Automatic Early Detection Research Articles

The effect of features combination on coloscopy images of cervical cancer using the support vector machine method

<p>Cervical cancer is cancer that grows in cells in the cervix. This cancer generally develops slowly and only shows symptoms when it has entered an advanced stage. Therefore, it is crucial to detect cervical cancer early before serious complications arise. One way to detect cervical cancer early is to use colposcopy, which is to look closely at the condition of the cervix to find changes in cells in the cervix that have the potential to become cancer. However, this method requires the expertise of an obstetrician. This research proposes the use of image processing techniques to create automatic early detection of cervical cancer based on coloscopy images. In this paper, we will discuss image selection using an approach in the form of comparing the weights of feature vectors and then using a data distribution threshold, features that are not too influential can be eliminated. Image classification uses the Support Vector Machine (SVM) method, which makes it possible to distinguish normal images from abnormal images. Classification with feature selection and merging results can improve the consistency of SVM model performance evenly across all four SVM kernels.</p>

Hybrid ensemble - deep transfer model for early cassava leaf disease classification

Cassava is a most important carbohydrate human food consumed in many African and Asian countries. Cassava leaf disease is the major issue which affects production. Automatic early cassava leaf disease detection through deep learning models and transfer learning models were used for multiclass classification with different approaches. Existing approaches deal with imbalanced dataset for predicting the classes. This research work develops an approach based on hybrid Ensemble - deep transfer model approach for early leaf disease detection. Data augmentation was applied to the raw data for balancing the dataset. Three distinct new hybrid models namely Ensemble(InceptionV3+DenseNet-BC-121-32 + Xception), Ensemble(ResNet50V2+DenseNet-BC-121-32), Ensemble(ResNet50V2+ResNet50) were developed. The proposed model shows high performance results. A broad comparison of the proposed model was performed with custom based Convolutional Neural Network and pre-trained models. Highest accuracy of 88.83% and 97.89% was obtained in ensemble based approach that combined InceptionV3, Xception, DenseNet-BC-121-32 for five class and two class classification respectively.

Identifying Buffet Oscillations Using Sliding-Window Dynamic Mode Decomposition

Buffet is a self-sustaining oscillation that can endanger an aircraft in transonic flight. Early automatic detection of the onset of buffet from simulation data is a necessary step in optimization algorithms used to design new aircraft geometries. We use the Dynamic Mode Decomposition (DMD) eigenvalues to detect the onset of buffet from transient simulations of the OAT15A airfoil across a range of angles of attack. During transients, different eigenvalues may be expressed at different times, and we process shorter segments of the data (sliding windows) in order to obtain robust, time-varying estimates of frequency of oscillation and decay/growth rates. We demonstrate that the onset of oscillation can be identified by analyzing DMD eigenvalues, and that physically different mechanisms of oscillation can additionally be identified from spatial profiles associated with DMD modes. The dominant buffeting modes can be detected through DMD before they become apparent in the physically relevant quantities, such as the coefficient of lift. We expect that such early detection of the onset of buffet could reduce the computational burden of aircraft design and expand the feasible design space.

DeepX-Ray: A Comparative Study of Deep Learning-Based Classification and Segmentation Techniques for Automated Detection and Diagnosis of COVID-19 from Chest X-ray Images

The relentless spread of the SARS-CoV-2 virus, causing COVID-19, has underscored the urgent need for efficient early detection and diagnosis methods to mitigate its impact. While traditional techniques like RT-PCR are valuable, they often suffer from time-consuming processes. In this study, DeepX-Ray is presented, a comprehensive investigation into deep learning-based classification and segmentation approaches for the automated detection of COVID-19 from chest X-ray images. Specifically, the authors focus on constructing a custom convolutional neural network (CNN) model to distinguish between COVID-19 and standard X-ray images and benchmark its performance against established models. For segmentation tasks, the effectiveness of various backbone architectures, including ResNet34, ResNet101, DenseNet201, and ResNet50 within the UNet model framework, is explored, with ResNet50 exhibiting superior performance. Furthermore, a novel dataset comprising 7657 images sourced from three publicly available authentic datasets is introduced, and the labelMe tool is employed by the authors to generate ground truth mask datasets for 4137 images to facilitate segmentation of infected lung areas. The authors' custom CNN model achieves an outstanding classification accuracy of 100%, while the segmentation approach attains a mean Intersection over Union (IoU) score of 96.19%. These results underscore the efficacy of the proposed model in enabling early automatic detection and diagnosis of COVID-19, particularly in resource-constrained and remote settings where establishing traditional laboratories may be impractical. This research significantly advances medical imaging techniques for combating the COVID-19 pandemic.

