visual prediction of the FFR and Instantaneous Wave-free Ratio (iFR), denoted vFFR and viFR respectively. Each judged each lesion on 2 occasions, so that every lesion had 50 vFFR, and 50 viFR assessments. The group consensus visual estimates (vFFR-group and viFR-group) and individual cardiologists' visual estimates (vFFR-individual and viFR-individual) were tested alongside invasively measured FFR and iFR for their ability to predict the placebo-controlled reduction in stress echo ischemia with stenting. ResultsPlacebo-controlled ischemia improvement with stenting was predicted by vFFR-group (p < 0.0001) and viFR-group (p < 0.0001), vFFR-individual (p < 0.0001) and viFR-individual (p < 0.0001). There were no significant differences between the predictive performance of the group visual estimates and their invasive counterparts: p = 0.53 for vFFR vs FFR and p = 0.56 for viFR vs iFR. ConclusionVisual assessment of the angiogram by contemporary experts, provides significant additional information on the amount of ischaemia which can be relieved by placebo-controlled stenting in single vessel coronary artery disease.

Vision-Aided Localization and Mapping in Forested Environments Using Stereo Images.

Forests are traditionally characterized by stand-level descriptors, such as basal area, mean diameter, and stem density. In recent years, there has been a growing interest in enhancing the resolution of forest inventory to examine the spatial structure and patterns of trees across landscapes. The spatial arrangement of individual trees is closely linked to various non-monetary forest aspects, including water quality, wildlife habitat, and aesthetics. Additionally, associating individual tree positions with dendrometric variables like diameter, taper, and species can provide data for highly optimized, site-specific silvicultural prescriptions designed to achieve diverse management objectives. Aerial photogrammetry has proven effective for mapping individual trees; however, its utility is limited due to the inability to directly estimate many dendrometric variables. In contrast, terrestrial mapping methods can directly observe essential individual tree characteristics, such as diameter, but their mapping accuracy is governed by the accuracy of the global satellite navigation system (GNSS) receiver and the density of the canopy obstructions between the receiver and the satellite constellation. In this paper, we introduce an integrated approach that combines a camera-based motion and tree detection system with GNSS positioning, yielding a stem map with twice the accuracy of using a consumer-grade GNSS receiver alone. We demonstrate that large-scale stem maps can be generated in real time, achieving a root mean squared position error of 2.16 m. We offer an in-depth explanation of a visual egomotion estimation algorithm designed to enhance the local consistency of GNSS-based positioning. Additionally, we present a least squares minimization technique for concurrently optimizing the pose track and the positions of individual tree stem[s].

Corroborating otolith age using oxygen isotopes and comparing outcomes to scale age: Consequences for estimation of growth and reference points in northern pike (Esox lucius)

AbstractAccurate age estimates are crucial for assessing the life‐histories of fish and providing management advice, but validation studies are rare for many species. We corroborated age estimates with annual cycles of oxygen isotopes (δ18O) in otoliths of 86 northern pike (Esox lucius) from the southern Baltic Sea, compared results with visual age estimates from scales and otoliths, and assessed bias introduced by different age‐estimation structures on von Bertalanffy growth models and age‐structured population models. Age estimates from otoliths were accurate, while age estimates from scales significantly underestimated the age of pike older than 6 years compared to the corroborated reference age. Asymptotic length () was larger, and the growth coefficient was lower for scale ages than for corroborated age and otolith age estimates. Consequentially, scale‐informed population models overestimated maximum sustainable yield (), biomass at (), relative frequency of trophy fish (), and optimal minimum length limit but underestimated fishing mortality at (). Using scale‐based ages to inform management regulations for pike may therefore result in conservative management and lost yield. The overestimated asymptotic length may instill unrealistic expectations of trophy potential in recreational anglers targeting large pike, while the overestimation in MSY would cause unrealistic expectations of yield potential in commercial fishers.

