RESUMO
Introduction: Freezing of gait (FoG) is a significant issue for those with Parkinson's disease (PD) since it is a primary contributor to falls and is linked to a poor superiority of life. The underlying apparatus is still not understood; however, it is postulated that it is associated with cognitive disorders, namely impairments in executive and visuospatial functions. During episodes of FoG, patients may experience the risk of falling, which significantly effects their quality of life. Methods: This research aims to systematically evaluate the effectiveness of machine learning approaches in accurately predicting a FoG event before it occurs. The system was tested using a dataset collected from the Kaggle repository and comprises 3D accelerometer data collected from the lower backs of people who suffer from episodes of FoG, a severe indication frequently realized in persons with Parkinson's disease. Data were acquired by measuring acceleration from 65 patients and 20 healthy senior adults while they engaged in simulated daily life tasks. Of the total participants, 45 exhibited indications of FoG. This research utilizes seven machine learning methods, namely the decision tree, random forest, Knearest neighbors algorithm, LightGBM, and CatBoost models. The Gated Recurrent Unit (GRU)-Transformers and Longterm Recurrent Convolutional Networks (LRCN) models were applied to predict FoG. The construction and model parameters were planned to enhance performance by mitigating computational difficulty and evaluation duration. Results: The decision tree exhibited exceptional performance, achieving sensitivity rates of 91% in terms of accuracy, precision, recall, and F1- score metrics for the FoG, transition, and normal activity classes, respectively. It has been noted that the system has the capacity to anticipate FoG objectively and precisely. This system will be instrumental in advancing consideration in furthering the comprehension and handling of FoG.
RESUMO
Diabetic retinopathy is a condition that affects the retina and causes vision loss due to blood vessel destruction. The retina is the layer of the eye responsible for visual processing and nerve signaling. Diabetic retinopathy causes vision loss, floaters, and sometimes blindness; however, it often shows no warning signals in the early stages. Deep learning-based techniques have emerged as viable options for automated illness classification as large-scale medical imaging datasets have become more widely available. To adapt to medical image analysis tasks, transfer learning makes use of pre-trained models to extract high-level characteristics from natural images. In this research, an intelligent recommendation-based fine-tuned EfficientNetB0 model has been proposed for quick and precise assessment for the diagnosis of diabetic retinopathy from fundus images, which will help ophthalmologists in early diagnosis and detection. The proposed EfficientNetB0 model is compared with three transfer learning-based models, namely, ResNet152, VGG16, and DenseNet169. The experimental work is carried out using publicly available datasets from Kaggle consisting of 3,200 fundus images. Out of all the transfer learning models, the EfficientNetB0 model has outperformed with an accuracy of 0.91, followed by DenseNet169 with an accuracy of 0.90. In comparison to other approaches, the proposed intelligent recommendation-based fine-tuned EfficientNetB0 approach delivers state-of-the-art performance on the accuracy, recall, precision, and F1-score criteria. The system aims to assist ophthalmologists in early detection, potentially alleviating the burden on healthcare units.
RESUMO
The research focuses on the segmentation and classification of leukocytes, a crucial task in medical image analysis for diagnosing various diseases. The leukocyte dataset comprises four classes of images such as monocytes, lymphocytes, eosinophils, and neutrophils. Leukocyte segmentation is achieved through image processing techniques, including background subtraction, noise removal, and contouring. To get isolated leukocytes, background mask creation, Erythrocytes mask creation, and Leukocytes mask creation are performed on the blood cell images. Isolated leukocytes are then subjected to data augmentation including brightness and contrast adjustment, flipping, and random shearing, to improve the generalizability of the CNN model. A deep Convolutional Neural Network (CNN) model is employed on augmented dataset for effective feature extraction and classification. The deep CNN model consists of four convolutional blocks having eleven convolutional layers, eight batch normalization layers, eight Rectified Linear Unit (ReLU) layers, and four dropout layers to capture increasingly complex patterns. For this research, a publicly available dataset from Kaggle consisting of a total of 12,444 images of four types of leukocytes was used to conduct the experiments. Results showcase the robustness of the proposed framework, achieving impressive performance metrics with an accuracy of 97.98% and precision of 97.97%. These outcomes affirm the efficacy of the devised segmentation and classification approach in accurately identifying and categorizing leukocytes. The combination of advanced CNN architecture and meticulous pre-processing steps establishes a foundation for future developments in the field of medical image analysis.