Automatic early detection of induced colic in horses using accelerometer devices.

To seek appropriate veterinary attention for horses with colic, owners must recognise early signs. Direct observation of horse behaviour has several drawbacks: it is time-consuming, hard to see subtle and common behavioural signs, and is based on intuition and subjective decisions. Due to recent advances in wearables and artificial intelligence, it may be possible to develop diagnostic software that can automatically detect colic signs. To develop a software algorithm to aid in the detection of colic signs and levels of pain. In vivo experiments. Transient colic was induced in eight experimental mares with luteolytic doses of prostaglandin. Veterinarians observed the horses before and throughout the interventions and assigned pain scores which were used to separate colic episodes into none (pain score ≤5), level 1 (pain score 6-10) or level 2 (pain score ≥11). Accelerometric data and videos were collected throughout the experiments and using accelerometric data, the horse's behaviour was classified into normal and 10 pain-related behaviours and an activity index was calculated. Models were designed that utilised behaviour and activity index characteristics both detecting the presence of colic and assessing its severity. To determine the accuracy of the model, the ground truth, that is the veterinarians' observation of colic signs and assessment of pain level, was compared with the automatic detection system. The cross-validation analysis demonstrated an accuracy of 91.2% for detecting colic and an accuracy of 93.8% in differentiating between level 1 colic and level 2 colic. The model was able to accurately classify 10 pain-related behaviours and distinguish them from normal behaviour with a high accuracy. We included a limited number of horses with severe pain related behaviours in the dataset. This constraint affects the accuracy of categorising colic severity rather than limiting the algorithms' capacity to identify early colic signs. Our system for early detection of colic in horses is unique and innovative, and it can distinguish between colic of varying severity.

RANCANG BANGUN ALAT PENDETEKSI DINI PONTENSI KEBAKARAN SECARA OTOMATIS pada RUANGAN TERINTGRASI dengan INTERNET of THINGS

In an effort to enhance safety and early detection of potential fires in indoor spaces, this research designs and constructs an automatic early fire detection device integrated with the Internet of Things (IoT). The device incorporates a flame sensor, MQ2 sensor, DHT11 temperature and humidity sensor, buzzer, LCD 16x2, and Node MCU to provide an effective and efficient solution. The flame sensor is utilized to detect the presence of fire, while the MQ2 sensor is employed to detect potentially hazardous gases associated with fires. The DHT11 sensor provides temperature and humidity information in the surrounding environment, aiding in identifying potential conditions conducive to fires. The Node MCU serves as the central processing unit of the system, collecting data from the sensors and transmitting it to the IoT platform via Wi-Fi connectivity. This data can be accessed and monitored in real-time through applications or online platforms. Upon detecting a potential fire, the buzzer issues an audible alert, and the LCD 16x2 displays detection information for the user. The primary advantage of this device lies in its ability to provide prompt and efficient detection information, accessible remotely through internet connectivity. With the integration of IoT, the device offers a smarter and more responsive solution for addressing potential fires in indoor spaces. The implementation of this device is anticipated to enhance safety and provide better protection against the risk of fires.

MMFNet: Forest Fire Smoke Detection Using Multiscale Convergence Coordinated Pyramid Network With Mixed Attention and Fast-Robust NMS

There is a problem in the field of early automatic detection of forest fire smoke that due to low concentration or tiny size, some smoke is difficult to capture. This paper proposes a multiscale convergence coordinated pyramid network with mixed attention and Fast-robust NMS (MMFNet) for fast detection of forest fire smoke. First, a multiscale convergence coordinated pyramid network is designed, which combines dual-attention feature pyramid network and coordinated convergence module. It improves the detection rate of targets of different sizes. Second, a mixed attention module is designed to focus more on the smoke in the image and enhance the extraction of horizontal and vertical features of smoke. Then a Fast-robust NMS is proposed to accelerate the convergence of bounding boxes and increase the accuracy of prediction box. Finally, a forest fire detection system of the Internet of Things using MMFNet is built. The experimental results show that our method achieves 80.72% AP, 88.52% AP50, 83.45% AP75, 46.88%AR and 154FPS, which is superior to the state-of-art forest fire smoke detection methods.

