Shark detection and classification with machine learning

Suitable shark conservation depends on well-informed population assessments. Direct methods such as scientific surveys and fisheries monitoring are adequate for defining population statuses, but species-specific indices of abundance and distribution coming from these sources are rare for most shark species. We can rapidly fill these information gaps by boosting media-based remote monitoring efforts with machine learning and automation.We created a database of 53,345 shark images covering 219 species of sharks, and packaged object-detection and image classification models into a Shark Detector bundle. The Shark Detector recognizes and classifies sharks from videos and images using transfer learning and convolutional neural networks (CNNs). We applied these models to common data-generation approaches of sharks: collecting occurrence records from photographs taken by the public or citizen scientists, processing baited remote camera footage and online videos, and data-mining Instagram. We examined the accuracy of each model and tested genus and species prediction correctness as a result of training data quantity.The Shark Detector can classify 47 species pertaining to 26 genera. It sorted heterogeneous datasets of images sourced from Instagram with 91% accuracy and classified species with 70% accuracy. It located sharks in baited remote footage and YouTube videos with 89% accuracy, and classified located subjects to the species level with 69% accuracy. All data-generation methods were processed without manual interaction.As media-based remote monitoring appears to dominate methods for observing sharks in nature, we developed an open-source Shark Detector to facilitate common identification applications. Prediction accuracy of the software pipeline increases as more images are added to the training dataset. We provide public access to the software on our GitHub page.     

Pain management using photobiomodulation: Mechanisms,location, and repeatability quantified by pain threshold and neural biomarkers in mice

Photobiomodulation (PBM) is a simple, efficient and cost‐effective treatment for both acute and chronic pain. We previously showed that PBM applied to the mouse head inhibited nociception in the foot. Nevertheless, the optimum parameters, location for irradiation, duration of the effect and the mechanisms of action remain unclear. In the present study, the pain threshold in the right hind paw of mice was studied, after PBM (810 nm CW laser, spot size 1 or 6 cm², 1.2–36 J/cm²) applied to various anatomical locations. The pain threshold, measured with von Frey filaments, was increased more than 3‐fold by PBM to the lower back (dorsal root ganglion, DRG), as well as to other neural structures along the pathway such as the head, neck and ipsilateral (right) paw. On the other hand, application of PBM to the contralateral (left) paw, abdomen and tail had no effect. The optimal effect occurred 2 to 3 hours post‐PBM and disappeared by 24 hours. Seven daily irradiations showed no development of tolerance. Type 1 metabotropic glutamate receptors decreased, and prostatic acid phosphatase and tubulin‐positive varicosities were increased as shown by immunofluorescence of DRG samples. These findings elucidate the mechanisms of PBM for pain and provide insights for clinical practice.      

Contour integration and cortical processing.

Our understanding of visual processing in general, and contour integration in particular, has undergone great change over the last 10 years. There is now an accumulation of psychophysical and neurophysiological evidence that the outputs of cells with conjoint orientation preference and spatial position are integrated in the process of explication of rudimentary contours. Recent neuroanatomical and neurophysiological results suggest that this process takes place at the cortical level V1. The code for contour integration may be a temporal one in that it may only manifest itself in the latter part of the spike train as a result of feedback and lateral interactions. Here we review some of the properties of contour integration from a psychophysical perspective and we speculate on their underlying neurophysiological substrate.     

Energy model for contrast detection: spatiotemporal characteristics of threshold vision

A model for contrast detection of spatiotemporal stimuli is proposed which consists of a spatiotemporal linear filter, an energy device and a threshold device. Assuming the existence of independent intrinsic noise, the probability of stimulus detection was approximated by a Weibull function of the response energy. With this assumption, the stimulus energy is a constant at fixed detection probability. This energy model for contrast detection satisfactorily accounted for the elliptical threshold contours of line pairs at stimulus separations within the range 2–30 min and at stimulus onset asynchronies within the range 20–140 ms. The threshold contour at a large stimulus onset asynchrony (300 ms) was in the form of a rounded square. This finding was explained by assuming that the probability of seeing the line pair was determined by the joint probability that at least one stimulus had been detected. With the energy model, the temporal and spatial autocorrelation functions of the response to a flashed line were evaluated. The autocorrelation functions thus determined were used to predict the temporal contrast sensitivity function to a flickering line stimulus and the spatial contrast sensitivity function to flashed gratings, which were in agreement with the experimental data. The data obtained were fitted adequately by an impulse response approximated by a spatiotemporal Gabor-like function. Received: 08 December 1997 / Accepted in revised form: 26 January 1999     

