RESUMO
BACKGROUND: Hepatic cystic echinococcosis is the main form of hepatic echinococcosis, which is a life-threatening liver disease caused by parasites that requires a precise diagnosis and proper treatment. OBJECTIVE: This study focuses on the automatic classification system of five different subtypes of hepatic cystic echinococcosis based on ultrasound images and deep learning algorithms. METHODS: Three popular deep convolutional neural networks (VGG19, Inception-v3, and ResNet18) with and without pretrained weights were selected to test their performance on the classification task, and the experiments were followed by a 5-fold cross-validation process. RESULTS: A total of 1820 abdominal ultrasound images covering five subtypes of hepatic cystic echinococcosis from 967 patients were used in the study. The classification accuracy for the models with pretrained weights (fine-tuning) ranged from 88.2 to 90.6%. The best accuracy of 90.6% was obtained by VGG19. For comparison, the models without pretrained weights (from scratch) achieved a lower accuracy, ranging from 69.4 to 75.1%. CONCLUSION: Deep convolutional neural networks with pretrained weights are capable of recognizing different subtypes of hepatic cystic echinococcosis from ultrasound images, which are expected to be applied in the computer-aided diagnosis systems in future work.
Assuntos
Aprendizado Profundo , Equinococose Hepática , Diagnóstico por Computador , Equinococose Hepática/diagnóstico por imagem , Humanos , Redes Neurais de Computação , UltrassonografiaRESUMO
BACKGROUND: Neonatal hyperbilirubinemia is a common clinical condition that requires medical attention in newborns, which may develop into acute bilirubin encephalopathy with a significant risk of long-term neurological deficits. The current clinical challenge lies in the separation of acute bilirubin encephalopathy and non-acute bilirubin encephalopathy neonates both with hyperbilirubinemia condition since both of them demonstrated similar T1 hyperintensity and lead to difficulties in clinical diagnosis based on the conventional radiological reading. This study aims to investigate the utility of T1-weighted MRI images for differentiating acute bilirubin encephalopathy and non-acute bilirubin encephalopathy neonates with hyperbilirubinemia. METHODS: 3 diagnostic approaches, including a visual inspection, a semi-quantitative method based on normalized the T1-weighted intensities of the globus pallidus and subthalamic nuclei, and a deep learning method with ResNet18 framework were applied to classify 47 acute bilirubin encephalopathy neonates and 32 non-acute bilirubin encephalopathy neonates with hyperbilirubinemia based on T1-weighted images. Chi-squared test and t-test were used to test the significant difference of clinical features between the 2 groups. RESULTS: The visual inspection got a poor diagnostic accuracy of 53.58 ± 5.71% indicating the difficulty of the challenge in real clinical practice. However, the semi-quantitative approach and ResNet18 achieved a classification accuracy of 62.11 ± 8.03% and 72.15%, respectively, which outperformed visual inspection significantly. CONCLUSION: Our study indicates that it is not sufficient to only use T1-weighted MRI images to detect neonates with acute bilirubin encephalopathy. Other more MRI multimodal images combined with T1-weighted MRI images are expected to use to improve the accuracy in future work. However, this study demonstrates that the semi-quantitative measurement based on T1-weighted MRI images is a simple and compromised way to discriminate acute bilirubin encephalopathy and non-acute bilirubin encephalopathy neonates with hyperbilirubinemia, which may be helpful in improving the current manual diagnosis.
Assuntos
Aprendizado Profundo , Kernicterus/diagnóstico por imagem , Imageamento por Ressonância Magnética/métodos , Doença Aguda , Distribuição de Qui-Quadrado , Diagnóstico Diferencial , Feminino , Globo Pálido/diagnóstico por imagem , Humanos , Hiperbilirrubinemia Neonatal/complicações , Recém-Nascido , Kernicterus/classificação , Masculino , Estudos Retrospectivos , Núcleo Subtalâmico/diagnóstico por imagemRESUMO
Image feature extraction is an important part of image processing and it is an important field of research and application of image processing technology. Uygur medicine is one of Chinese traditional medicine and researchers pay more attention to it. But large amounts of Uygur medicine data have not been fully utilized. In this study, we extracted the image color histogram feature of herbal and zooid medicine of Xinjiang Uygur. First, we did preprocessing, including image color enhancement, size normalizition and color space transformation. Then we extracted color histogram feature and analyzed them with statistical method. And finally, we evaluated the classification ability of features by Bayes discriminant analysis. Experimental results showed that high accuracy for Uygur medicine image classification was obtained by using color histogram feature. This study would have a certain help for the content-based medical image retrieval for Xinjiang Uygur medicine.
