Search results for: mask R-CNN

Authors: Sandabad Sara, Sayd Tahri Yassine, Hammouch Ahmed

Abstract:

The development of science has allowed computer scientists to touch the medicine and bring aid to radiologists as we are presenting it in our article. Our work focuses on the detection and localization of tumors areas in the human brain; this will be a completely automatic without any human intervention. In front of the huge volume of MRI to be treated per day, the radiologist can spend hours and hours providing a tremendous effort. This burden has become less heavy with the automation of this step. In this article we present an automatic and effective tumor detection, this work consists of two steps: the first is the image filtering using the filter Nl-means, then applying the expectation maximization algorithm (EM) for retrieving the tumor mask from the brain MRI and extracting the tumor area using the mask obtained from the second step. To prove the effectiveness of this method multiple evaluation criteria will be used, so that we can compare our method to frequently extraction methods used in the literature.

Keywords: MRI, Em algorithm, brain, tumor, Nl-means

Procedia PDF Downloads 296

Authors: Hanaa A. Al-Gaoudi, Hassan Ebeid

Abstract:

The research investigates the materials and processes used in the manufacturing of an Egyptian polychrome cartonnage mummy mask with the aim of dating this object and establishing trade patterns of certain materials that were used and available at the time of ancient Egypt. This anonymous-source object was held in the basement storage of the Egyptian Museum in Cairo (EMC) and has never been on display. Furthermore, there is no information available regarding its owner, provenance, date, and even the time of its possession by the museum. Moreover, the object is in a very poor condition where almost two-thirds of the mask was bent and has never received any previous conservation treatment. This research has utilized well-established multi-analytical methods to identify the considerable diversity of materials that have been used in the manufacturing of this object. These methods include Computed Tomography Scan (CT scan) to acquire detailed pictures of the inside physical structure and condition of the bended layers. Dino-Lite portable digital microscope, scanning electron microscopy with energy dispersive X-ray spectrometer (SEM-EDX), and the non-invasive imaging technique of multispectral imaging (MSI) to obtain information about the physical characteristics and condition of the painted layers and to examine the microstructure of the materials. Portable XRF Spectrometer (PXRF) and X-Ray powder diffraction (XRD) to identify mineral phases and the bulk element composition in the gilded layer, ground, and pigments; Fourier-transform infrared (FTIR) to identify organic compounds and their molecular characterization; accelerator mass spectrometry (AMS 14C) to date the object. Preliminary results suggest that there are no human remains inside the object, and the textile support is linen fibres with tabby weave 1/1 and these fibres are in a very bad condition. Several pigments have been identified, such as Egyptian blue, Magnetite, Egyptian green frit, Hematite, Calcite, and Cinnabar; moreover, the gilded layers are pure gold and the binding media in the pigments is Arabic gum and animal glue in the textile support layer.

Keywords: analytical methods, Egyptian museum, mummy mask, pigments, textile

Procedia PDF Downloads 96

Authors: Behnam Madadnia

Abstract:

For several years, 3D-shaped electronics have been rising, with many uses in home appliances, automotive, and manufacturing. One of the biggest challenges in the fabrication of 3D shape electronics, which are made by thermoforming, is repeatable and accurate component positioning, and typically there is no control over the final position of the component. This paper aims to address this issue and present a reliable approach for guiding the electronic components in the desired place during thermoforming. We have proposed a heat-control mask in the thermoforming machine to control the heating of the polymer, not allowing specific parts to be formable, which can assure the conductive traces' mechanical stability during thermoforming of the substrate. We have verified our approach's accuracy by applying our method on a real industrial semi-sphere mold for positioning 7 LEDs and one touch sensor. We measured the LEDs' position after thermoforming to prove the process's repeatability. The experiment results demonstrate that the proposed method is capable of positioning electronic components in thermoformed 3D electronics with high precision.

Keywords: 3D-shaped electronics, electronic components, thermoforming, component positioning

Procedia PDF Downloads 64

Authors: Siamak Yaghoubi, Mohammad Reza Abootorabi, Hamid Kayalha

Abstract:

Objective: The study was aimed to evaluate the applicability of the Cobra Perilaryngeal Airway (Cobra PLATM) for patients under general anesthesia and also compare result with the Laryngeal Mask Airway (LMA). Methods: Seventy three obese and overweight patients were included in the study. The patients were randomly assigned to either LMA or Cobra PLATM. Time required for intubation, successful intubation attempt, airway sealing pressure, the incidences of complications including blood staining, sore throat and dysphagia were assessed and noted. Results: Thirty six and thirty seven patients were allocated randomly to either LMA or Cobra PLATM, respectively. Most of the patients were male and were in Mallampati Class II airway in both groups. The first attempt and overall insertion success for the Cobra PLATM was significantly more frequent compared to the LMA (p<0.05). Tube insertion was more successful (Cobra PLATM, 94%; LMA™, 77%; P = 0.027) with the Cobra PLATM. The insertion times were similar with the Cobra PLATM and LMA™ (Cobra PLATM, 29.94±16.35s; LMA™, 27.00±7.88s). The airway sealing pressure in the Cobra PLATM (24.80±0.90 H2O) was significantly more than LMA™ (19.13 ±0.58 H2O, p<0.001). Sore throat was more frequent in the LMA™ groups that did not reach statistical significance (Fisher’s exact test, P = 0.33). Incidences of blood staining on airway tube were seen for both groups that was higher in the Cobra PLATM group (Fisher’s exact test, P = 0.02). Incidence of dysphagia was not different between the two groups. Conclusion: The CobraPLA™ was found to be safe and low complications, better airway sealing and high rate of the first insertion success for suing in obese and overweight patients. The study recommended using the CobraPLA™ as a rescue device in an emergency situation among obese and overweight patients.

Keywords: CobraPLA™, flexible laryngeal mask airway, obese patients, perilaryngeal airway

Procedia PDF Downloads 356

Authors: Oladapo Tolulope Ibitoye

Abstract:

Neural networks have become prominent and widely engaged in algorithmic-based machine learning networks. They are perfect in solving day-to-day issues to a certain extent. Neural networks are computing systems with several interconnected nodes. One of the numerous areas of application of neural networks is object detection. This is a prominent area due to the coronavirus disease pandemic and the post-pandemic phases. Wearing a face mask in public slows the spread of the virus, according to experts’ submission. This calls for the development of a reliable and effective model for detecting face masks on people's faces during compliance checks. The existing neural network models for facemask detection are characterized by their black-box nature and large dataset requirement. The highlighted challenges have compromised the performance of the existing models. The proposed model utilized Faster R-CNN Model on Inception V3 backbone to reduce system complexity and dataset requirement. The model was trained and validated with very few datasets and evaluation results shows an overall accuracy of 96% regardless of skin tone.