Assuntos
Aprendizado Profundo , Humanos , Curadoria de Dados , Leucócitos , Redes Neurais de Computação , Células Sanguíneas , Processamento de Imagem Assistida por Computador/métodosRESUMO
At present, the field of the Internet of Things (IoT) is one of the fastest-growing areas in terms of Artificial Intelligence (AI) and Machine Learning (ML) techniques [...].
RESUMO
The human gastrointestinal (GI) tract is an important part of the body. According to World Health Organization (WHO) research, GI tract infections kill 1.8 million people each year. In the year 2019, almost 5 million individuals were detected with gastrointestinal disease. Radiation therapy has the potential to improve cure rates in GI cancer patients. Radiation oncologists direct X-ray beams at the tumour while avoiding the stomach and intestines. The current objective is to direct the X-ray beam toward the malignancy while avoiding the stomach and intestines in order to improve dose delivery to the tumour. This study offered a technique for segmenting GI tract organs (small bowel, big intestine, and stomach) to assist radio oncologists to treat cancer patients more quickly and accurately. The suggested model is a U-Net model designed from scratch and used for the segmentation of a small size of images to extract the local features more efficiently. Furthermore, in the proposed model, six transfer learning models were employed as the backbone of the U-Net topology. The six transfer learning models used are Inception V3, SeResNet50, VGG19, DenseNet121, InceptionResNetV2, and EfficientNet B0. The suggested model was analysed with model loss, dice coefficient, and IoU. The results specify that the suggested model outperforms all transfer learning models, with performance parameter values as 0.122 model loss, 0.8854 dice coefficient, and 0.8819 IoU.
RESUMO
Human skin disease, the most infectious dermatological ailment globally, is initially diagnosed by sight. Some clinical screening and dermoscopic analysis of skin biopsies and scrapings for accurate classification are medically compulsory. Classification of skin diseases using medical images is more challenging because of the complex formation and variant colors of the disease and data security concerns. Both the Convolution Neural Network (CNN) for classification and a federated learning approach for data privacy preservation show significant performance in the realm of medical imaging fields. In this paper, a custom image dataset was prepared with four classes of skin disease, a CNN model was suggested and compared with several benchmark CNN algorithms, and an experiment was carried out to ensure data privacy using a federated learning approach. An image augmentation strategy was followed to enlarge the dataset and make the model more general. The proposed model achieved a precision of 86%, 43%, and 60%, and a recall of 67%, 60%, and 60% for acne, eczema, and psoriasis. In the federated learning approach, after distributing the dataset among 1000, 1500, 2000, and 2500 clients, the model showed an average accuracy of 81.21%, 86.57%, 91.15%, and 94.15%. The CNN-based skin disease classification merged with the federated learning approach is a breathtaking concept to classify human skin diseases while ensuring data security.
Assuntos
Psoríase , Dermatopatias , Humanos , Dermatopatias/diagnóstico por imagem , Pele/diagnóstico por imagem , Psoríase/diagnóstico por imagem , Internet , Aprendizado de MáquinaRESUMO
Breast cancer can develop when breast cells replicate abnormally. It is now a worldwide issue that concerns people's safety all around the world. Every day, women die from breast cancer, which is especially common in the United States. Mammography, CT, MRI, ultrasound, and biopsies may all be used to detect breast cancer. Histopathology (biopsy) is often carried out to examine the image and discover breast cancer. Breast cancer detection at an early stage saves lives. Deep and machine learning models aid in the detection of breast cancer. The aim of the research work is to encourage medical research and the development of technology by employing deep learning models to recognize cancer cells that are small in size. For histological annotation and diagnosis, the proposed technique makes use of the BreCaHAD dataset. Color divergence is caused by differences in slide scanners, staining procedures, and biopsy materials. To avoid overfitting, we used data augmentation with 19 factors, such as scale, rotation, and gamma. The proposed hybrid dilation deep learning model is of two sorts. It illustrates edges, curves, and colors, and it improves the key traits. It utilizes dilation convolution and max pooling for multi-scale information. The proposed dilated unit processes the image and sends the processed features to the Alexnet, and it can recognize minute objects and thin borders by using the dilated residual expanding kernel model. An AUC of 96.15 shows that the new strategy is better than the old one.