Machine Learning to Classify Cardiotocography for Fetal Hypoxia Detection.

Fetal hypoxia can cause damaging consequences on babies' such as stillbirth and cerebral palsy. Cardiotocography (CTG) has been used to detect intrapartum fetal hypoxia during labor. It is a non-invasive machine that measures the fetal heart rate and uterine contractions. Visual CTG suffers inconsistencies in interpretations among clinicians that can delay interventions. Machine learning (ML) showed potential in classifying abnormal CTG, allowing automatic interpretation. In the absence of a gold standard, researchers used various surrogate biomarkers to classify CTG, where some were clinically irrelevant. We proposed using Apgar scores as the surrogate benchmark of babies' ability to recover from birth. Apgar scores measure newborns' ability to recover from active uterine contraction, which measures appearance, pulse, grimace, activity and respiration. The higher the Apgar score, the healthier the baby is.We employ signal processing methods to pre-process and extract validated features of 552 raw CTG. We also included CTG-specific characteristics as outlined in the NICE guidelines. We employed ML techniques using 22 features and measured performances between ML classifiers. While we found that ML can distinguish CTG with low Apgar scores, results for the lowest Apgar scores, which are rare in the dataset we used, would benefit from more CTG data for better performance. We need an external dataset to validate our model for generalizability to ensure that it does not overfit a specific population.Clinical Relevance- This study demonstrated the potential of using a clinically relevant benchmark for classifying CTG to allow automatic early detection of hypoxia to reduce decision-making time in maternity units.

DeepDrRVO: A GAN-auxiliary two-step masked transformer framework benefits early recognition and differential diagnosis of retinal vascular occlusion from color fundus photographs

Retinal vascular occlusion (RVO) are common causes of visual impairment. Accurate recognition and differential diagnosis of RVO are unmet medical needs for determining appropriate treatments and health care to properly manage the ocular condition and minimize the damaging effects. To leverage deep learning as a potential solution to detect RVO reliably, we developed a deep learning model on color fundus photographs (CFPs) using a two-step masked SwinTransformer with a Few-Sample Generator (FSG)-auxiliary training framework (called DeepDrRVO) for early and differential RVO diagnosis. The DeepDrRVO was trained on the training set from the in-house cohort and achieved consistently high performance in early recognition and differential diagnosis of RVO in the validation set from the in-house cohort with an accuracy of 86.3%, and other three independent multi-center cohorts with the accuracy of 92.6%, 90.8%, and 100%. Further comparative analysis showed that the proposed DeepDrRVO outperforms conventional state-of-the-art classification models, such as ResNet18, ResNet50d, MobileNetv3, and EfficientNetb1. These results highlight the potential benefits of the deep learning model in automatic early RVO detection and differential diagnosis for improving clinical outcomes and providing insights into diagnosing other ocular diseases with a few-shot learning challenge. The DeepDrRVO is publicly available on https://github.com/ZhouSunLab-Workshops/DeepDrRVO.

Automatic Detection of Cage-Free Dead Hens with Deep Learning Methods

Poultry farming plays a significant role in ensuring food security and economic growth in many countries. However, various factors such as feeding management practices, environmental conditions, and diseases lead to poultry mortality (dead birds). Therefore, regular monitoring of flocks and timely veterinary assistance is crucial for maintaining poultry health, well-being, and the success of poultry farming operations. However, the current monitoring method relies on manual inspection by farm workers, which is time-consuming. Therefore, developing an automatic early mortality detection (MD) model with higher accuracy is necessary to prevent the spread of infectious diseases in poultry. This study aimed to develop, evaluate, and test the performance of YOLOv5-MD and YOLOv6-MD models in detecting poultry mortality under various cage-free (CF) housing settings, including camera height, litter condition, and feather coverage. The results demonstrated that the YOLOv5s-MD model performed exceptionally well, achieving a high mAP@0.50 score of 99.5%, a high FPS of 55.6, low GPU usage of 1.04 GB, and a fast-processing time of 0.4 h. Furthermore, this study also evaluated the models’ performances under different CF housing settings, including different levels of feather coverage, litter coverage, and camera height. The YOLOv5s-MD model with 0% feathered covering achieved the best overall performance in object detection, with the highest mAP@0.50 score of 99.4% and a high precision rate of 98.4%. However, 80% litter covering resulted in higher MD. Additionally, the model achieved 100% precision and recall in detecting hens’ mortality at the camera height of 0.5 m but faced challenges at greater heights such as 2 m. These findings suggest that YOLOv5s-MD can detect poultry mortality more accurately than other models, and its performance can be optimized by adjusting various CF housing settings. Therefore, the developed model can assist farmers in promptly responding to mortality events by isolating affected birds, implementing disease prevention measures, and seeking veterinary assistance, thereby helping to reduce the impact of poultry mortality on the industry, ensuring the well-being of poultry and the overall success of poultry farming operations.