Dense fog simulation in southern Brazil using the WRF model with high spatial resolution

Fog is an atmospheric phenomenon that reduces horizontal visibility to 1000 m or less. These phenomena affect different human activities, mainly those associated with air, maritime, and road transport. Rio Grande city, located in southern Brazil, has one of the main Brazilian ports, and fog occurrences at this location cause several inconveniences, such as the interruption of port activities and increases in operating costs. Thus, a better understanding of fog formation and dissipation, and the consequent improvement in its forecasts, can bring significant help to the port activities in that region. Therefore, this work seeks to evaluate the WRF model’s ability to simulate fog events in Rio Grande city. Four study cases of dense fogs were performed for this goal, as well as a sensitivity test to select the best vertical resolution configuration of the model. Meteorological data from the Rio Grande pilotage station and the Brazilian Coastal Monitoring System buoys were analyzed. Satellite images and synoptic charts were also analyzed for the study cases. The numerical simulations were performed using the WRF model with three nested domains, where 1 km was the highest horizontal resolution. The Fog Stability Index (FSI) and two estimates of horizontal visibility were evaluated. Simulations show promising results, highlighting the ability of the model to identify the fog occurrence in the area of interest and represent aspects of its dissipation process. Furthermore, it is noteworthy that the FSI and the evaluated visibility estimates were sensitive to the occurrence of the events.

BEENE: deep learning-based nonlinear embedding improves batch effect estimation.

Analyzing large-scale single-cell transcriptomic datasets generated using different technologies is challenging due to the presence of batch-specific systematic variations known as batch effects. Since biological and technological differences are often interspersed, detecting and accounting for batch effects in RNA-seq datasets are critical for effective data integration and interpretation. Low-dimensional embeddings, such as principal component analysis (PCA) are widely used in visual inspection and estimation of batch effects. Linear dimensionality reduction methods like PCA are effective in assessing the presence of batch effects, especially when batch effects exhibit linear patterns. However, batch effects are inherently complex and existing linear dimensionality reduction methods could be inadequate and imprecise in the presence of sophisticated nonlinear batch effects. We present Batch Effect Estimation using Nonlinear Embedding (BEENE), a deep nonlinear auto-encoder network which is specially tailored to generate an alternative lower dimensional embedding suitable for both linear and nonlinear batch effects. BEENE simultaneously learns the batch and biological variables from RNA-seq data, resulting in an embedding that is more robust and sensitive than PCA embedding in terms of detecting and quantifying batch effects. BEENE was assessed on a collection of carefully controlled simulated datasets as well as biological datasets, including two technical replicates of mouse embryogenesis cells, peripheral blood mononuclear cells from three largely different experiments and five studies of pancreatic islet cells. BEENE is freely available as an open source project at https://github.com/ashiq24/BEENE.

Be careful what you explain: Benefits and costs of explainable AI in a simulated medical task

We investigated the impact of explainability instructions with respect to system limitations on trust behavior and trust attitude when using an artificial intelligence (AI) support agent to perform a simulated medical task. In an online experiment (N = 128), participants performed a visual estimation task in a simulated medical setting (i.e., estimate the percentage of bacteria in a visual stimulus). All participants were supported by an AI that gave perfect recommendations for all but one color of bacteria (i.e., error-prone color with 50% reliability). We manipulated between-subjects whether participants knew about the error-prone color (XAI condition) or not (nonXAI condition). The analyses revealed that participants showed higher trust behavior (i.e., lower deviation from the AI recommendation) for the non-error-prone trials in the XAI condition. Moreover, participants showed lower trust behavior for the error-prone color in the XAI condition than in the nonXAI condition. However, this behavioral adaptation only applied to the subset of error-prone trials in which the AI gave correct recommendations, and not to the actual erroneous trials. Thus, designing explainable AI systems can also come with inadequate behavioral adaptations, as explainability was associated with benefits (i.e., more adequate behavior in non-error-prone trials), but also costs (stronger changes to the AI recommendations in correct error-prone trials).

