DeepSmile: Anomaly Detection Software for Facial Movement Assessment.
Ontology highlight
ABSTRACT:
SUBMITTER: Rodriguez Martinez EA
PROVIDER: S-EPMC9858579 | biostudies-literature | 2023 Jan
REPOSITORIES: biostudies-literature
Publications
DeepSmile: Anomaly Detection Software for Facial Movement Assessment.
Rodríguez Martínez Eder A EA Polezhaeva Olga O Marcellin Félix F Colin Émilien É Boyaval Lisa L Sarhan François-Régis FR Dakpé Stéphanie S
Diagnostics (Basel, Switzerland) 20230110 2
Facial movements are crucial for human interaction because they provide relevant information on verbal and non-verbal communication and social interactions. From a clinical point of view, the analysis of facial movements is important for diagnosis, follow-up, drug therapy, and surgical treatment. Current methods of assessing facial palsy are either (i) objective but inaccurate, (ii) subjective and, thus, depending on the clinician's level of experience, or (iii) based on static data. To address ...[more]
Similar Datasets
Project description:In this paper, to solve the problem of detecting network anomalies, a method of forming a set of informative features formalizing the normal and anomalous behavior of the system on the basis of evaluating the Hurst (H) parameter of the network traffic has been proposed. Criteria to detect and prevent various types of network anomalies using the Three Sigma Rule and Hurst parameter have been defined. A rescaled range (RS) method to evaluate the Hurst parameter has been chosen. The practical value of the proposed method is conditioned by a set of the following factors: low time spent on calculations, short time required for monitoring, the possibility of self-training, as well as the possibility of observing a wide range of traffic types. For new DPI (Deep Packet Inspection) system implementation, algorithms for analyzing and captured traffic with protocol detection and determining statistical load parameters have been developed. In addition, algorithms that are responsible for flow regulation to ensure the QoS (Quality of Services) based on the conducted static analysis of flows and the proposed method of detection of anomalies using the parameter Hurst have been developed. We compared the proposed software DPI system with the existing SolarWinds Deep Packet Inspection for the possibility of network traffic anomaly detection and prevention. The created software components of the proposed DPI system increase the efficiency of using standard intrusion detection and prevention systems by identifying and taking into account new non-standard factors and dependencies. The use of the developed system in the IoT communication infrastructure will increase the level of information security and significantly reduce the risks of its loss.
| S-EPMC7146318 | biostudies-literature
Project description:Importance:Facial vascular anomalies are surgical challenges due to their vascularity and facial nerve distortion. To assist facial vascular anomaly surgical treatment, presurgical percutaneous facial nerve stimulation and recording of compound motor action potentials can be used to map the facial nerve branches. During surgery, the nerve map and continuous intraoperative motor end plate potential monitoring can be used to reduce nerve injury. Objective:To investigate if preoperative facial nerve mapping (FNM) is associated with intraoperative facial nerve injury risk and safe surgical approach options compared with standard nerve integrity monitoring (NIM). Design, Setting, and Participants:This investigation was a historically controlled study at a tertiary vascular anomaly center in Seattle, Washington. Participants were 92 pediatric patients with facial vascular anomalies undergoing definitive anomaly surgery (from January 1, 1999, through January 1, 2015), with 2 years' follow-up. In retrospective review, a consecutive FNM patient cohort after 2005 (FNM group) was compared with a consecutive historical cohort (1999-2005) (NIM group). Main Outcomes and Measures:Postoperative facial nerve function and selected surgical approach. For NIM and FNM comparisons, statistical analysis calculated odds ratios of nerve injury and operative approach, and time-to-event methods analyzed operative time. Results:The NIM group had 31 patients (median age, 3.3 years [interquartile range, 2.2-11.4 years]; 20 [65%] male), and the FNM group had 61 patients (median age, 4.4 years [interquartile range, 1.5-11.0 years]; 26 [43%] male). In both groups, lymphatic malformation resection was most common (19 of 31 [61%] in the NIM group and 32 of 61 [52%] in the FNM group), and the median anomaly volumes were similar (52.4 mL; interquartile range, 12.8-183.3 mL in the NIM group and 65.4 mL; interquartile range, 18.8-180.2 mL in the FNM group). Weakness in the facial nerve branches at 2 years after surgery was more common in the NIM group (6 of 31 [19%]) compared with the FNM group (1 of 61 [2%]) (percentage difference, 17%; 95% CI, 3%-32%). Anterograde facial nerve dissection was used more in the NIM group (27 of 31 [87%]) compared with the FNM group (28 of 61 [46%]) (percentage difference, 41%; 95% CI, 24%-58%). Treatment with retrograde dissection without identification of the main trunk of the facial nerve was performed in 21 of 61 (34%) in the FNM group compared with 0 of 31 (0%) in the NIM group. Operative time was significantly shorter in the FNM group, and patients in the FNM group were more likely to complete surgery sooner (adjusted hazard ratio, 5.36; 95% CI, 2.00-14.36). Conclusions and Relevance:Facial nerve mapping before facial vascular anomaly surgery was associated with less intraoperative facial nerve injury and shorter operative time. Mapping enabled direct identification of individual intralesional and perilesional nerve branches, reducing the need for traditional anterograde facial nerve dissection, and allowed for safe removal of some lesions after partial nerve dissection through transoral or direct excision.