RESUMO
Skin is the primary protective layer of the internal organs of the body. Nowadays, due to increasing pollution and multiple other factors, various types of skin diseases are growing globally. With variable shapes and multiple types, the classification of skin lesions is a challenging task. Motivated by this spreading deformity in society, a lightweight and efficient model is proposed for the highly accurate classification of skin lesions. Dynamic-sized kernels are used in layers to obtain the best results, resulting in very few trainable parameters. Further, both ReLU and leakyReLU activation functions are purposefully used in the proposed model. The model accurately classified all of the classes of the HAM10000 dataset. The model achieved an overall accuracy of 97.85%, which is much better than multiple state-of-the-art heavy models. Further, our work is compared with some popular state-of-the-art and recent existing models.
RESUMO
Chest X-ray (CXR) scans are emerging as an important diagnostic tool for the early spotting of COVID and other significant lung diseases. The recognition of visual symptoms is difficult and can take longer time by radiologists as CXR provides various signs of viral infection. Therefore, artificial intelligence-based method for automated identification of COVID by utilizing X-ray images has been found to be very promising. In the era of deep learning, effective utilization of existing pretrained generalized models is playing a decisive role in terms of time and accuracy. In this paper, the benefits of weights of existing pretrained model VGG16 and InceptionV3 have been taken. Base model has been created using pretrained models (VGG16 and InceptionV3). The last fully connected (FC) layer has been added as per the number of classes for classification of CXR in binary and multi-class classification by appropriately using transfer learning. Finally, combination of layers is made by integrating the FC layer weights of both the models (VGG16 and InceptionV3). The image dataset used for experimentation consists of healthy, COVID, pneumonia viral, and pneumonia bacterial. The proposed weight fusion method has outperformed the existing models in terms of accuracy, achieved 99.5% accuracy in binary classification over 20 epochs, and 98.2% accuracy in three-class classification over 100 epochs.
Assuntos
COVID-19 , Pneumonia , Inteligência Artificial , COVID-19/diagnóstico por imagem , Humanos , Inteligência , Pneumonia/diagnóstico por imagem , Projetos de PesquisaRESUMO
Microvascular problems of diabetes, such as diabetic retinopathy and macular edema, can be seen in the eye's retina, and the retinal images are being used to screen for and diagnose the illness manually. Using deep learning to automate this time-consuming process might be quite beneficial. In this paper, a deep neural network, i.e., convolutional neural network, has been proposed for predicting diabetes through retinal images. Before applying the deep neural network, the dataset is preprocessed and normalised for classification. Deep neural network is constructed by using 7 layers, 5 kernels, and ReLU activation function, and MaxPooling is implemented to combine important features. Finally, the model is implemented to classify whether the retinal image belongs to a diabetic or nondiabetic class. The parameters used for evaluating the model are accuracy, precision, recall, and F1 score. The implemented model has achieved a training accuracy of more than 95%, which is much better than the other states of the art algorithms.
Assuntos
Diabetes Mellitus , Retinopatia Diabética , Algoritmos , Retinopatia Diabética/diagnóstico por imagem , Humanos , Redes Neurais de Computação , Retina/diagnóstico por imagemRESUMO
Neurological imbalance sometimes resulted in stress, which is experienced by the number of people at some moment in their life. A considerable measurement scheme can quantify the stress level in an individual, in which music has always been considered as the best therapy for stress relief in healthy human being as well in severe medical conditions. In this work, the impact of four types of music interventions with the lyrics of Hindi music and varying spectral centroid has been studied for an analysis of stress relief in males and females. The self-reported data for stress using state-trait anxiety (STA) and electroencephalography (EEG) signals for 14 channels in response to music interventions have been considered. Features such as Hjorth (activity, mobility, and complexity), variance, standard deviation, skew, kurtosis, and mean have been extracted from five bands (delta, theta, alpha, beta, and gamma) of each channel of the recorded EEG signals from 9 males and 9 females of the age category between 18 and 25 years. The support vector machine classifier has been used to classify three subsets: (i) male and female, (ii) baseline and female, and (iii) baseline and male. The noteworthy accuracy of 100% was found at the delta band for the first subset, beta and gamma bands for the second subset, and beta, gamma, and delta bands for the third subset. STA score has shown more deviation in the male category than in female, which gives a clear insight into the impact of music intervention with varying spectral centroid that has a higher impact to relieve stress in the male category than the female category.