Keywords: convolutional neural network, face detection, face mask, masked faces

Procedia PDF Downloads 44

Authors: Hyunbin Choi, Jihyeon Noh, Changwon Lim

Abstract:

In this paper, we propose SiamMask++, an architecture that performs layer-wise aggregation and depth-wise cross-correlation and introduce multi-RPN module and multi-MASK module to improve EAO (Expected Average Overlap), a representative performance evaluation metric for Visual Object Tracking (VOT) challenge. The proposed architecture, SiamMask++, has two versions, namely, bi_SiamMask++, which satisfies the real time (56fps) on systems equipped with GPUs (Titan XP), and rf_SiamMask++, which combines mask refinement modules for EAO improvements. Tests are performed on VOT2016, VOT2018 and VOT2019, the representative datasets of Visual Object Tracking tasks labeled as rotated bounding boxes. SiamMask++ perform better than SiamMask on all the three datasets tested. SiamMask++ is achieved performance of 62.6% accuracy, 26.2% robustness and 39.8% EAO, especially on the VOT2018 dataset. Compared to SiamMask, this is an improvement of 4.18%, 37.17%, 23.99%, respectively. In addition, we do an experimental in-depth analysis of how much the introduction of features and multi modules extracted from the backbone affects the performance of our model in the VOT task.

Keywords: visual object tracking, video, deep learning, layer wise aggregation, Siamese network

Procedia PDF Downloads 114

Authors: Eman A. Gani, Loay E. George, Faisel G. Mohammed, Kamal H. Sager

Abstract:

Digital retinal images are more appropriate for automatic screening of diabetic retinopathy systems. Unfortunately, a significant percentage of these images are poor quality that hinders further analysis due to many factors (such as patient movement, inadequate or non-uniform illumination, acquisition angle and retinal pigmentation). The retinal images of poor quality need to be enhanced before the extraction of features and abnormalities. So, the segmentation of retinal image is essential for this purpose, the segmentation is employed to smooth and strengthen image by separating the background and damaged areas from the overall image thus resulting in retinal image enhancement and less processing time. In this paper, methods for segmenting colored retinal image are proposed to improve the quality of retinal image diagnosis. The methods generate two segmentation masks; i.e., background segmentation mask for extracting the background area and poor quality mask for removing the noisy areas from the retinal image. The standard retinal image databases DIARETDB0, DIARETDB1, STARE, DRIVE and some images obtained from ophthalmologists have been used to test the validation of the proposed segmentation technique. Experimental results indicate the introduced methods are effective and can lead to high segmentation accuracy.

Keywords: retinal images, fundus images, diabetic retinopathy, background segmentation, damaged areas segmentation

Procedia PDF Downloads 372

Authors: May M. Youssef

Abstract:

This article introduces a design, for additive manufacturing technology, face shield as Personal Protective Equipment from the respiratory viruses such as coronavirus 2. The face shields help to reduce ocular exposure and play a vital role in diverting away from the respiratory COVID-19 air droplets around the users' face. The proposed face shield comprises three assembled polymer parts. The frame with a transparency overhead projector sheet visor is suitable for frontline health care workers and ordinary citizens. The frame design allows tightening the shield around the user’s head and permits rubber elastic straps to be used if required. That ergonomically designed with a unique face mask support used in case of wearing extra protective mask was created using computer aided design (CAD) software package. The finite element analysis (FEA) structural verification of the proposed design is performed by an advanced simulation technique. Subsequently, the prototype model was fabricated by a 3D printing using Fused Deposition Modeling (FDM) as a globally developed face shield product. This study provides a different face shield designs for global production, which showed to be suitable and effective toward supply chain shortages and frequent needs of personal protective goods during coronavirus disease and similar viruses.

Keywords: additive manufacturing, Coronavirus-19, face shield, personal protective equipment, 3D printing

Procedia PDF Downloads 162

Authors: Haolun Sun, Ping Wang, Mei Wu, Meng Zhang, Bin Hou, Ling Yang, Xiaohua Ma, Yue Hao

Abstract:

Gallium nitride (GaN) is an attractive material for next-generation power devices. It is noted that the performance of GaN-based high electron mobility transistors (HEMTs) is always limited by the self-heating effect. In response to the problem, integrating devices with polycrystalline diamond (PCD) has been demonstrated to be an efficient way to alleviate the self-heating issue of the GaN-based HEMTs. Among all the heat-spreading schemes, using PCD to cap the epitaxial layer before the HEMTs process is one of the most effective schemes. Now, the mainstream method of fabricating the PCD-capped HEMTs is to deposit the diamond heat-spreading layer on the AlGaN surface, which is covered by a thin nucleation dielectric/passivation layer. To achieve the pattern etching of the diamond heat spreader and device preparation, we selected SiN as the hard mask for diamond etching, which was deposited by plasma-enhanced chemical vapor deposition (PECVD). The conventional diamond etching method first uses F-based etching to remove the SiN from the special window region, followed by using O₂/Ar plasma to etch the diamond. However, the results of the scanning electron microscope (SEM) and focused ion beam microscopy (FIB) show that there are lots of diamond pillars on the etched diamond surface. Through our study, we found that it was caused by the high roughness of the diamond surface and the existence of the overlap between the diamond grains, which makes the etching of the SiN hard mask insufficient and leaves micro-masks on the diamond surface. Thus, a cyclic etching method was proposed to solve the problem of the residual SiN, which was left in the F-based etching. We used F-based etching during the first step to remove the SiN hard mask in the specific region; then, the O₂/Ar plasma was introduced to etch the diamond in the corresponding region. These two etching steps were set as one cycle. After the first cycle, we further used cyclic etching to clear the pillars, in which the F-based etching was used to remove the residual SiN, and then the O₂/Ar plasma was used to etch the diamond. Whether to take the next cyclic etching depends on whether there are still SiN micro-masks left. By using this method, we eventually achieved the self-terminated etching of the diamond and the smooth surface after the etching. These results demonstrate that the cyclic etching method can be successfully applied to the integrated preparation of polycrystalline diamond thin films and GaN HEMTs.