Automatic early detection of wildfire smoke with visible-light cameras and EfficientDet

The ability of EfficientDet, a framework developed in 2019 for object detection, to automatically detect smoke plumes at a distance of several kilometers is demonstrated. Recent articles have raised concerns about the effectiveness of EfficientDet in fire detection applications, with over 40% of false positives reported. The proposed EfficientDet model achieved a true detection rate of 80.4% and a false-positive rate of 1.13% on a testing set. The data set used in this study, which includes 14,125 smoke and 21,203 non-smoke images, is one of the largest, or even the largest, of its kind reported in the literature for images containing smoke plumes. Our results surpass those of a previous study that used the same data set and are more reliable and realistic than those reported by others, which may seem better, but were calculated using smaller and less representative data sets.

Automatic Early Warning System Design with Firefighter Synchronization Based on Internet of Things (IoT)

The fire department or DamKar is one of the important agencies which is highly needed by Indonesian people, especially when a fire occurs. However, DamKar is often judged to be late in arriving at the location because help comes to the fire when the fire has already take a set. One of the reasons why delay happens in DamKar is because the report is received late by the DamKar officer. One of the fire prevention measures is to take the preventive action or early prevention from indications of a fire. Automatic early detection is needed in an emergency and requires speed and accuracy in overcoming the problem. designs an early detection system for fires that directs to DamKar and warehouse owners. This system can detect and provide temperature information in real time. This system works if there is a drastic change in temperature and there is a puff as soon as detected by the sensor The information is in the form of a notification "Excessive CO gas detected" if the temperature is in the range of 25 °C and the gas content is 100 PPM, if the temperature exceeds 35°C and the gas content is 500 PPM there will be a notification "The warehouse temperature is too hot", then a danger notification " indicated fire above”. From the experimental results, it is found that the mesh communication system can work properly

Automatic early detection of rice leaf diseases using hybrid deep learning and machine learning methods

Automatic early detection of rice leaf diseases using hybrid deep learning and machine learning methods

Mathematical Assessment of Machine Learning Models Used for Brain Tumor Diagnosis

The brain is an intrinsic and complicated component of human anatomy. It is a collection of connective tissues and nerve cells that regulate the principal actions of the entire body. Brain tumor cancer is a serious mortality factor and a highly intractable disease. Even though brain tumors are not considered a fundamental cause of cancer deaths worldwide, about 40% of other cancer types are metastasized to the brain and transform into brain tumors. Computer-aided devices for diagnosis through magnetic resonance imaging (MRI) have remained the gold standard for the diagnosis of brain tumors, but this conventional method has been greatly challenged with inefficiencies and drawbacks related to the late detection of brain tumors, high risk in biopsy procedures, and low specificity. To circumvent these underlying hurdles, machine learning models have recently been developed to enhance computer-aided diagnosis tools for advanced, precise, and automatic early detection of brain tumors. This study takes a novel approach to evaluate machine learning models (support vector machine (SVM), random forest (RF), gradient-boosting model (GBM), convolutional neural network (CNN), K-nearest neighbor (KNN), AlexNet, GoogLeNet, CNN VGG19, and CapsNet) used for the early detection and classification of brain tumors by deploying the multicriteria decision-making method called fuzzy preference ranking organization method for enrichment evaluations (PROMETHEE), based on selected parameters, in this study: prediction accuracy, precision, specificity, recall, processing time, and sensitivity. To validate the results of our proposed approach, we performed a sensitivity analysis and cross-checking analysis with the PROMETHEE model. The CNN model, with an outranking net flow of 0.0251, is considered the most favorable model for the early detection of brain tumors. The KNN model, with a net flow of -0.0154, is the least appealing option. The findings of this study support the applicability of the proposed approach for making optimal choices regarding the selection of machine learning models. The decision maker is thus afforded the opportunity to expand the range of considerations which they must rely on in selecting the preferred models for early detection of brain tumors.