Assuntos
Música , Doenças Neurodegenerativas , Adolescente , Adulto , Eletroencefalografia/métodos , Feminino , Humanos , Masculino , Máquina de Vetores de Suporte , Adulto JovemRESUMO
Sign language is essential for deaf and mute people to communicate with normal people and themselves. As ordinary people tend to ignore the importance of sign language, which is the mere source of communication for the deaf and the mute communities. These people are facing significant downfalls in their lives because of these disabilities or impairments leading to unemployment, severe depression, and several other symptoms. One of the services they are using for communication is the sign language interpreters. But hiring these interpreters is very costly, and therefore, a cheap solution is required for resolving this issue. Therefore, a system has been developed that will use the visual hand dataset based on an Arabic Sign Language and interpret this visual data in textual information. The dataset used consists of 54049 images of Arabic sign language alphabets consisting of 1500\ images per class, and each class represents a different meaning by its hand gesture or sign. Various preprocessing and data augmentation techniques have been applied to the images. The experiments have been performed using various pretrained models on the given dataset. Most of them performed pretty normally and in the final stage, the EfficientNetB4 model has been considered the best fit for the case. Considering the complexity of the dataset, models other than EfficientNetB4 do not perform well due to their lightweight architecture. EfficientNetB4 is a heavy-weight architecture that possesses more complexities comparatively. The best model is exposed with a training accuracy of 98 percent and a testing accuracy of 95 percent.
Assuntos
Surdez , Língua de Sinais , Gestos , Humanos , Idioma , Aprendizado de MáquinaRESUMO
The concept of fuzzy set, intuitionistic set, and mediative fuzzy set as a generalization of a crisp set have been introduced in many real-life applications. The concept of crisp relation between elements of sets can be extended to fuzzy relations. Extended relations will be considered as relations on fuzzy sets. In this work, we developed the concept of mediative fuzzy relation and meditative fuzzy projection in the context of fuzzy relation and fuzzy projection. We extended the basic operations of fuzzy projection into intuitionistic fuzzy projection and then in the mediative fuzzy projection. We have shown the credibility and impact of mediative index factor involves in the mediative fuzzy projection in context of prediction work in relation to the proposed model. Further, we applied the mediative fuzzy projection in the medical diagnosis in post-COVID-19 patients. The obtained results have also been discussed with their geometrical representation.
RESUMO
Coronavirus disease is a viral infection caused by a novel coronavirus (CoV) which was first identified in the city of Wuhan, China somewhere in the early December 2019. It affects the human respiratory system by causing respiratory infections with symptoms (mild to severe) like fever, cough, and weakness but can further lead to other serious diseases and has resulted in millions of deaths until now. Therefore, an accurate diagnosis for such types of diseases is highly needful for the current healthcare system. In this paper, a state of the art deep learning method is described. We propose COVDC-Net, a Deep Convolutional Network-based classification method which is capable of identifying SARS-CoV-2 infected amongst healthy and/or pneumonia patients from their chest X-ray images. The proposed method uses two modified pre-trained models (on ImageNet) namely MobileNetV2 and VGG16 without their classifier layers and fuses the two models using the Confidence fusion method to achieve better classification accuracy on the two currently publicly available datasets. It is observed through exhaustive experiments that the proposed method achieved an overall classification accuracy of 96.48% for 3-class (COVID-19, Normal and Pneumonia) classification tasks. For 4-class classification (COVID-19, Normal, Pneumonia Viral, and Pneumonia Bacterial) COVDC-Net method delivered 90.22% accuracy. The experimental results demonstrate that the proposed COVDC-Net method has shown better overall classification accuracy as compared to the existing deep learning methods proposed for the same task in the current COVID-19 pandemic.
RESUMO
The disorder of Alzheimer's (AD) is defined as a gradual deterioration of cognitive functions, such as the failure of spatial cognition and short-term memory. Besides difficulties in memory, a person with this disease encounters visual processing difficulties and even awareness and identifying of their beloved ones. Nowadays, recent technologies made this possible to connect everything that exists around us on Earth through the Internet, this is what the Internet of Things (IoT) made possible which can capture and save a massive amount of data that are considered very important and useful information which then can be valuable in training of the various state-of-the-art machine and deep learning algorithms. Assistive mobile health applications and IoT-based wearable devices are helping and supporting the ongoing health screening of a patient with AD. In the early stages of AD, the wearable devices and IoT approach aim to keep AD patients mentally active in all of life's daily activities, independent from their caregivers or any family member of the patient. These technological solutions have great potential in improving the quality of life of an AD patient as this helps to reduce pressure on healthcare and to minimize the operational cost. The purpose of this study is to explore the State-of-the-Art wearable technologies for people with AD. Significance, challenges, and limitations that arise and what will be the future of these technological solutions and their acceptance. Therefore, this study also provides the challenges and gaps in the current literature review and future directions for other researchers working in the area of developing wearable devices.