Ventilatory threshold: measurement and variation with age

The purpose of this study is to present measurement of ventilatory threshold (VeT) and maximal oxygen uptake (VO2max) in a large group of predominantly older subjects using a bicycle ergometer and an automated measuring system. One hundred and twenty-seven healthy elderly subjects (mean age: 68) and 44 young and middle-aged subjects (mean age: 39) underwent a maximal exercise test with breath-by-breath measurement of ventilation and gas exchange variables. Ventilatory threshold was determined by visual inspection of the breakpoints in the VE/VO2 and PETO2 data curves. Additional measures were made in a subset of subjects to determine the reproducibility and interobserver variability of VeT and the relationship between VeT and the venous lactate threshold (LaT). Day-to-day reproducibility of VeT was good with a mean difference in VO2 at VeT on two occasions of 40.23 +/- 125 ml/min. Interobserver variability was low (intraclass correlation coefficient of r = 0.941) and VeT was found to correlate to LaT (r = 0.79, P less than 0.05) with LaT occurring a mean 2.3 min after VeT. VeT declined significantly with age in both males and females but less rapidly than VO2max. Both VO2max and VeT were found to vary with age, sex, height, and weight in a stepwise multiple-linear regression analysis. Age-associated changes in skeletal muscle composition may be in part responsible for the less precipitous decline in VeT with age compared with VO2max.     

Area summation in human vision at and above detection threshold

The initial image-processing stages of visual cortex are well suited to a local (patchwise) analysis of the viewed scene. But the world's structures extend over space as textures and surfaces, suggesting the need for spatial integration. Most models of contrast vision fall shy of this process because (i) the weak area summation at detection threshold is attributed to probability summation (PS) and (ii) there is little or no advantage of area well above threshold. Both of these views are challenged here. First, it is shown that results at threshold are consistent with linear summation of contrast following retinal inhomogeneity, spatial filtering, nonlinear contrast transduction and multiple sources of additive Gaussian noise. We suggest that the suprathreshold loss of the area advantage in previous studies is due to a concomitant increase in suppression from the pedestal. To overcome this confound, a novel stimulus class is designed where: (i) the observer operates on a constant retinal area, (ii) the target area is controlled within this summation field, and (iii) the pedestal is fixed in size. Using this arrangement, substantial summation is found along the entire masking function, including the region of facilitation. Our analysis shows that PS and uncertainty cannot account for the results, and that suprathreshold summation of contrast extends over at least seven target cycles of grating.     

Vitellogenin dynamics during egg-laying: daily variation, repeatability and relationship with egg size

Daily variation in circulating levels of the avian yolk precursor, vitellogenin (VTG), throughout the laying cycle was investigated in female zebra finches Taeniopygia guttata and compared with predicted ovarian follicle demand (based on a model of follicular development for this species). In general, the pattern of variation in plasma VTG matched the predicted demand from the developing ovarian follicle hierarchy. Plasma VTG was non-detectable in non-breeders, but increased rapidly with onset of yolk development, remaining high (1.43–1.82 μg/ml, zinc) through to the 3-egg stage. Plasma levels then declined at the 5-egg stage (to 0.78±0.32 μg/ml) and were undetectable at clutch completion. This result is consistent with the hypothesis that yolk precursor production is costly and that selection has matched supply and demand. While inter-individual variation in plasma VTG was marked (e.g. 0.47–4.26 μg/ml at the 1-egg stage), it also exhibited high intra-individual repeatability (r=0.87–0.93). Finally, we examined the relationship between plasma VTG and primary reproductive effort. While individual variation in plasma VTG was independent of clutch size, laying interval and laying rate, there was a complex, diet-dependent relationship between VTG and egg size, with low plasma VTG levels being associated with both very small (<0.90 g) and very large (>1.15 g) egg sizes.     

Protein complex detection with semi-supervised learning in protein interaction networks

Background

Protein-protein interactions (PPIs) play fundamental roles in nearly all biological processes. The systematic analysis of PPI networks can enable a great understanding of cellular organization, processes and function. In this paper, we investigate the problem of protein complex detection from noisy protein interaction data, i.e., finding the subsets of proteins that are closely coupled via protein interactions. However, protein complexes are likely to overlap and the interaction data are very noisy. It is a great challenge to effectively analyze the massive data for biologically meaningful protein complex detection.