| S-EPMC5876811 | biostudies-literature
Project description:Measurement of eye movements often helps to diagnose ocular motor disorders in the clinic, and is also used as a research tool in ocular motor, vision and vestibular research. Eye movements, however, are usually recorded without simultaneous video recordings, making offline interpretation difficult. We developed a tool that converts the measured eye movement data into a three-dimensional (3D) movie of eye movements. Having useful functions such as slow-play, pause and exaggeration of the movements, this new software provides a research and teaching tool to aid interpretation of the recorded eye movements.
| S-EPMC3156268 | biostudies-literature
Project description:BackgroundObjective dynamic assessments are central to the evaluation of facial rejuvenation treatments. This study used three-dimensional digital stereophotogrammetry to generate a quantitative dynamic assessment of facial strain and hyaluronic acid filler efficacy.MethodsThirty women (aged 41 to 65 years) with moderate to severe bilateral nasolabial folds and marionette lines received Restylane Refyne, Restylane Defyne, or both, and were compared to a "younger" untreated group (n = 20; aged 25 to 35 years). Three-dimensional videos were collected at baseline and at day 42. Dynamic strain was analyzed at the marionette lines and nasolabial folds as subjects went from a neutral position through a series of facial expressions.ResultsBaseline nasolabial folds and marionette lines showed statistically significant (p ≤ 0.05) differences in the level of stretch between younger and older untreated subjects, with higher stretch profiles observed in the older cohort. In the older cohort, filler treatment reduced peak strain (stretch) in the nasolabial folds and marionette lines compared with baseline across all tested expressions, resembling the youthful strain profile of the younger untreated control group. Treatment was well tolerated.ConclusionsQuantitative dynamic strain analysis is an innovative method for evaluating the dynamic face following dermal fillers and provides objective evidence that such treatment results in stretch levels resembling a younger phenotype in areas prone to the effects of facial aging. The strain reduction following filler treatment objectively conveys a dermal tightening effect, likely secondary to the volumization of treated areas. Additional studies will refine the technology and associated procedures to optimize quantitative dynamic strain analysis for treatment planning and patient outcome optimization.Clinical question/level of evidenceTherapeutic, IV.
| S-EPMC7004449 | biostudies-literature
Project description:Hyperspectral images include information from a wide range of spectral bands deemed valuable for computer vision applications in various domains such as agriculture, surveillance, and reconnaissance. Anomaly detection in hyperspectral images has proven to be a crucial component of change and abnormality identification, enabling improved decision-making across various applications. These abnormalities/anomalies can be detected using background estimation techniques that do not require the prior knowledge of outliers. However, each hyperspectral anomaly detection (HS-AD) algorithm models the background differently. These different assumptions may fail to consider all the background constraints in various scenarios. We have developed a new approach called Greedy Ensemble Anomaly Detection (GE-AD) to address this shortcoming. It includes a greedy search algorithm to systematically determine the suitable base models from HS-AD algorithms and hyperspectral unmixing for the first stage of a stacking ensemble and employs a supervised classifier in the second stage of a stacking ensemble. It helps researchers with limited knowledge of the suitability of the HS-AD algorithms for the application scenarios to select the best methods automatically. Our evaluation shows that the proposed method achieves a higher average F1-macro score with statistical significance compared to the other individual methods used in the ensemble. This is validated on multiple datasets, including the Airport-Beach-Urban (ABU) dataset, the San Diego dataset, the Salinas dataset, the Hydice Urban dataset, and the Arizona dataset. The evaluation using the airport scenes from the ABU dataset shows that GE-AD achieves a 14.97% higher average F1-macro score than our previous method (HUE-AD), at least 17.19% higher than the individual methods used in the ensemble, and at least 28.53% higher than the other state-of-the-art ensemble anomaly detection algorithms. As using the combination of greedy algorithm and stacking ensemble to automatically select suitable base models and associated weights have not been widely explored in hyperspectral anomaly detection, we believe that our work will expand the knowledge in this research area and contribute to the wider application of this approach.
| S-EPMC11204925 | biostudies-literature
Project description:Relative scores such as Local Outlying Factor and mass ratio have been shown to be better scores than global scores in detecting anomalies. While this is true, our analysis reveals for the first time that these relative scores have a key shortcoming: anomalies have greatly different relative scores if they are located in different regions where the curvatures of the density surface are very different. As a result, the low-score anomalies could be ranked lower than some normal points. This revelation motivates (i) a new score called Neighbourhood Contrast (NC) which produces approximately the same high scores for all anomalies, regardless of varying curvatures of the density surface in different regions; and (ii) an anomaly detection method based on NC. Our experiments show that the proposed method which employs the new score significantly outperforms methods using the aforementioned relative scores on benchmark datasets.