Keywords: AlGaN/GaN heterojunction, O₂/Ar plasma, cyclic etching, polycrystalline diamond

Procedia PDF Downloads 82

Authors: M. Dhana Lakshmi, S. Sakthivel Murugan

Abstract:

As light passes from source to observer in the water medium, it is scattered by the suspended particulate matter. This scattering effect will plague the captured images with non-uniform illumination, blurring details, halo artefacts, weak edges, etc. To overcome this, pixel normalization with an Amended Unsharp Mask (AUM) filter is proposed to enhance the degraded image. To validate the robustness of the proposed technique irrespective of atmospheric light, the considered datasets are collected on dual locations. For those images, the maxima and minima pixel intensity value is computed and normalized; then the AUM filter is applied to strengthen the blurred edges. Finally, the enhanced image is obtained with good illumination and contrast. Thus, the proposed technique removes the effect of scattering called de-hazing and restores the perceptual information with enhanced edge detail. Both qualitative and quantitative analyses are done on considering the standard non-reference metric called underwater image sharpness measure (UISM), and underwater image quality measure (UIQM) is used to measure color, sharpness, and contrast for both of the location images. It is observed that the proposed technique has shown overwhelming performance compared to other deep-based enhancement networks and traditional techniques in an adaptive manner.

Keywords: underwater drone imagery, pixel normalization, thresholding, masking, unsharp mask filter

Procedia PDF Downloads 161

Authors: V. Mantey, N. Findlay, I. Maddox

Abstract:

The quality and quantity of global satellite data have been increasing exponentially in recent years as spaceborne systems become more affordable and the sensors themselves become more sophisticated. This is a valuable resource for many applications, including disaster management and relief. However, while more information can be valuable, the volume of data available is impossible to manually examine. Therefore, the question becomes how to extract as much information as possible from the data with limited manpower. Buildings are a key feature of interest in satellite imagery with applications including telecommunications, population models, and disaster relief. Machine learning tools are fast becoming one of the key resources to solve this problem, and models have been developed to detect buildings in optical satellite imagery. However, by and large, most models focus on affluent regions where buildings are generally larger and constructed further apart. This work is focused on the more difficult problem of detection in populated regions. The primary challenge with detecting small buildings in densely populated regions is both the spatial and spectral resolution of the optical sensor. Densely packed buildings with similar construction materials will be difficult to separate due to a similarity in color and because the physical separation between structures is either non-existent or smaller than the spatial resolution. This study finds that training models until they are overfitting the input sample can perform better in these areas than a more robust, generalized model. An overfitted model takes less time to fine-tune from a generalized pre-trained model and requires fewer input data. The model developed for this study has also been fine-tuned using existing, open-source, building vector datasets. This is particularly valuable in the context of disaster relief, where information is required in a very short time span. Leveraging existing datasets means that little to no manpower or time is required to collect data in the region of interest. The training period itself is also shorter for smaller datasets. Requiring less data means that only a few quality areas are necessary, and so any weaknesses or underpopulated regions in the data can be skipped over in favor of areas with higher quality vectors. In this study, a landcover classification model was developed in conjunction with the building detection tool to provide a secondary source to quality check the detected buildings. This has greatly reduced the false positive rate. The proposed methodologies have been implemented and integrated into a configurable production environment and have been employed for a number of large-scale commercial projects, including continent-wide DEM production, where the extracted building footprints are being used to enhance digital elevation models. Overfitted machine learning models are often considered too specific to have any predictive capacity. However, this study demonstrates that, in cases where input data is scarce, overfitted models can be judiciously applied to solve time-sensitive problems.

Keywords: building detection, disaster relief, mask-RCNN, satellite mapping

Procedia PDF Downloads 148

Authors: Jayshree P. Zend, Ashatai B. Pawar

Abstract:

The study was carried out in Parbhani district of Maharashtra state, India with the objectives to study environmental problems faced by flour mill workers, prevalence of work-related health hazards and the physiological cost of workers while performing work in flour mill in traditional method as well as improved method. The use of flour presser, dust controlling bag and noise and dust controlling mask developed by AICRP College of Home Science, VNMKV, Parbhani was considered as an improved method. This investigation consisted survey and experiment which was conducted in the respective locations of flour mills. Healthy, non-smoking 30 flour mill workers ranged between the age group of 20-50 yrs comprising 16 female and 14 male working at flour mill for 4-8 hrs/ day and 6 days/ week and had minimum five years experience of work in flour mill were selected for the study. Pulmonary function test of flour mill workers was carried out by trained technician at Dr. ShankarraoChavan Government Medical College, Nanded by using Electronic Spirometer. The data regarding heart rate (resting, working and recovery), energy expenditure, musculoskeletal problems and occupational health hazards and accidents were recorded by using pretested questionnaire. Scientific equipment used in the experiment were polar sport test heart rate monitor, Hygrometer, Goniometer, Dialed Thermometer, Sound Level Meter, Lux Meter, Ambient Air Sampler and Air Quality Monitor. The collected data were subjected to appropriate statistical analysis such as 't' test and correlation coefficient test. Results indicated that improved method i.e. use of noise and dust controlling mask, flour presser and dust controlling bag were effective in reducing physiological cost of work of flour mill workers. Lung function test of flour mill workers showed decreased values of all parameters, hence the results of present study support paying attention to use of personal protective noise and dust controlling mask by flour mill workers and also to the working conditions in flour mill especially ventilation and illumination level needs to be enhanced in flour mill. The study also emphasizes the need to develop some mechanism for lifting load of grains and unloading in the hopper. It is also suggested that the flour mill workers should use flour presser suitable to their height to avoid frequent bending and should use dust controlling bag to flour outlet of machine to reduce inhalable flour dust level in the flour mill.