CORRECTION OF OPTOACOUSTIC IMAGES USING BANACH'S THEOREM

The main aim of the work was the development and research of a numerical algorithm designed to correct artifacts and distortions that occur during image reconstruction in optical-acoustic tomography tasks. The task was to develop an algorithm capable of compensating for the peculiarities of the reconstruction method. The proposed iterative scheme for enhancing optoacoustic images is based on Banach's fixed point theorem. The problem of eliminating distortions and artifacts in optoacoustic images has been modeled and tested for four iterative schemes. Two-dimensional and three-dimensional numerical models were used to investigate the efficiency of the proposed algorithm. The optical-acoustic experiment simulated a biological environment with an object to be reconstructed built into it. The linear (2D case) or planar (3D case) detecting acoustic arrays located on the surface of the studied samples were considered. The reconstruction quality was determined using both the quantitative and visual estimates of the obtained results. The efficiency of the iterative algorithm of enhancing the reconstruction quality was quantitatively estimated using the structure similarity index SSIM and relative reconstruction error E. It was shown that the developed algorithm makes it possible to considerably improve the image quality as compared with the traditional optical-acoustic tomography methods even with a small number of iterations. Algorithms of the k-Wave Matlab toolbox software package were used for comparing. The results obtained in the work can be important from the standpoint of the prospects of their further practical application in the tasks of biomedical visualization.

A new green approach for Lavandula stoechas aroma recovery and stabilization coupling supercritical CO2 and natural deep eutectic solvents

This work investigated a green approach to obtain and stabilize Lavandula stoechas L. volatile organic compounds with sensory aroma characteristics by using alternative solvents, namely supercritical carbon dioxide (scCO2) and deep eutectic solvents (DES). The CO2 extracts were dispersed in different DES mixtures (betaine:ethylene glycol (1:3), betaine:glycerol (1:2), and glycerol:glucose (4:1)) and their stability was monitored during 6 months of storage at room temperature by monitoring the headspace (HS) profile. The CO2 extract was used as the control. It was initially determined that there was a dominant presence of oxygenated monoterpenes (67.33–77.50%) in the extracts. During storage, significant changes occurred in the samples’ HS, such as the decrease in terpene hydrocarbons which also affected the presence of oxygenated terpenes, which increased in certain cases. Moreover, the highest formation of new components was recorded in the control which could be an indicator of decreased stability. The DESs-CO2 were more stable than the CO2 control and among them, betaine:ethylene glycol stood out as the most adequate systems for maintaining the stability of L. stoechas HS components. For the visual estimation of similarities and dissimilarities among the samples, chemometric pattern recognition approaches were applied including the hierarchical cluster analysis, principal component analysis, and sum of ranking differences.

Analysis of Food Cost and Food Perception with Food Waste of The Supplemental Food for School Children in Elementary School Children at Kulon Progo District

Background: The success of the Supplemental Food for School Children program is determined by the child's acceptance of the food provide. Objective: to analyze appropriate costs and perceptions of food with additional food leftovers for elementary school children in Kulon Progo Regency. Methods: The research design is a cross-sectional design. This research was conducted on 89 elementary school children from grade IV to grade VI in 4 elementary schools that received PMT AS in Kulon Progo District with simple random. The research instrument used a food cost calculation form, a food perception questionnaire with a smiley face, a food weighing device, a comstock scale form with a 6-point visual estimation method. Analysis of bivariate using chi square test and multivariate using logistic regression test.Results: The bivariate analysis results showed a significant relationship between the cost of food and the perception of The Supplemental Food for School Children with p = 0.002. There is a significant relationship between food perception and food waste of The Supplemental Food for School Children with p = 0.000. Finally, there is a significant relationship between food costs and food waste with p = 0.007. The results of multivariate analysis, food costs and perceptions contributed 29% to reducing food waste. Conclusion: Appropriate food costs and perceptions of preferred the food are associated with the least leftover food waste for school children. Key words : Food Cost, Food Perception, Food Waste