Assuntos
Análise Discriminante , Medicamentos de Ervas Chinesas/análise , Teorema de Bayes , Cor , Medicina Tradicional ChinesaRESUMO
To properly treat and care for hepatic cystic echinococcosis (HCE), it is essential to make an accurate diagnosis before treatment. OBJECTIVE: The objective of this study was to assess the diagnostic accuracy of computer-aided diagnosis techniques in classifying HCE ultrasound images into five subtypes. METHODS: A total of 1820 HCE ultrasound images collected from 967 patients were included in the study. A multi-kernel learning method was developed to learn the texture and depth features of the ultrasound images. Combined kernel functions were built-in Support Vector Machine (MK-SVM) for the classification work. The experimental results were evaluated using five-fold cross-validation. Finally, our approach was compared with three other machine learning algorithms: the decision tree classifier, random forest, and gradient boosting decision tree. RESULTS: Among all the methods used in the study, the MK-SVM achieved the highest accuracy of 96.6% on the fused feature set. CONCLUSION: The multi-kernel learning method effectively learns different image features from ultrasound images by utilizing various kernels. The MK-SVM method, which combines the learning of texture features and depth features separately, has significant application value in HCE classification tasks.
Assuntos
Equinococose Hepática , Aprendizado de Máquina , Ultrassonografia , Humanos , Equinococose Hepática/diagnóstico por imagem , Ultrassonografia/métodos , Masculino , Fígado/diagnóstico por imagem , Feminino , Adulto , Pessoa de Meia-Idade , Máquina de Vetores de Suporte , Reprodutibilidade dos Testes , Algoritmos , Idoso , Interpretação de Imagem Assistida por Computador/métodosRESUMO
Hepatic cystic echinococcosis (HCE) is a zoonotic disease that occurs when the larvae of Echinococcus granulosus parasitize the livers of humans and mammals. HCE has five subtypes, and accurate subtype classification is critical for choosing a treatment strategy. To evaluate the clinical utility of artificial intelligence (AI) based on convolutional neural networks (CNNs) in the classification of HCE subtypes via ultrasound imaging, we collected ultrasound images from 4,012 HCE patients at the First Affiliated Hospital of Xinjiang Medical University between 2008 and 2020. Specifically, 1,820 HCE images from 967 patients were used as the training and validation sets for the construction of the AI model, and the remaining 6,808 images from 3,045 patients were used as the test set to evaluate the performance of the AI models. The 6,808 images were randomly divided into six groups, and each group contained equal proportions of the five subtypes. The data of each group were analyzed by a resident physician. The accuracy of HCE subtype classification by the AI model and by manual inspection was compared. The AI HCE classification model showed good performance in the diagnosis of subtypes CE1, CE2, CE4, and CE5. The overall accuracy of the AI classification (90.4%) was significantly greater than that of manual classification by physicians (86.1%; P <0.05). The CNN can better identify the five subtypes of HCE on ultrasound images and should help doctors with little experience in more accurately diagnosing HCE.
Assuntos
Inteligência Artificial , Equinococose Hepática , Ultrassonografia , Humanos , Equinococose Hepática/diagnóstico por imagem , Equinococose Hepática/classificação , Ultrassonografia/métodos , Masculino , Pessoa de Meia-Idade , Feminino , Adulto , Redes Neurais de Computação , Animais , Fígado/diagnóstico por imagem , Fígado/parasitologia , IdosoRESUMO
Xinjiang local liver hydatid disease is an infectious parasitic disease in Xinjiang pastoral areas. Based on the image features, selecting the appropriate distance algorithms to retrieve the image quickly and accurately, different distance algorithms have been induced in this area, which can greatly assist the doctors to early detect, diagnose and cure the liver hydatid disease. This paper compared the performance of different distance algorithms to retrieve the image when using the liver hydatid disease medical image texture features. The results showed that: for the liver hydatid disease medical images retrieval based on gray level cocurrence matrix (GLCM) texture features, the Mahalanobis distance algorithm is superior to other distance algorithms.
Assuntos
Algoritmos , Equinococose Hepática/diagnóstico por imagem , Processamento de Imagem Assistida por Computador/métodos , Armazenamento e Recuperação da Informação/métodos , Tomografia Computadorizada por Raios X/métodos , China , Bases de Dados Factuais , HumanosRESUMO
BACKGROUND: Hepatic cystic echinococcosis (HCE) still has a high misdiagnosis rate, and misdiagnosis may lead to wrong treatments seriously harmful for the patients. Precise diagnosis of HCE relies heavily on the experience of clinical experts with auxiliary diagnostic tools using medical images. PURPOSE: This paper intends to improve the diagnostic accuracy for HCE by employing a method which combines deep learning with ensemble method. METHODS: We proposed a method, namely HCEs-Net, for classification of five HCE subtypes using ultrasound images. It takes first the snap-shot strategy to obtain sub-models from the pre-trained VGG19, ResNet18, ViT-Base, and ConvNeXt-T models, then a stacking process to ensemble those sub-models. Afterwards, it uses the tree-structured Pazren estimator (TPE) to optimize the hyperparameters. The experiments were evaluated by the five-fold cross-validation process. RESULTS: A total of 3083 abdominal ultrasound images from 972 patients covering five subtypes of HCE were utilized in this study. The experiments were conducted to predict the HCE subtype, and results of modeling performance evaluation were reported in terms of precision, recall, F1-score, and AUC. The stacking model based on three ConvNeXt-T sub-models showed the best performance, with precision 85.9%, recall 85.5%, F1-score 85.7%, and AUC 0.971 which are higher than the compared state-of-the-art models. CONCLUSION: The stacking model of three ConvNeXt-T sub-models shows comparable or superior performance to the other methods, including VGG19, ResNet18 and ViT-Base. It has the potential to enhance clinical diagnosis for HCE.