Keywords: physiological cost, energy expenditure, musculoskeletal problems

Procedia PDF Downloads 372

Authors: Bernard Eze Orji

Abstract:

The masquerade, which is the most popular indigenous art form in Africa, is obviously identified by its elaborate, weird, and opulent costumes. The costume is the major essential accouterments in the art of the masquerade. From time past, masquerades have performed and enjoyed the freedom associated with its inscrutability and mystification solely because of its costumes. Noninitiates and women watched masquerades from a distance due to the reverence attached to its costumes and performances. In fact, whether in performance or as an item of art, the masquerade costume was seen as an embodiment of a tradition of liveliness, showiness, secrecy, and sacredness. This liveliness and showiness transformed masked characters who are believed to be possessed by spirits of ancestors and animals that inhabited the costumes. However, with the translocation of masquerade in new festivals such as carnival and state-sponsored cultural days, its costumes have been reduced to a mere item of entertainment and aesthetic values. The sacredness and reverence which hitherto elevated masquerade art to the point of wonderment have given way to an aesthetic appreciation of ingenious and individual creativity deployed in these festivals. This is as a result of the realistic and artistic creations that pervade masquerade costumes and masks in these festivals. It is a common sight to see such masquerades of animal and human genera like a lion, elephant, hippopotamus, and antelope; Agbogho Mmuo, Adamma, and Nchiekwa, respectively. This creative flair has emerged to expunge the ritual narratives associated with masquerades in the past. The study utilized performance analysis and aesthetic theory to establish that the creative ingenuity deployed by fine artists and mask designers who combine traditional artifacts to achieve modern masterpieces for the masquerades of the new festivals have reduced the ritual trappings and hype ascribed to masquerades in indigenous societies.

Keywords: costume and mask designs, entertainment, masquerade, ritual

Procedia PDF Downloads 98

Authors: Marcela De Oliveira, Marina P. Da Silva, Fernando C. G. Da Rocha, Jorge M. Santos, Jaime S. Cardoso, Paulo N. Lisboa-Filho

Abstract:

Multiple Sclerosis (MS) is an immune-mediated disease of the central nervous system characterized by neurodegeneration, inflammation, demyelination, and axonal loss. Magnetic resonance imaging (MRI), due to the richness in the information details provided, is the gold standard exam for diagnosis and follow-up of neurodegenerative diseases, such as MS. Brain atrophy, the gradual loss of brain volume, is quite extensive in multiple sclerosis, nearly 0.5-1.35% per year, far off the limits of normal aging. Thus, the brain volume quantification becomes an essential task for future analysis of the occurrence atrophy. The analysis of MRI has become a tedious and complex task for clinicians, who have to manually extract important information. This manual analysis is prone to errors and is time consuming due to various intra- and inter-operator variability. Nowadays, computerized methods for MRI segmentation have been extensively used to assist doctors in quantitative analyzes for disease diagnosis and monitoring. Thus, the purpose of this work was to evaluate the brain volume in MRI of MS patients. We used MRI scans with 30 slices of the five patients diagnosed with multiple sclerosis according to the McDonald criteria. The computational methods for the analysis of images were carried out in two steps: segmentation of the brain and brain volume quantification. The first image processing step was to perform brain extraction by skull stripping from the original image. In the skull stripper for MRI images of the brain, the algorithm registers a grayscale atlas image to the grayscale patient image. The associated brain mask is propagated using the registration transformation. Then this mask is eroded and used for a refined brain extraction based on level-sets (edge of the brain-skull border with dedicated expansion, curvature, and advection terms). In the second step, the brain volume quantification was performed by counting the voxels belonging to the segmentation mask and converted in cc. We observed an average brain volume of 1469.5 cc. We concluded that the automatic method applied in this work can be used for the brain extraction process and brain volume quantification in MRI. The development and use of computer programs can contribute to assist health professionals in the diagnosis and monitoring of patients with neurodegenerative diseases. In future works, we expect to implement more automated methods for the assessment of cerebral atrophy and brain lesions quantification, including machine-learning approaches. Acknowledgements: This work was supported by a grant from Brazilian agency Fundação de Amparo à Pesquisa do Estado de São Paulo (number 2019/16362-5).

Keywords: brain volume, magnetic resonance imaging, multiple sclerosis, skull stripper

Procedia PDF Downloads 103

Authors: Abe Degale D., Cheng Jian

Abstract:

When creating automated video surveillance systems, violent action recognition is crucial. In recent years, hand-crafted feature detectors have been the primary method for achieving violence detection, such as the recognition of fighting activity. Researchers have also looked into learning-based representational models. On benchmark datasets created especially for the detection of violent sequences in sports and movies, these methods produced good accuracy results. The Hockey dataset's videos with surveillance camera motion present challenges for these algorithms for learning discriminating features. Image recognition and human activity detection challenges have shown success with deep representation-based methods. For the purpose of detecting violent images and identifying aggressive human behaviours, this research suggested a deep representation-based model using the transfer learning idea. The results show that the suggested approach outperforms state-of-the-art accuracy levels by learning the most discriminating features, attaining 99.34% and 99.98% accuracy levels on the Hockey and Movies datasets, respectively.

Keywords: violence detection, faster RCNN, transfer learning and, surveillance video

Procedia PDF Downloads 67

Authors: Anshika Kankane, Dongshik Kang

Abstract:

Marine debris has a significant influence on coastal environments, damaging biodiversity, and causing loss and damage to marine and ocean sector. A functional cost-effective and automatic approach has been used to look up at this problem. Computer vision combined with a deep learning-based model is being proposed to identify and categorize marine debris of seven kinds on different beach locations of Japan. This research compares state-of-the-art deep learning models with a suggested model architecture that is utilized as a feature extractor for debris categorization. The model is being proposed to detect seven categories of litter using a manually constructed debris dataset, with the help of Mask R-CNN for instance segmentation and a shape matching network called HOGShape, which can then be cleaned on time by clean-up organizations using warning notifications of the system. The manually constructed dataset for this system is created by annotating the images taken by fixed KaKaXi camera using CVAT annotation tool with seven kinds of category labels. A pre-trained HOG feature extractor on LIBSVM is being used along with multiple templates matching on HOG maps of images and HOG maps of templates to improve the predicted masked images obtained via Mask R-CNN training. This system intends to timely alert the cleanup organizations with the warning notifications using live recorded beach debris data. The suggested network results in the improvement of misclassified debris masks of debris objects with different illuminations, shapes, viewpoints and litter with occlusions which have vague visibility.