Remote visual estimation of shoulder range of motion has generally high interobserver reliability but limited accuracy

Surgeon visual estimation of shoulder range of motion (ROM) is commonplace in the outpatient office setting and routinely reported in clinical research, but the reliability and accuracy of this practice remain unclear. The purpose of this study is to establish the reliability and accuracy of remote visual estimation of shoulder ROM in healthy volunteers and symptomatic patients among a large group of shoulder surgeons. Our hypothesis is that remote visual estimation would be reliable and accurate compared with the digital goniometer method. Fifty shoulder surgeon members of the PacWest Shoulder and Elbow Society independently determined the active shoulder forward flexion (FF), internal rotation at 90° abduction (IR90), external rotation at 90° abduction, external rotation at the side , and maximal spinal level reached with internal rotation (IRspine) through visual estimation of video recordings taken from 10 healthy volunteers and 10 symptomatic patients. Variations in measurements were quantified using the interobserver reliability through calculation of the intraclass correlation coefficient. Accuracy was determined through comparison with digital goniometer measurements obtained with an on-screen protractor application using Bland-Altman mean differences and 95% limits of agreement. The interobserver reliability among examiners showed moderate to excellent correlation, with intraclass correlation coefficient ranging from 0.768 to 0.928 for the healthy volunteers and 0.739 to 0.878 for the symptomatic patients. Accuracy was limited, with upper limits of agreement exceeding the established minimal clinically important differences (MCIDs) for FF (20° vs. MCID of 14°) and IR90 (25° vs. 18°) in the healthy volunteers and for FF (33° vs. 16°), external rotation at 90° abduction (21° vs. 18°), and IR90 (31° vs. 20°) in the symptomatic patients. Despite generally high intersurgeon reliability in the visual estimation of shoulder ROM, there was questionable accuracy when compared to digital goniometer measurements,with measurement errors often exceeding established MCID values. Given the potential implications for the clinical response to treatment and the significance of research findings, the adoption of validated instruments to measure ROM and the standardization of examination procedures should be considered.

Sustainable and healthy food consumption patterns in a multicultural university cafeteria by plate waste visual estimation

ABSTRACT Consumers play a decisive role in setting trends in the market, where healthier interventions shall be implemented toward sustainable lifestyles. University cafeterias are important in the students’ food consumption. They should focus on the quality of the dish, engaging students in the process of selecting and accepting the menu avoiding food waste and changing consumption and production patterns. This study evaluated consumption behaviors and acceptability of three lunch menus based on the Mediterranean Diet in a student cafeteria catering using the plate waste visual estimation method, participants also completed a consumer behavior questionnaire. 71 students from 12 Latin American countries participated in the study. More than half of the participants (56.3%) confirmed that they repeat lunchtime meal options, with the highest frequency at once or twice a week. Menus were positively accepted with a data mean of between 4.58 to 5.49, indicating that they consumed more than 50% of the dishes served, with a plate waste of less than 12.5%. These findings show the importance of the enhancement of sustainable consumption patterns using sensory evaluation through low levels of food waste, with the use of sustainable local products helping in meal preparation.

The Ability ofMilitary Critical Care Air Transport Members to Visually Estimate Percent Systolic Pressure Variation.