Assuntos
Equinococose , Humanos , Dieldrin , Projetos de PesquisaRESUMO
BACKGROUND: Automatic classification of brain tumors is an important issue in computeraided diagnosis (CAD) for medical applications since it can efficiently improve the clinician's diagnostic performance and the current study focused on the CAD system of the brain tumors. METHODS: Existing studies mainly focused on a single classifier either based on traditional machinelearning algorithms or deep learning algorithms with unsatisfied results. In this study, we proposed an ensemble of pre-trained convolutional neural networks to classify brain tumors into three types from their T1-weighted contrast-enhanced MRI (CE-MRI) images, which are meningioma, glioma, and pituitary tumor. Three pre-trained convolutional neural networks (Inception-v3, Resnet101, Densenet201) with the best classification performance (i.e. accuracy of 96.21%, 97.00%, 96.54%, respectively) on the CE-MRI benchmark dataset were selected as backbones of the ensemble model. The features extracted by backbone networks in the ensemble model were further classified by a support vector machine. RESULTS: The ensemble system achieved an average classification accuracy of 98.14% under a five-fold cross-validation process, outperforming any single deep learning model in the ensemble system and other methods in the previous studies. Performance metrics for each brain tumor type, including area under the curve, sensitivity, specificity, precision, and F-score, were calculated to show the ensemble system's performance. Our work addressed a practical issue by evaluating the model with fewer training samples. The classification accuracy was reduced to 97.23%, 96.87%, and 93.96% when 75%, 50%, and 25% training data was used to train the ensemble model, respectively. CONCLUSION: Our ensemble model has a great capacity and achieved the best performance in any single convolutional neural networks for brain tumors classification and is potentially applicable in real clinical practice.
Assuntos
Neoplasias Encefálicas , Humanos , Neoplasias Encefálicas/diagnóstico por imagem , Imageamento por Ressonância Magnética , Redes Neurais de Computação , Diagnóstico por Computador , AlgoritmosRESUMO
Ovarian cancer is one of the most common gynecologic malignancies. Accurate classification of ovarian cancer types (serous carcinoma, mucous carcinoma, endometrioid carcinoma, transparent cell carcinoma) is an essential part in the different diagnosis. Computer-aided diagnosis (CADx) can provide useful advice for pathologists to determine the diagnosis correctly. In our study, we employed a Deep Convolutional Neural Networks (DCNN) based on AlexNet to automatically classify the different types of ovarian cancers from cytological images. The DCNN consists of five convolutional layers, three max pooling layers, and two full reconnect layers. Then we trained the model by two group input data separately, one was original image data and the other one was augmented image data including image enhancement and image rotation. The testing results are obtained by the method of 10-fold cross-validation, showing that the accuracy of classification models has been improved from 72.76 to 78.20% by using augmented images as training data. The developed scheme was useful for classifying ovarian cancers from cytological images.
Assuntos
Adenocarcinoma Mucinoso/diagnóstico , Carcinoma Endometrioide/diagnóstico , Citodiagnóstico , Neoplasias Ovarianas/diagnóstico , Adenocarcinoma Mucinoso/patologia , Carcinoma Endometrioide/patologia , Diagnóstico Diferencial , Feminino , Humanos , Processamento de Imagem Assistida por Computador , Redes Neurais de Computação , Neoplasias Ovarianas/classificação , Neoplasias Ovarianas/patologiaRESUMO
Cervical cancer (CC) is one of the most common gynecologic malignancies in the world. The incidence and mortality keep high in some remote and poor medical condition regions in China. In order to improve the current situation and promote the pathologists' diagnostic accuracy of CC in such regions, we tried to propose an intelligent and efficient classification model for CC based on convolutional neural network (CNN) with relatively simple architecture compared with others. The model was trained and tested by two groups of image datasets, respectively, which were original image group with a volume of 3012 datasets and augmented image group with a volume of 108432 datasets. Each group has a number of fixed-size RGB images (227*227) of keratinizing squamous, non-keratinizing squamous, and basaloid squamous. The method of three-folder cross-validation was applied to the model. And the classification accuracy of the models, overall, 93.33% for original image group and 89.48% for augmented image group. The improvement of 3.85% has been achieved by using augmented images as input data for the model. The results got from paired-samples ttest indicated that two models' classification accuracy has a significant difference (P<0.05). The developed scheme we proposed was useful for classifying CCs from cytological images and the model can be served as a pathologist assistance to improve the doctor's diagnostic level of CC, which has a great meaning and huge potential application in poor medical condition areas in China.