Assuntos
Doença de Alzheimer , Internet das Coisas , Dispositivos Eletrônicos Vestíveis , Doença de Alzheimer/diagnóstico , Atenção à Saúde , Humanos , Qualidade de VidaRESUMO
An important aspect of the diagnosis procedure in daily clinical practice is the analysis of dental radiographs. This is because the dentist must interpret different types of problems related to teeth, including the tooth numbers and related diseases during the diagnostic process. For panoramic radiographs, this paper proposes a convolutional neural network (CNN) that can do multitask classification by classifying the X-ray images into three classes: cavity, filling, and implant. In this paper, convolutional neural networks are taken in the form of a NASNet model consisting of different numbers of max-pooling layers, dropout layers, and activation functions. Initially, the data will be augmented and preprocessed, and then, the construction of a multioutput model will be done. Finally, the model will compile and train the model; the evaluation parameters used for the analysis of the model are loss and the accuracy curves. The model has achieved an accuracy of greater than 96% such that it has outperformed other existing algorithms.
Assuntos
Redes Neurais de Computação , Doenças Estomatognáticas , Algoritmos , Humanos , Radiografia Panorâmica/métodos , Raios XRESUMO
It is critical to establish a reliable method for detecting people infected with COVID-19 since the pandemic has numerous harmful consequences worldwide. If the patient is infected with COVID-19, a chest X-ray can be used to determine this. In this work, an X-ray showing a COVID-19 infection is classified by the capsule neural network model we trained to recognise. 6310 chest X-ray pictures were used to train the models, separated into three categories: normal, pneumonia, and COVID-19. This work is considered an improved deep learning model for the classification of COVID-19 disease through X-ray images. Viewpoint invariance, fewer parameters, and better generalisation are some of the advantages of CapsNet compared with the classic convolutional neural network (CNN) models. The proposed model has achieved an accuracy greater than 95% during the model's training, which is better than the other state-of-the-art algorithms. Furthermore, to aid in detecting COVID-19 in a chest X-ray, the model could provide extra information.
Assuntos
COVID-19 , Aprendizado Profundo , COVID-19/diagnóstico por imagem , Humanos , Redes Neurais de Computação , Tórax/diagnóstico por imagem , Raios XRESUMO
Breast cancer is a global cause for concern owing to its high incidence around the world. The alarming increase in breast cancer cases emphasizes the management of disease at multiple levels. The management should start from the beginning that includes stringent cancer screening or cancer registry to effective diagnostic and treatment strategies. Breast cancer is highly heterogeneous at morphology as well as molecular levels and needs different therapeutic regimens based on the molecular subtype. Breast cancer patients with respective subtype have different clinical outcome prognoses. Breast cancer heterogeneity emphasizes the advanced molecular testing that will help on-time diagnosis and improved survival. Emerging fields such as liquid biopsy and artificial intelligence would help to under the complexity of breast cancer disease and decide the therapeutic regimen that helps in breast cancer management. In this review, we have discussed various risk factors and advanced technology available for breast cancer diagnosis to combat the worst breast cancer status and areas that need to be focused for the better management of breast cancer.
Assuntos
Neoplasias da Mama , Inteligência Artificial , Neoplasias da Mama/diagnóstico , Neoplasias da Mama/epidemiologia , Neoplasias da Mama/prevenção & controle , Detecção Precoce de Câncer , Feminino , Humanos , Incidência , Fatores de RiscoRESUMO
Malaria comes under one of the dangerous diseases in many countries. It is the primary reason for most of the causalities across the world. It is presently rated as a significant cause of the high mortality rate worldwide compared with other diseases that can be reduced significantly by its earlier detection. Therefore, to facilitate the early detection/diagnosis of malaria to reduce the mortality rate, an automated computational method is required with a high accuracy rate. This study is a solid starting point for researchers who want to look into automated blood smear analysis to detect malaria. In this paper, a comprehensive review of different computer-assisted techniques has been outlined as follows: (i) acquisition of image dataset, (ii) preprocessing, (iii) segmentation of RBC, and (iv) feature extraction and selection, and (v) classification for the detection of malaria parasites using blood smear images. This study will be helpful for: (i) researchers can inspect and improve the existing computational methods for early diagnosis of malaria with a high accuracy rate that may further reduce the interobserver and intra-observer variations; (ii) microbiologists to take the second opinion from the automated computational methods for effective diagnosis of malaria; and (iii) finally, several issues remain addressed, and future work has also been discussed in this work.