Results

Many people try to solve the problem by using the traditional unsupervised graph clustering methods. Here, we stand from a different point of view, redefining the properties and features for protein complexes and designing a “semi-supervised” method to analyze the problem. In this paper, we utilize the neural network with the “semi-supervised” mechanism to detect the protein complexes. By retraining the neural network model recursively, we could find the optimized parameters for the model, in such a way we can successfully detect the protein complexes. The comparison results show that our algorithm could identify protein complexes that are missed by other methods. We also have shown that our method achieve better precision and recall rates for the identified protein complexes than other existing methods. In addition, the framework we proposed is easy to be extended in the future.

Conclusions

Using a weighted network to represent the protein interaction network is more appropriate than using a traditional unweighted network. In addition, integrating biological features and topological features to represent protein complexes is more meaningful than using dense subgraphs. Last, the “semi-supervised” learning model is a promising model to detect protein complexes with more biological and topological features available.

     

Neural modeling with dynamically adjustable threshold and refractory period.

A variant of the FitzHugh-Nagumo model is proposed in order to fully make use of the computational properties of intraneuronal dynamics. The mechanisms of threshold and refractory periods resulting from the double dynamical processes are qualitatively studied through computer simulation. The results show that the variant neuron model has the property that its threshold, refractory period and response amplitude are dynamically adjustable. This paper has also discussed some problems relating to collective property, learning and implementation of the neural network based on the neuron model proposed. It is noted that the implicit way to describe threshold and refractory period is advantageous to adaptive learning in neural networks and that molecular electronics probably provides an effective approach to implementing the above neuron model.     

Contour integration and segmentation with self-organized lateral connections

Contour integration in low-level vision is believed to occur based on lateral interaction between neurons with similar orientation tuning. How such interactions could arise in the brain has been an open question. Our model suggests that the interactions can be learned through input-driven self-organization, i.e., through the same mechanism that underlies many other developmental and functional phenomena in the visual cortex. The model also shows how synchronized firing mediated by these lateral connections can represent the percept of a contour, resulting in performance similar to that of human contour integration. The model further demonstrates that contour integration performance can differ in different parts of the visual field, depending on what kinds of input distributions they receive during development. The model thus grounds an important perceptual phenomenon onto detailed neural mechanisms so that various structural and functional properties can be measured and predictions can be made to guide future experiments.     

Accurate detection of dairy cow mastitis with deep learning technology: a new and comprehensive detection method based on infrared thermal images

Mastitis is one of the most common diseases in dairy cows and has a negative impact on their welfare and life, causing significant economic losses to the dairy industry. Many attempts have been made to develop a detection method for mastitis using thermal infrared thermography. However, the use of this detection technique to determine the health of the cow's udder is susceptible to external factors, resulting in inaccurate detection of dairy cow mastitis. Therefore, this study explored a new and comprehensive detection method of dairy cow mastitis based on infrared thermal images. This method combined the left and right udder skin surface temperature (USST) difference detection method with the ocular surface temperature and USST difference detection method with improvements. The effect of external factors on dairy cow USST was effectively reduced. In addition, after comparing different target localisation algorithms, this paper used the You Only Look Once v5 (YOLOv5) deep learning network model to obtain the temperature information of eyes and udders, and mastitis detection of dairy cows was performed. A total of 105 dairy cows passing through a passage were randomly selected from the thermal infrared video and detected by the new and comprehensive detection method, and the results of cow mastitis detection were compared with somatic cell count. The results showed that the accuracy, specificity, and sensitivity of mastitis detection were 87.62, 84.62, and 96.30%, respectively. Using the YOLOv5 deep learning network model to locate the key parts of the cow had a good effect, with an average accuracy of 96.1%, and an average frame rate of 116.3f/s. The detection accuracy of dairy cow mastitis by deep learning technology combined with the detection method in this paper reached 85.71%. The results showed that the new and comprehensive detection method based on infrared thermal images can be used for the detection of dairy cow mastitis with high detection accuracy. This method can reduce the influence of external factors and can be integrated into the automatic identification system of dairy mastitis based on YOLOv5 to realise on-site monitoring of dairy mastitis.     

The repeatability of fluctuating asymmetry: a revision and extension

Fluctuating asymmetry (FA) is an imperfect measure of the developmental stability of an individual. Estimates of the heritability of FA will underestimate the heritability of developmental stability; the extent of this bias can be estimated and corrected by estimating the repeatability of developmental stability. This note corrects an error in a previous derivation of this repeatability using the absolute value of the difference between sides as the measure of FA, and derives the repeatability of developmental stability for another common measure of FA, the variance among sides within an individual.     