| S-EPMC7206296 | biostudies-literature
Project description:ObjectiveMyasthenia gravis (MG) is an autoimmune disease leading to fatigable muscle weakness. Extra-ocular and bulbar muscles are most commonly affected. We aimed to investigate whether facial weakness can be quantified automatically and used for diagnosis and disease monitoring.MethodsIn this cross-sectional study, we analyzed video recordings of 70 MG patients and 69 healthy controls (HC) with two different methods. Facial weakness was first quantified with facial expression recognition software. Subsequently, a deep learning (DL) computer model was trained for the classification of diagnosis and disease severity using multiple cross-validations on videos of 50 patients and 50 controls. Results were validated using unseen videos of 20 MG patients and 19 HC.ResultsExpression of anger (p = 0.026), fear (p = 0.003), and happiness (p < 0.001) was significantly decreased in MG compared to HC. Specific patterns of decreased facial movement were detectable in each emotion. Results of the DL model for diagnosis were as follows: area under the curve (AUC) of the receiver operator curve 0.75 (95% CI 0.65-0.85), sensitivity 0.76, specificity 0.76, and accuracy 76%. For disease severity: AUC 0.75 (95% CI 0.60-0.90), sensitivity 0.93, specificity 0.63, and accuracy 80%. Results of validation, diagnosis: AUC 0.82 (95% CI: 0.67-0.97), sensitivity 1.0, specificity 0.74, and accuracy 87%. For disease severity: AUC 0.88 (95% CI: 0.67-1.0), sensitivity 1.0, specificity 0.86, and accuracy 94%.InterpretationPatterns of facial weakness can be detected with facial recognition software. Second, this study delivers a 'proof of concept' for a DL model that can distinguish MG from HC and classifies disease severity.
| S-EPMC10424649 | biostudies-literature
Project description:The purposes of this study are to propose an unsupervised anomaly detection method based on a deep neural network (DNN) model, which requires only normal images for training, and to evaluate its performance with a large chest radiograph dataset. We used the auto-encoding generative adversarial network (α-GAN) framework, which is a combination of a GAN and a variational autoencoder, as a DNN model. A total of 29,684 frontal chest radiographs from the Radiological Society of North America Pneumonia Detection Challenge dataset were used for this study (16,880 male and 12,804 female patients; average age, 47.0 years). All these images were labeled as "Normal," "No Opacity/Not Normal," or "Opacity" by board-certified radiologists. About 70% (6,853/9,790) of the Normal images were randomly sampled as the training dataset, and the rest were randomly split into the validation and test datasets in a ratio of 1:2 (7,610 and 15,221). Our anomaly detection system could correctly visualize various lesions including a lung mass, cardiomegaly, pleural effusion, bilateral hilar lymphadenopathy, and even dextrocardia. Our system detected the abnormal images with an area under the receiver operating characteristic curve (AUROC) of 0.752. The AUROCs for the abnormal labels Opacity and No Opacity/Not Normal were 0.838 and 0.704, respectively. Our DNN-based unsupervised anomaly detection method could successfully detect various diseases or anomalies in chest radiographs by training with only the normal images.
| S-EPMC8289984 | biostudies-literature
Project description:Diffusion-MRI (dMRI) measures molecular diffusion, which allows to characterize microstructural properties of the human brain. Gliomas strongly alter these microstructural properties. Delineation of brain tumors currently mainly relies on conventional MRI-techniques, which are, however, known to underestimate tumor volumes in diffusely infiltrating glioma. We hypothesized that dMRI is well suited for tumor delineation, and developed two different deep-learning approaches. The first diffusion-anomaly detection architecture is a denoising autoencoder, the second consists of a reconstruction and a discrimination network. Each model was exclusively trained on non-annotated dMRI of healthy subjects, and then applied on glioma patients' data. To validate these models, a state-of-the-art supervised tumor segmentation network was modified to generate groundtruth tumor volumes based on structural MRI. Compared to groundtruth segmentations, a dice score of 0.67 ± 0.2 was obtained. Further inspecting mismatches between diffusion-anomalous regions and groundtruth segmentations revealed, that these colocalized with lesions delineated only later on in structural MRI follow-up data, which were not visible at the initial time of recording. Anomaly-detection methods are suitable for tumor delineation in dMRI acquisitions, and may further enhance brain-imaging analysis by detection of occult tumor infiltration in glioma patients, which could improve prognostication of disease evolution and tumor treatment strategies.
| S-EPMC10663596 | biostudies-literature
Project description:Density estimation is a widely used method for unsupervised anomaly detection. However, the presence of anomalies in training data may severely impact the density estimation process, thereby hampering the use of more sophisticated density estimation methods such as those based on deep neural networks. In this work, we propose RobustRealNVP, a robust deep density estimation framework for unsupervised anomaly detection. Our approach differs from existing flow-based models from two perspectives. First, RobustRealNVP discards data points with low estimated densities during optimization to prevent them from corrupting the density estimation process. Furthermore, it imposes Lipschitz regularization to ensure smoothness in the estimated density function. We demonstrate the robustness of our algorithm against anomalies in training data from both theoretical and empirical perspectives. The results show that our algorithm outperforms state-of-the-art unsupervised anomaly detection methods.
| S-EPMC9615909 | biostudies-literature