Keywords: computer vision, debris, deep learning, fixed live camera images, histogram of gradients feature extractor, instance segmentation, manually annotated dataset, multiple template matching

Procedia PDF Downloads 70

Authors: Hang Zhou, Limin Zhu

Abstract:

This paper investigates the fabrication of microstructures on Si surfaces by using electrochemical anodic oxidation with agarose stamps. The fabricating process is based on a selective anodic oxidation reaction that occurs in the contact area between a stamp and a Si substrate. The stamp which is soaked in electrolyte previously acts as a current flow channel. After forming the oxide patterns as an etching mask, a KOH aqueous is used for the wet etching of Si. A complicated microstructure array of 1 cm2 was fabricated by the method with high accuracy.

Keywords: microstructures, anodic oxidation, silicon, agarose stamps

Procedia PDF Downloads 272

Authors: Matthew Baucum

Abstract:

With the increasing popularity of fMRI as an experimental method, psychology and neuroscience can greatly benefit from advanced techniques for summarizing and synthesizing large amounts of data from brain imaging studies. One promising avenue is automated meta-analyses, in which natural language processing methods are used to identify the brain regions consistently associated with certain semantic concepts (e.g. “social”, “reward’) across large corpora of studies. This study builds on this approach by demonstrating how, in fMRI meta-analyses, individual voxels can be treated as vectors in a semantic space and evaluated for their “proximity” to terms of interest. In this technique, a low-dimensional semantic space is built from brain imaging study texts, allowing words in each text to be represented as vectors (where words that frequently appear together are near each other in the semantic space). Consequently, each voxel in a brain mask can be represented as a normalized vector sum of all of the words in the studies that showed activation in that voxel. The entire brain mask can then be visualized in terms of each voxel’s proximity to a given term of interest (e.g., “vision”, “decision making”) or collection of terms (e.g., “theory of mind”, “social”, “agent”), as measured by the cosine similarity between the voxel’s vector and the term vector (or the average of multiple term vectors). Analysis can also proceed in the opposite direction, allowing word cloud visualizations of the nearest semantic neighbors for a given brain region. This approach allows for continuous, fine-grained metrics of voxel-term associations, and relies on state-of-the-art “open vocabulary” methods that go beyond mere word-counts. An analysis of over 11,000 neuroimaging studies from an existing meta-analytic fMRI database demonstrates that this technique can be used to recover known neural bases for multiple psychological functions, suggesting this method’s utility for efficient, high-level meta-analyses of localized brain function. While automated text analytic methods are no replacement for deliberate, manual meta-analyses, they seem to show promise for the efficient aggregation of large bodies of scientific knowledge, at least on a relatively general level.

Keywords: FMRI, machine learning, meta-analysis, text analysis

Procedia PDF Downloads 421

Authors: Anoop T. R., Otman Basir, Robert F. Hess, Eileen E. Birch, Brooke A. Koritala, Reed M. Jost, Becky Luu, David Stager, Ben Thompson

Abstract:

Strabismus refers to a misalignment of the eyes. Early detection and treatment of strabismus in childhood can prevent the development of permanent vision loss due to abnormal development of visual brain areas. We developed a two-stage method for strabismus detection and classification based on photographs of the face. The first stage detects the presence or absence of strabismus, and the second stage classifies the type of strabismus. The first stage comprises face detection using Haar cascade, facial landmark estimation, face alignment, aligned face landmark detection, segmentation of the eye region, and detection of strabismus using VGG 16 convolution neural networks. Face alignment transforms the face to a canonical pose to ensure consistency in subsequent analysis. Using facial landmarks, the eye region is segmented from the aligned face and fed into a VGG 16 CNN model, which has been trained to classify strabismus. The CNN determines whether strabismus is present and classifies the type of strabismus (exotropia, esotropia, and vertical deviation). If stage 1 detects strabismus, the eye region image is fed into stage 2, which starts with the estimation of pupil center coordinates using mask R-CNN deep neural networks. Then, the distance between the pupil coordinates and eye landmarks is calculated along with the angle that the pupil coordinates make with the horizontal and vertical axis. The distance and angle information is used to characterize the degree and direction of the strabismic eye misalignment. This model was tested on 100 clinically labeled images of children with (n = 50) and without (n = 50) strabismus. The True Positive Rate (TPR) and False Positive Rate (FPR) of the first stage were 94% and 6% respectively. The classification stage has produced a TPR of 94.73%, 94.44%, and 100% for esotropia, exotropia, and vertical deviations, respectively. This method also had an FPR of 5.26%, 5.55%, and 0% for esotropia, exotropia, and vertical deviation, respectively. The addition of one more feature related to the location of corneal light reflections may reduce the FPR, which was primarily due to children with pseudo-strabismus (the appearance of strabismus due to a wide nasal bridge or skin folds on the nasal side of the eyes).

Keywords: strabismus, deep neural networks, face detection, facial landmarks, face alignment, segmentation, VGG 16, mask R-CNN, pupil coordinates, angle deviation, horizontal and vertical deviation

Procedia PDF Downloads 56

Authors: Jonathan Gong

Abstract:

Purpose AI-driven solutions are at the forefront of many pathology and medical imaging methods. Using algorithms designed to better the experience of medical professionals within their respective fields, the efficiency and accuracy of diagnosis can improve. In particular, X-rays are a fast and relatively inexpensive test that can diagnose diseases. In recent years, X-rays have not been widely used to detect and diagnose COVID-19. The under use of Xrays is mainly due to the low diagnostic accuracy and confounding with pneumonia, another respiratory disease. However, research in this field has expressed a possibility that artificial neural networks can successfully diagnose COVID-19 with high accuracy. Models and Data The dataset used is the COVID-19 Radiography Database. This dataset includes images and masks of chest X-rays under the labels of COVID-19, normal, and pneumonia. The classification model developed uses an autoencoder and a pre-trained convolutional neural network (DenseNet201) to provide transfer learning to the model. The model then uses a deep neural network to finalize the feature extraction and predict the diagnosis for the input image. This model was trained on 4035 images and validated on 807 separate images from the ones used for training. The images used to train the classification model include an important feature: the pictures are cropped beforehand to eliminate distractions when training the model. The image segmentation model uses an improved U-Net architecture. This model is used to extract the lung mask from the chest X-ray image. The model is trained on 8577 images and validated on a validation split of 20%. These models are calculated using the external dataset for validation. The models’ accuracy, precision, recall, f1-score, IOU, and loss are calculated. Results The classification model achieved an accuracy of 97.65% and a loss of 0.1234 when differentiating COVID19-infected, pneumonia-infected, and normal lung X-rays. The segmentation model achieved an accuracy of 97.31% and an IOU of 0.928. Conclusion The models proposed can detect COVID-19, pneumonia, and normal lungs with high accuracy and derive the lung mask from a chest X-ray with similarly high accuracy. The hope is for these models to elevate the experience of medical professionals and provide insight into the future of the methods used.