Inappropriate fluid management during patient transport may lead to casualty morbidity. Percent systolic pressure variation (%SPV) is one of several technologies that perform a dynamic assessment of fluid responsiveness (FT-DYN). Trained anesthesia providers can visually estimate and use %SPV to limit the incidence of erroneous volume management decisions to 1-4%. However, the accuracy of visually estimated %SPV by other specialties is unknown. The aim of this article is to determine the accuracy of estimated %SPV and the incidence of erroneous volume management decisions for Critical Care Air Transport (CCAT) team members before and after training to visually estimate and utilize %SPV. In one sitting, CCAT team providers received didactics defining %SPV and indicators of fluid responsiveness and treatment with %SPV ≤7 and ≥14.5 defining a fluid nonresponsive and responsive patient, respectively; they were then shown ten 45-second training arterial waveforms on a simulated Propaq M portable monitor's screen. Study subjects were asked to visually estimate %SPV for each arterial waveform and queried whether they would treat with a fluid bolus. After each training simulation, they were told the true %SPV. Seven days post-training, the subjects were shown a different set of ten 45-second testing simulations and asked to estimate %SPV and choose to treat, or not. Nonparametric limits of agreement for differences between true and estimated %SPV were analyzed using Bland-Altman graphs. In addition, three errors were defined: (1) %SPV visual estimate errors that would label a volume responsive patient as nonresponsive, or vice versa; (2) incorrect treatment decisions based on estimated %SPV (algorithm application errors); and (3) incorrect treatment decisions based on true %SPV (clinically significant treatment errors). For the training and testing simulations, these error rates were compared between, and within, provider groups. Sixty-one physicians (MDs), 64 registered nurses (RNs), and 53 respiratory technicians (RTs) participated in the study. For testing simulations, the incidence and 95% CI for %SPV estimate errors with sufficient magnitude to result in a treatment error were 1.4% (0.5%, 3.2%), 1.6% (0.6%, 3.4%), and 4.1% (2.2%, 6.9%) for MDs, RNs, and RTs, respectively. However, clinically significant treatment errors were statistically more common for all provider types, occurring at a rate of 7%, 10%, and 23% (all P < .05). Finally, students did not show clinically relevant reductions in their errors between training and testing simulations. Although most practitioners correctly visually estimated %SPV and all students completed the training in interpreting and applying %SPV, all groups persisted in making clinically significant treatment errors with moderate to high frequency. This suggests that the treatment errors were more often driven by misapplying FT-DYN algorithms rather than by inaccurate visual estimation of %SPV. Furthermore, these errors were not responsive to training, suggesting that a decision-making cognitive aid may improve CCAT teams' ability to apply FT-DYN technologies.

Visual rating and the use of image analysis for assessing canopy density in a pecan provenance collection during leaf fall

A collection representing the native range of pecan was planted at the USDA − ARS Southeastern Fruit and Tree Nut Research Station, Byron, GA. The collection (867 trees) is a valuable genetic resource for characterizing important horticultural traits. Canopy density during leaf fall is important as the seasonal canopy dynamics provides insights to environmental cues and breeding potential of germplasm. The ability of visual raters to estimate canopy density on a subset of the provenance collection (76 trees) as an indicator of leaf shed during autumn along with image analysis values was explored. Mean canopy density using the image analysis software was less compared to visual estimates (11.9% vs 18.4%, respectively). At higher canopy densities, the raters overestimated foliage density, but overall agreement between raters and measured values was good (ρc = 0.849 to 0.915), and inter-rater reliability was high (R2 = 0.910 to 0.953). The provenance from Missouri (MO-L), the northernmost provenance, had the lowest canopy density in November, and results show that the higher the latitude of the provenance, the lower the canopy density. Based on regression, the source provenance latitude explained 0.609 of the variation using image analysis, and 0.551 to 0.640 when based on the rater estimates of canopy density. Visual assessment of pecan canopy density due to late season leaf fall for comparing pecan genotypes provides accurate and reliable estimates and could be used in future studies of the whole provenance collection.

Artificial intelligence-based polyp size measurement in gastrointestinal endoscopy using the auxiliary waterjet as a reference.