Meniscal lesion detection and characterization in adult knee MRI: A deep learning model approach with external validation

PurposeEvaluation of a deep learning approach for the detection of meniscal tears and their characterization (presence/absence of migrated meniscal fragment).MethodsA large annotated adult knee MRI database was built combining medical expertise of radiologists and data scientists’ tools. Coronal and sagittal proton density fat suppressed-weighted images of 11,353 knee MRI examinations (10,401 individual patients) paired with their standardized structured reports were retrospectively collected. After database curation, deep learning models were trained and validated on a subset of 8058 examinations. Algorithm performance was evaluated on a test set of 299 examinations reviewed by 5 musculoskeletal specialists and compared to general radiologists’ reports. External validation was performed using the publicly available MRNet database. Receiver Operating Characteristic (ROC) curves results and Area Under the Curve (AUC) values were obtained on internal and external databases.ResultsA combined architecture of meniscal localization and lesion classification 3D convolutional neural networks reached AUC values of 0.93 (95% CI 0.82, 0.95) for medial and 0.84 (95% CI 0.78, 0.89) for lateral meniscal tear detection, and 0.91 (95% CI 0.87, 0.94) for medial and 0.95 (95% CI 0.92, 0.97) for lateral meniscal tear migration detection. External validation of the combined medial and lateral meniscal tear detection models resulted in an AUC of 0.83 (95% CI 0.75, 0.90) without further training and 0.89 (95% CI 0.82, 0.95) with fine tuning.ConclusionOur deep learning algorithm demonstrated high performance in knee menisci lesion detection and characterization, validated on an external database.     

V-BANet: Land cover change detection using effective deep learning technique

Urban development alters landscapes, frequently degrading environmental services and quality of life. High-resolution remote sensing images provide a chance to detect subtle changes in land cover and can capture the features of a ground object. However, traditional approaches usually experience difficulties when processing large and quickly expanding datasets, low levels of automation, limited computational efficiency, and inconsistent identification accuracies and standards brought on by inconsistent operators. Conducting change detection in a more accurate, automated, and standardized manner has become crucial and increasingly difficult due to the quick collection of remote sensing data. Therefore, in this paper, V-Net and Bilateral Attention Network (V-BANet) based deep learning is implemented to segment the landscapes and extract the features from the images. Initially, the bi-temporal images are segmented using V-Net to independently identify the objects in each image. Then spatial and channel attention blocks are employed in Bilateral Attention Network to learn more discriminative features from the images. Finally, the features' relationships are discovered by contrasting the original feature map in one image with the updated feature map in the other. Objective and subjective experiments are performed on a public bi-temporal high-resolution ONERA Satellite Change Detection (OSCD) dataset and the LEVIR-CD dataset. Moreover, the proposed approach reached an accuracy of 99.29% and IoU of 98.31% with the OSCD Dataset and 99.42% accuracy and 98.83% IoU with the LEVIR-CD Dataset. The experimental outcomes with each specified dataset demonstrated that the suggested methodology outperformed several state-of-the-art techniques and produced superior results.     

Local dynamic stability of treadmill walking: Intrasession and week-to-week repeatability

Repetitive falls degrade the quality of life of elderly people and of patients suffering of various neurological disorders. In order to prevent falls while walking, one should rely on relevant early indicators of impaired dynamic balance. The local dynamic stability (LDS) represents the sensitivity of gait to small perturbations: divergence exponents (maximal Lyapunov exponents) assess how fast a dynamical system diverges from neighbor points. Although numerous findings attest the validity of LDS as a fall risk index, reliability results are still sparse. The present study explores the intrasession and intersession repeatability of gait LDS using intraclass correlation coefficients (ICC) and standard error of measurement (SEM). Ninety-five healthy individuals performed 5 min treadmill walking in two sessions separated by 9 days. Trunk acceleration was measured with a 3D accelerometer. Three time scales were used to estimate LDS: over 4–10 strides (λ_4–10), over one stride (λ₁) and over one step (λ_0.5). The intrasession repeatability was assessed from three repetitions of either 35 strides or 70 strides taken within the 5 min tests. The intersession repeatability compared the two sessions, which totalized 210 strides. The intrasession ICCs (70-strides estimates/35-strides estimates) were 0.52/0.18 for λ_4–10 and 0.84/0.77 for λ₁ and λ_0.5. The intersession ICCs were around 0.60. The SEM results revealed that λ_0.5 measured in medio-lateral direction exhibited the best reliability, sufficient to detect moderate changes at individual level (20%). However, due to the low intersession repeatability, one should average several measurements taken on different days in order to better approximate the true LDS.