Keywords: artificial intelligence, convolutional neural networks, deep learning, image processing, machine learning

Procedia PDF Downloads 98

Authors: Andreas Aceranti, Simonetta Vernocchi, Marco Colorato, Daniel Zaccariello

Abstract:

Our study explores the usefulness of implementing facial expression recognition capabilities and using the Facial Action Coding System (FACS) in contexts where the other person is wearing a mask. In the communication process, the subjects use a plurality of distinct and autonomous reporting systems. Among them, the system of mimicking facial movements is worthy of attention. Basic emotion theorists have identified the existence of specific and universal patterns of facial expressions related to seven basic emotions -anger, disgust, contempt, fear, sadness, surprise, and happiness- that would distinguish one emotion from another. However, due to the COVID-19 pandemic, we have come up against the problem of having the lower half of the face covered and, therefore, not investigable due to the masks. Facial-emotional behavior is a good starting point for understanding: (1) the affective state (such as emotions), (2) cognitive activity (perplexity, concentration, boredom), (3) temperament and personality traits (hostility, sociability, shyness), (4) psychopathology (such as diagnostic information relevant to depression, mania, schizophrenia, and less severe disorders), (5) psychopathological processes that occur during social interactions patient and analyst. There are numerous methods to measure facial movements resulting from the action of muscles, see for example, the measurement of visible facial actions using coding systems (non-intrusive systems that require the presence of an observer who encodes and categorizes behaviors) and the measurement of electrical "discharges" of contracting muscles (facial electromyography; EMG). However, the measuring system invented by Ekman and Friesen (2002) - "Facial Action Coding System - FACS" is the most comprehensive, complete, and versatile. Our study, carried out on about 1,500 subjects over three years of work, allowed us to highlight how the movements of the hands and upper part of the face change depending on whether the subject wears a mask or not. We have been able to identify specific alterations to the subjects’ hand movement patterns and their upper face expressions while wearing masks compared to when not wearing them. We believe that finding correlations between how body language changes when our facial expressions are impaired can provide a better understanding of the link between the face and body non-verbal language.

Keywords: facial action coding system, COVID-19, masks, facial analysis

Procedia PDF Downloads 46

Authors: Mohammad Shameel, Peter K. K. Loh, James H. Ng

Abstract:

Cybercrime is a new category of crime which poses a different challenge for crime investigators and incident responders. Attackers can mask their identities using a suite of tools and with the help of the deep web, which makes them difficult to track down. Scouring the deep web manually takes time and is inefficient. There is a growing need for a tool to scour the deep web to obtain useful evidence or intel automatically. In this paper, we will explain the background and motivation behind the research, present a survey of existing research on related tools, describe the design of our own crime/threat intelligence hunting tool prototype, demonstrate its capability with some test cases and lastly, conclude with proposals for future enhancements.

Keywords: cybercrime, deep web, threat intelligence, web crawler

Procedia PDF Downloads 141

Authors: Ebrahim Shayesteh, Nikolaos Styliaras, Alin George Raducu, Ozan Sahin, Daniel Pombo VáZquez, Jonas Funkquist, Sotirios Thanopoulos

Abstract:

One of the important requirements in many code generation projects is defining some of the model parameters tunable. This helps to update the model parameters without performing the code generation again. This paper studies the concept of embedded code generation by MATLAB/Simulink coder targeting the TwinCAT Simulink system. The generated runtime modules are then tested and deployed to the TwinCAT 3 engineering environment. However, defining the parameters tunable in MATLAB/Simulink code generation targeting TwinCAT is not very straightforward. This paper focuses on this subject and reviews some of the techniques tested here to make the parameters tunable in generated runtime modules. Three techniques are proposed for this purpose, including normal tunable parameters, callback functions, and mask subsystems. Moreover, some test Simulink models are developed and used to evaluate the results of proposed approaches. A brief summary of the study results is presented in the following. First of all, the parameters defined tunable and used in defining the values of other Simulink elements (e.g., gain value of a gain block) could be changed after the code generation and this value updating will affect the values of all elements defined based on the values of the tunable parameter. For instance, if parameter K=1 is defined as a tunable parameter in the code generation process and this parameter is used to gain a gain block in Simulink, the gain value for the gain block is equal to 1 in the gain block TwinCAT environment after the code generation. But, the value of K can be changed to a new value (e.g., K=2) in TwinCAT (without doing any new code generation in MATLAB). Then, the gain value of the gain block will change to 2. Secondly, adding a callback function in the form of “pre-load function,” “post-load function,” “start function,” and will not help to make the parameters tunable without performing a new code generation. This means that any MATLAB files should be run before performing the code generation. The parameters defined/calculated in this file will be used as fixed values in the generated code. Thus, adding these files as callback functions to the Simulink model will not make these parameters flexible since the MATLAB files will not be attached to the generated code. Therefore, to change the parameters defined/calculated in these files, the code generation should be done again. However, adding these files as callback functions forces MATLAB to run them before the code generation, and there is no need to define the parameters mentioned in these files separately. Finally, using a tunable parameter in defining/calculating the values of other parameters through the mask is an efficient method to change the value of the latter parameters after the code generation. For instance, if tunable parameter K is used in calculating the value of two other parameters K1 and K2 and, after the code generation, the value of K is updated in TwinCAT environment, the value of parameters K1 and K2 will also be updated (without any new code generation).