Measurement of colorectal polyp size during endoscopy is mainly performed visually. In this work, we propose a novel polyp size measurement system (Poseidon) based on artificial intelligence (AI) using the auxiliary waterjet as a measurement reference. Visual estimation, biopsy forceps-based estimation, and Poseidon were compared using a computed tomography colonography-based silicone model with 28 polyps of defined sizes. Four experienced gastroenterologists estimated polyp sizes visually and with biopsy forceps. Furthermore, the gastroenterologists recorded images of each polyp with the waterjet in proximity for the application of Poseidon. Additionally, Poseidon's measurements of 29 colorectal polyps during routine clinical practice were compared with visual estimates. In the silicone model, visual estimation had the largest median percentage error of 25.1 % (95 %CI 19.1 %-30.4 %), followed by biopsy forceps-based estimation: median 20.0 % (95 %CI 14.4 %-25.6 %). Poseidon gave a significantly lower median percentage error of 7.4 % (95 %CI 5.0 %-9.4 %) compared with other methods. During routine colonoscopies, Poseidon presented a significantly lower median percentage error (7.7 %, 95 %CI 6.1 %-9.3 %) than visual estimation (22.1 %, 95 %CI 15.1 %-26.9 %). In this work, we present a novel AI-based method for measuring colorectal polyp size with significantly higher accuracy than other common sizing methods.

A comprehensive methodology for the visual impact assessment of mines and quarries

Since the 60s, several multidisciplinary studies have been carried out worldwide to investigate the perception of landscape modification and define appropriate methodologies of Visual Impact Assessment (VIA). The criteria typically applied to evaluate the visual impact due to landscape alteration can be categorized into two classes: direct and indirect approaches (i.e. respectively perception test and numerical quantification of landscape modification). Both the approaches intend to overcome the limit of the judgment subjectivity in the evaluation process and its dependence on the observer's specific characteristics. As a matter of fact, the effect of landscape modification needs to be objectively estimated when the VIA procedures are mandated by regulatory policies and accurate evaluation techniques are required to support decision-making. The Lvi indicator (Level of visual impact) has been formerly proposed as a tool to enable the objective quantification of the visual impact magnitude produced by extensive surface excavation (i.e. mines and quarries). This article discusses the integration of the Lvi indicator into a comprehensive Landscape Assessment Model (LAM), which includes the three succeeding steps of the visual impact assessment procedure: 1. the identification of the Key Observation Points (KOPs) (i.e. intervisibility analysis), 2. the quantification of the visual impact magnitude from the selected KOPs (i.e. visual impact estimation), 3. the comparison with predefined levels of acceptable landscape modification (i.e. visual impact evaluation). The proposed assessment procedure has been applied to a quarry of inert materials located in Sardinia (Italy) to highlight and discuss the practical implications of the proposed procedure and its inherent limitations.

Statistical analysis of organelle movement using state-space models

BackgroundOrganelle motility is essential for the correct cellular function of various eukaryotic cells. In plant cells, chloroplasts move towards the intracellular area irradiated by a weak light to maximise photosynthesis. To initiate this process, an unknown signal is transferred from the irradiated area to distant chloroplasts. Quantification of this chloroplast movement has been performed using visual estimations that are analyst-dependent and labour-intensive. Therefore, an objective and faster method is required.ResultsIn this study, we developed the cellssm package of R (https://github.com/hnishio/cellssm.git), which is a user-friendly tool for state-space modelling to statistically analyse the directional movement of cells or organelles. Our method showed a high accuracy in estimating the start time of chloroplast movement in the liverwort Marchantia polymorpha over a short period. The tool indicated that chloroplast movement accelerates during transport to the irradiated area and that signal transfer speed is uneven within a cell. We also developed a method to estimate the common dynamics among multiple chloroplasts in each cell, which clarified different characteristics among cells.ConclusionsWe demonstrated that state-space modelling is a powerful method to understand organelle movement in eukaryotic cells. The cellssm package can be applied to various directional movements (both accumulation and avoidance) at cellular and subcellular levels to estimate the true transition of states behind the time-series data.

The Application of Deep Learning for the Segmentation and Classification of Coronary Arteries.