A Cloud-Based Deep Learning Framework for Early Detection of Pushing at Crowded Event Entrances

Crowding at the entrances of large events may lead to critical and life-threatening situations, particularly when people start pushing each other to reach the event faster. Automatic and timely identification of pushing behavior would help organizers and security forces to intervene early and mitigate dangerous situations. In this paper, we propose a cloud-based deep learning framework for automatic early detection of pushing in crowded event entrances. The proposed framework initially modifies and trains the EfficientNetV2B0 Convolutional Neural Network model. Subsequently, it integrates the adapted model with an accurate and fast pre-trained deep optical flow model with the color wheel method to analyze video streams and identify pushing patches in real-time. Moreover, the framework uses live capturing technology and a cloud-based environment to collect video streams of crowds in real-time and provide early-stage results. A novel dataset is generated based on five real-world experiments and their associated ground truth data to train the adapted EfficientNetV2B0 model. The experimental setups simulated a crowded event entrance, while the ground truths for each video experiment was generated manually by social psychologists. Several experiments on the videos and the generated dataset are carried out to evaluate the accuracy and annotation delay time of the proposed framework. The experimental results show that the proposed framework identified pushing behaviors with an accuracy rate of 87% within a reasonable delay time.

Two-Stage Classifier for Resource Constrained On-Board Cardiac Arrhythmia Detection

Early automatic detection of cardiac arrhythmia in an ambulatory health monitoring setup can reduce the mortality rates. In the existing approaches on multiclass arrhythmia detection (MCAD) techniques, adopting either time-domain morphological features (TDMFs), fails to extract hidden pattern from the single-lead electrocardiogram (ECG) resulting in poor accuracy (Ac), or adopting deep-learning (DL) techniques becomes too complex or heavy in architecture, thus hindering their easy implementation in low resource architecture for wearable applications. This work presents a judicious hybridization of both approaches in an end-to-end MCAD using two-stage classifier to achieve higher Ac and low latency while detecting 15 classes of cardiac rhythms [14 types of arrhythmia beats or normal sinus rhythms (NSRs)]. The memory resource saving is achieved through the beat skeletonization process and two-stage feature minimization. The stage#1 classifier uses 14 TDMF features fed to a random forest (RF) regressor and provides probability values for NSR, atrial, or ventricular conduction problem. The stage#2 classifier, another RF, accepts stacked autoencoder (SAE) features, TDMF features, and output of stage#1 to achieve a blind test Ac of 99.21% using only 13 features while evaluated with full Massachusetts Institute of Technology-Beth Israel (MIT-BIH) arrhythmia database (mitdb) records. A stand-alone on-board application of the two-stage MCAD using ARMv6 single-core microcontroller yields a detection latency of 782 ms and consumes 390 kB of memory per cardiac cycle, thus proving a feasible solution for wearable cardiac monitors. It also provides the competitive performance with the State of the art (SOTA) methods and outperforms the limited available research on hardware implementation in terms of number of classes handled and Ac.

SVD-CLAHE boosting and balanced loss function for Covid-19 detection from an imbalanced Chest X-Ray dataset