Keywords: code generation, MATLAB, tunable parameters, TwinCAT

Procedia PDF Downloads 199

Authors: Adel Edwar Waheeb Louka

Abstract:

Purpose AI-driven solutions are at the forefront of many pathology and medical imaging methods. Using algorithms designed to better the experience of medical professionals within their respective fields, the efficiency and accuracy of diagnosis can improve. In particular, X-rays are a fast and relatively inexpensive test that can diagnose diseases. In recent years, X-rays have not been widely used to detect and diagnose COVID-19. The under use of Xrays is mainly due to the low diagnostic accuracy and confounding with pneumonia, another respiratory disease. However, research in this field has expressed a possibility that artificial neural networks can successfully diagnose COVID-19 with high accuracy. Models and Data The dataset used is the COVID-19 Radiography Database. This dataset includes images and masks of chest X-rays under the labels of COVID-19, normal, and pneumonia. The classification model developed uses an autoencoder and a pre-trained convolutional neural network (DenseNet201) to provide transfer learning to the model. The model then uses a deep neural network to finalize the feature extraction and predict the diagnosis for the input image. This model was trained on 4035 images and validated on 807 separate images from the ones used for training. The images used to train the classification model include an important feature: the pictures are cropped beforehand to eliminate distractions when training the model. The image segmentation model uses an improved U-Net architecture. This model is used to extract the lung mask from the chest X-ray image. The model is trained on 8577 images and validated on a validation split of 20%. These models are calculated using the external dataset for validation. The models’ accuracy, precision, recall, f1-score, IOU, and loss are calculated. Results The classification model achieved an accuracy of 97.65% and a loss of 0.1234 when differentiating COVID19-infected, pneumonia-infected, and normal lung X-rays. The segmentation model achieved an accuracy of 97.31% and an IOU of 0.928. Conclusion The models proposed can detect COVID-19, pneumonia, and normal lungs with high accuracy and derive the lung mask from a chest X-ray with similarly high accuracy. The hope is for these models to elevate the experience of medical professionals and provide insight into the future of the methods used.

Keywords: artificial intelligence, convolutional neural networks, deeplearning, image processing, machine learningSarapin, intraarticular, chronic knee pain, osteoarthritisFNS, trauma, hip, neck femur fracture, minimally invasive surgery

Procedia PDF Downloads 27

Authors: Yinjun Tu, Qiang Fang, Glenn I. Matthews, Shuenn-Yuh Lee

Abstract:

This paper reports an efficient and rigorous self-adaptive stimulus artifacts removal approach for a mixed surface EMG (Electromyography) and stimulus signal during muscle stimulation. The recording of EMG and the stimulation of muscles were performing simultaneously. It is difficult to generate muscle fatigue feature from the mixed signal, which can be further used in closed loop system. A self-adaptive method is proposed in this paper, the stimulation frequency was calculated and verified firstly. Then, a mask was created based on this stimulation frequency to remove the undesired stimulus. 20 EMG signal recordings were analyzed, and the ANOVA (analysis of variance) approach illustrated that the decreasing trend of median power frequencies was successfully generated from the 'cleaned' EMG signal.

Keywords: EMG, FES, stimulus artefacts, self-adaptive

Procedia PDF Downloads 373

Authors: Ravija Gunawardana, Banuka Athuraliya

Abstract:

Machine learning has emerged as a powerful tool for disease diagnosis and prediction. The use of machine learning algorithms has the potential to improve the accuracy of disease prediction, thereby enabling medical professionals to provide more effective and personalized treatments. This study focuses on developing a machine-learning model for disease prediction using symptoms and X-ray images. The importance of this study lies in its potential to assist medical professionals in accurately diagnosing diseases, thereby improving patient outcomes. Respiratory diseases are a significant cause of morbidity and mortality worldwide, and chest X-rays are commonly used in the diagnosis of these diseases. However, accurately interpreting X-ray images requires significant expertise and can be time-consuming, making it difficult to diagnose respiratory diseases in a timely manner. By incorporating machine learning algorithms, we can significantly enhance disease prediction accuracy, ultimately leading to better patient care. The study utilized the Mask R-CNN algorithm, which is a state-of-the-art method for object detection and segmentation in images, to process chest X-ray images. The model was trained and tested on a large dataset of patient information, which included both symptom data and X-ray images. The performance of the model was evaluated using a range of metrics, including accuracy, precision, recall, and F1-score. The results showed that the model achieved an accuracy rate of over 90%, indicating that it was able to accurately detect and segment regions of interest in the X-ray images. In addition to X-ray images, the study also incorporated symptoms as input data for disease prediction. The study used three different classifiers, namely Random Forest, K-Nearest Neighbor and Support Vector Machine, to predict diseases based on symptoms. These classifiers were trained and tested using the same dataset of patient information as the X-ray model. The results showed promising accuracy rates for predicting diseases using symptoms, with the ensemble learning techniques significantly improving the accuracy of disease prediction. The study's findings indicate that the use of machine learning algorithms can significantly enhance disease prediction accuracy, ultimately leading to better patient care. The model developed in this study has the potential to assist medical professionals in diagnosing respiratory diseases more accurately and efficiently. However, it is important to note that the accuracy of the model can be affected by several factors, including the quality of the X-ray images, the size of the dataset used for training, and the complexity of the disease being diagnosed. In conclusion, the study demonstrated the potential of machine learning algorithms for disease prediction using symptoms and X-ray images. The use of these algorithms can improve the accuracy of disease diagnosis, ultimately leading to better patient care. Further research is needed to validate the model's accuracy and effectiveness in a clinical setting and to expand its application to other diseases.

Keywords: K-nearest neighbor, mask R-CNN, random forest, support vector machine

Procedia PDF Downloads 105

Authors: Peter Modregger, Özgül Öztürk

Abstract:

Glass fabric reinforced thermoplastic (GFRT) are composite materials, which combine low weight and resilient mechanical properties rendering them especially suitable for automobile construction. However, defects in the glass fabric as well as in the polymer matrix can occur during manufacturing, which may compromise component lifetime or even safety. One type of these defects is local fiber shifts, which can be difficult to detect. Recently, we have experimentally demonstrated the reliable detection of local fiber shifts by X-ray scattering based on the edge-illumination (EI) principle. EI constitutes a novel X-ray imaging technique that utilizes two slit masks, one in front of the sample and one in front of the detector, in order to simultaneously provide absorption, phase, and scattering contrast. The principle of contrast formation is as follows. The incident X-ray beam is split into smaller beamlets by the sample mask, resulting in small beamlets. These are distorted by the interaction with the sample, and the distortions are scaled up by the detector masks, rendering them visible to a pixelated detector. In the experiment, the sample mask is laterally scanned, resulting in Gaussian-like intensity distributions in each pixel. The area under the curves represents absorption, the peak offset refraction, and the width of the curve represents the scattering occurring in the sample. Here, scattering is caused by the numerous glass fiber/polymer matrix interfaces. In our recent publication, we have shown that the standard deviation of the absorption and scattering values over a selected field of view can be used to distinguish between intact samples and samples with local fiber shift defects. The quantification of defect detection performance was done by using p-values (p=0.002 for absorption and p=0.009 for scattering) and contrast-to-noise ratios (CNR=3.0 for absorption and CNR=2.1 for scattering) between the two groups of samples. This was further improved for the scattering contrast to p=0.0004 and CNR=4.2 by utilizing a harmonic decomposition analysis of the images. Thus, we concluded that local fiber shifts can be reliably detected by the X-ray scattering contrasts provided by EI. However, a potential application in, for example, production monitoring requires fast data acquisition times. For the results above, the scanning of the sample masks was performed over 50 individual steps, which resulted in long total scan times. In this paper, we will demonstrate that reliable detection of local fiber shift defects is also possible by using single images, which implies a speed up of total scan time by a factor of 50. Additional performance improvements will also be discussed, which opens the possibility for real-time acquisition. This contributes a vital step for the translation of EI to industrial applications for a wide variety of materials consisting of numerous interfaces on the micrometer scale.