In recent years, the prevalence of coronary artery disease (CAD) has become one of the leading causes of death around the world. Accurate stenosis detection of coronary arteries is crucial for timely treatment. Cardiologists use visual estimations when reading coronary angiography images to diagnose stenosis. As a result, they face various challenges which include high workloads, long processing times and human error. Computer-aided segmentation and classification of coronary arteries, as to whether stenosis is present or not, significantly reduces the workload of cardiologists and human errors caused by manual processes. Moreover, deep learning techniques have been shown to aid medical experts in diagnosing diseases using biomedical imaging. Thus, this study proposes the use of automatic segmentation of coronary arteries using U-Net, ResUNet-a, UNet++, models and classification using DenseNet201, EfficientNet-B0, Mobilenet-v2, ResNet101 and Xception models. In the case of segmentation, the comparative analysis of the three models has shown that U-Net achieved the highest score with a 0.8467 Dice score and 0.7454 Jaccard Index in comparison with UNet++ and ResUnet-a. Evaluation of the classification model's performances has shown that DenseNet201 performed better than other pretrained models with 0.9000 accuracy, 0.9833 specificity, 0.9556 PPV, 0.7746 Cohen's Kappa and 0.9694 Area Under the Curve (AUC).

Genetic mapping of the Rcs2 locus in soybean cultivar Kent for resistance to frogeye leaf spot

AbstractThe fungal disease frogeye leaf spot (FLS) can reduce yield up to 30% in susceptible soybean [Glycine max (L) Merr] cultivars. The Rcs2 locus in the cultivar Kent provides resistance to all the named Cercospora sojina races currently found in the United States; however, this locus for resistance to FLS has not been molecularly mapped yet. F2:3 and recombinant inbred line (RIL) populations derived from a cross of Forrest × Kent were phenotyped in greenhouse assays with isolate S23 of C. sojina, and reactions were determined using a visual estimate of disease severity. A bulked segregant analysis and linkage mapping with Kompetitive allele‐specific PCR (KASP) markers in the F2:3 population located the Rcs2 locus on chromosome 11. The locus was mapped in the RIL population derived from the same pedigree and refined to a 336 kb interval by evaluating recombination break points in the RILs. This region contains 31 gene models, and Kent has polymorphisms in 11 of these genes and/or promoter regions of the Williams82.a2.v1 genome. The Rcs2 locus and linked markers reported will be useful in marker‐assisted selection for FLS resistance. This research will also guide future efforts to understand the landscape of FLS resistance in United States soybean germplasm and to discover the causal genes for Rcs2 resistance.

Accuracy of Gravimetric Method of Blood Loss Estimation during Open Myomectomy

Background: Estimation of blood loss during surgery is a critical component that may affect patients’ management. Objective: The study evaluated the accuracy of the gravimetric method of blood loss estimation in women with an open myomectomy. Materials and Methods: This was a cross-sectional analytical study that evaluated the accuracy of the gravimetric method of measuring blood loss during open myomectomy for patients with symptomatic fibroid at a Teaching Hospital in Nigeria. The diagnosis of uterine fibroids was based on clinical examination and ultrasound scan. A proforma was used to record information from those who gave written consent. The gravimetric method and visual estimation were compared with the actual blood loss (ABL) as the gold standard. Statistical analysis was done using Stata version 16. Student's t-test was used to compare the mean blood loss estimation of the different methods. The significant level was set at P-value &lt;0.05. Results: A total of 78 women who had open myomectomy participated in the study. The mean ABL was 787.11 ± 63.17 mL. The mean blood loss by the gravimetric method was 755.95 ± 492.72 mL, while the mean estimation error by the gravimetric method was 31.16 ± 263.23 mL. The mean difference between the estimated ABL and the loss estimated by the gravimetric method was 31.18 mL. This was not statistically significant (P-value = 0.30). However, there was a statistically significant difference between the ABL and blood loss by visual method (mean difference = 161.64, P-value = 0.02). Conclusion: The gravimetric estimation of blood loss during surgery correlated with actual estimated blood loss. It should be considered in the routine estimation of blood loss during surgery.