Covid-19 disease has had a disastrous effect on the health of the global population, for the last two years. Automatic early detection of Covid-19 disease from Chest X-Ray (CXR) images is a very crucial step for human survival against Covid-19. In this paper, we propose a novel data-augmentation technique, called SVD-CLAHE Boosting and a novel loss function Balanced Weighted Categorical Cross Entropy (BWCCE), in order to detect Covid 19 disease efficiently from a highly class-imbalanced Chest X-Ray image dataset. Our proposed SVD-CLAHE Boosting method is comprised of both oversampling and under-sampling methods. First, a novel Singular Value Decomposition (SVD) based contrast enhancement and Contrast Limited Adaptive Histogram Equalization (CLAHE) methods are employed for oversampling the data in minor classes. Simultaneously, a Random Under Sampling (RUS) method is incorporated in major classes, so that the number of images per class will be more balanced. Thereafter, Balanced Weighted Categorical Cross Entropy (BWCCE) loss function is proposed in order to further reduce small class imbalance after SVD-CLAHE Boosting. Experimental results reveal that ResNet-50 model on the augmented dataset (by SVD-CLAHE Boosting), along with BWCCE loss function, achieved 95% F1 score, 94% accuracy, 95% recall, 96% precision and 96% AUC, which is far better than the results by other conventional Convolutional Neural Network (CNN) models like InceptionV3, DenseNet-121, Xception etc. as well as other existing models like Covid-Lite and Covid-Net. Hence, our proposed framework outperforms other existing methods for Covid-19 detection. Furthermore, the same experiment is conducted on VGG-19 model in order to check the validity of our proposed framework. Both ResNet-50 and VGG-19 model are pre-trained on the ImageNet dataset. We publicly shared our proposed augmented dataset on Kaggle website (https://www.kaggle.com/tr1gg3rtrash/balanced-augmented-covid-cxr-dataset), so that any research community can widely utilize this dataset. Our code is available on GitHub website online (https://github.com/MrinalTyagi/SVD-CLAHE-and-BWCCE).

Automatic early glaucoma detection by extracting parapapillary atrophy and optic disc from fundus image using SVM

Automatic early glaucoma detection by extracting parapapillary atrophy and optic disc from fundus image using SVM

Automatic Early Detection of Wildfire Smoke With Visible Light Cameras Using Deep Learning and Visual Explanation

The present work focus was developing a system for early automatic detection of smoke plumes in visible-light images. The system used a realistic dataset gathered in 274 different days from a total of nine real surveillance cameras, with most smoke plumes being viewed from afar and 85% of them occupying less than 5% of the image area. We employed the innovative strategy of using the whole image for classification but “asking” the neural networks to indicate, in a multidimensional output, which image regions contained a smoke plume. The multidimensional output helped to focus the detector on the smoke regions. At the same time, the use of the whole image prevented wrong image classification caused by a constrained view of the landscape under analysis. Another strategy used was to rectify the detection results using a visual explanation algorithm, Gradient-weighted Class Activation Mapping (Grad-CAM), to ensure that detections corresponded to the smoke regions in an image. The detection algorithms tested were residual neural networks (ResNet) and EfficientNet of various sizes because these two types have given good results in the past in multiple domains. The training was done using transfer learning. Our dataset contained a total of 14125 and 21203 images with and without smoke, respectively, making it, to the best of the author’s knowledge, one of the largest or even the largest reported dataset in the scientific literature in terms of the number of images with smoke collected from large distances of various kilometers. This dataset was fundamental to achieve realistic results concerning smoke detection efficiency. Our best result in the test set was an Area Under Receiver Operating Characteristic curve (AUROC) of 0.949 obtained with an EfficientNet-B0.

Early Detection of Wild Rocket Tracheofusariosis Using Hyperspectral Image-Based Machine Learning

Fusarium oxysporum f. sp. raphani is responsible for wilting wild rocket (Diplotaxis tenuifolia L. [D.C.]). A machine learning model based on hyperspectral data was constructed to monitor disease progression. Thus, pathogenesis after artificial inoculation was monitored over a 15-day period by symptom assessment, qPCR pathogen quantification, and hyperspectral imaging. The host colonization by a pathogen evolved accordingly with symptoms as confirmed by qPCR. Spectral data showed differences as early as 5-day post infection and 12 hypespectral vegetation indices were selected to follow disease development. The hyperspectral dataset was used to feed the XGBoost machine learning algorithm with the aim of developing a model that discriminates between healthy and infected plants during the time. The multiple cross-prediction strategy of the pixel-level models was able to detect hyperspectral disease profiles with an average accuracy of 0.8. For healthy pixel detection, the mean Precision value was 0.78, the Recall was 0.88, and the F1 Score was 0.82. For infected pixel detection, the average evaluation metrics were Precision: 0.73, Recall: 0.57, and F1 Score: 0.63. Machine learning paves the way for automatic early detection of infected plants, even a few days after infection.