Keywords: defects in composites, X-ray scattering, local fiber shifts, X-ray edge Illumination

Procedia PDF Downloads 34

Authors: Maria Cristina B. Cabanag, Elijinese Marie S. Culangen

Abstract:

Pregnancy is a period in a woman's life wherein the body may undergo different physiological changes. These changes can be attributed to the interplay of hormones in the body but can mask a more sinister type of disease such as malignancy on rare occasions. Colorectal cancer (CRC) in pregnancy is an epidemiologically rare disease worldwide. To our knowledge, no available studies were reported in the Philippines at the time of this writing, posing a dilemma for its appropriate diagnosis and management. Signs and symptoms of colorectal malignancy may camouflage a normal pregnancy and, when overlooked, impedes an appropriate approach. This case of a 38-year-old elderly primigravid who presented with hematochezia on her 25th week of gestation. She was diagnosed with rectal adenocarcinoma later in pregnancy which warranted a predicament regarding her appropriate care and management. This paper explores the repertoire of the different diagnostic and treatment approaches to colorectal cancer in the second trimester of pregnancy, with the least possible maternal and fetal hazards.

Keywords: cancer in pregnancy, chemotherapy in pregnancy, colorectal cancer, hematochezia in pregnancy

Procedia PDF Downloads 130

Authors: Nazia Tarannum, M. Ayaz Ahmad

Abstract:

The main goal of this article is to report some interesting features for the fabrication/design of PEDOT:PSS based humidity sensor. Here first we fabricated humidity sensor and then studied its electro-mechanical characteristics. In general the humidity plays an important role in various private and government sectors all over the world. Monitoring and controlling the humidity is a great task for the reliable operation of various systems. The PEDOT:PSS is very much promising humidity sensor and also is fabricated by performing various analyses. The interdigited electrode (IDE) has channel length 200 microns prepared by lithography. Lithography of IDE was done on PPR coated glass substrate using negative mask and exposing it with UV light for 10 secs via DSA. During the above said fabrication, we have taken account for the following steps: •Plasma ashing of IDE •Spincoating of PEDOT:PSS was done @3000 rpm on IDE substrace •Baked the substrace at 130 °C up to time limit 15 mins. •Resistance measurement using Labtracer 2.9 software via Keithley 2400source meter.

Keywords: fabrication method, PEDOT:PSS material, humidity sensor, electro-mechanical

Procedia PDF Downloads 317

Authors: Esra Yuksel, Dilek Duman, Levent Yeniay, Sezgin Ulukaya

Abstract:

Sentinel lymph node biopsy (SLNB) is a minimal invasive technique with lower morbidity in axillary staging of breast cancer. Isosulfan blue stain is frequently used in SLNB and regarded as safe. The present case report aimed to report severe decrement in SpO2 following isosulfan blue administration, as well as skin and urine signs and inconsistency with clinical picture in a 67-year-old ,77 kg, ASA II female case that underwent SLNB under general anesthesia. Ten minutes after subcutaneous administration of 10 ml 1% isosulfan blue by the surgeons into the patient, who were hemodynamically stable, SpO2 first reduced to 87% from 99%, and then to 75% in minutes despite 100% oxygen support. Meanwhile, blood pressure and EtCO2 monitoring was unremarkable. After specifying that anesthesia device worked normally, airway pressure did not increase and the endotracheal tube has been placed accurately, the blood sample was taken from the patient for arterial gas analysis. A severe increase was thought in MetHb concentration since SpO2 persisted to be 75% although the concentration of inspired oxygen was 100%, and solution of 2500 mg ascorbic acid in 500 ml 5% Dextrose was given to the patient via intravenous route until the results of arterial blood gas were obtained. However, arterial blood gas results were as follows: pH: 7.54, PaCO2: 23.3 mmHg, PaO2: 281 mmHg, SaO2: %99, and MetHb: %2.7. Biochemical analysis revealed a blood MetHb concentration of 2%.However, since arterial blood gas parameters were good, hemodynamics of the patient was stable and methemoglobin concentration was not so high, the patient was extubated after surgery when she was relaxed, cooperated and had adequate respiration. Despite the absence of respiratory or neurological distress, SpO2 value was increased only up to 85% within 2 hours with 5 L/min oxygen support via face mask in the surgery room as the patient was extubated. At that time, the skin of particularly the upper part of her body has turned into blue, more remarkable on the face. The color of plasma of the blood taken from the patient for biochemical analysis was blue. The color of urine coming throughout the urinary catheter placed in intensive care unit was also blue. Twelve hours after 5 L/min. oxygen inhalation via a mask, the SpO2 reached to 90%. During monitoring in intensive care unit on the postoperative 1st day, facial color and urine color of the patient was still blue, SpO2 was 92%, and arterial blood gas levels were as follows: pH: 7.44, PaO2: 76.1 mmHg, PaCO2: 38.2 mmHg, SaO2: 99%, and MetHb 1%. During monitoring in clinic on the postoperative 2nd day, SpO2 was 95% without oxygen support and her facial and urine color turned into normal. The patient was discharged on the 3rd day without any problem.In conclusion, SLNB is a less invasive alternative to axillary dissection. However, false pulse oximeter reading due to pigment interference is a rare complication of this procedure. Arterial blood gas analysis should be used to confirm any fall in SpO2 reading during monitoring.

Keywords: isosulfan blue, pulse oximetry, SLNB, methemoglobinemia

Procedia PDF Downloads 297