Automatic detection and classification of disease in citrus fruit and leaves using a customized CNN based model / Detección y clasificación automática de enfermedades en frutas y hojas de cítricos utilizando un modelo basado en CNN personalizado

India's commercial advancement and development depend heavily on agriculture. A common fruit grown in tropical settings is citrus. A professional judgment is required while analyzing an illness because different diseases have slight variati ons in their symptoms. In order to recognize and classify diseases in citrus fruits and leaves, a customized CNN - based approach that links CNN with LSTM was developed in this research. By using a CNN - based method, it is possible to automatically differenti ate from healthier fruits and leaves and those that have diseases such fruit blight, fruit greening, fruit scab, and melanoses. In terms of performance, the proposed approach achieves 96% accuracy, 98% sensitivity, 96% Recall, and an F1 - score of 92% for ci trus fruit and leave identification and classification and the proposed method was compared with KNN, SVM, and CNN and concluded that the proposed CNN - based model is more accurate and effective at identifying illnesses in citrus fruits and leaves.

El avance y desarrollo comercial de India dependen en gran medida de la agricultura. Un tipo de fruta comunmente cultivada en en tornos tropicales es el cítrico. Se requiere un juicio profesional al analizar una enfermedad porque diferentes enfermedades tienen ligeras variaciones en sus síntomas. Para reconocer y clasificar enfermedades en frutas y hojas de cítricos, se desarrolló e n esta investigación un enfoque personalizado basado en CNN que vincula CNN con LSTM. Al utilizar un método basado en CNN, es posible diferenciar automáticamente entre frutas y hojas más saludables y aquellas que tienen enfermedades como la plaga de frutas , el verdor de frutas, la sarna de frutas y las melanosis. En términos de desempeño, el enfoque propuesto alcanza una precisión del 96%, una sensibilidad del 98%, una recuperación del 96% y una puntuación F1 del 92% para la identificación y clasificación d e frutas y hojas de cítricos, y el método propuesto se comparó con KNN, SVM y CNN y se concluyó que el modelo basado en CNN propuesto es más preciso y efectivo para identificar enfermedades en frutas y hojas de cítricos.

Research progress on medical image dataset expansion methods / 生物医学工程学杂志

Computer-aided diagnosis (CAD) systems play a very important role in modern medical diagnosis and treatment systems, but their performance is limited by training samples. However, the training samples are affected by factors such as imaging cost, labeling cost and involving patient privacy, resulting in insufficient diversity of training images and difficulty in data obtaining. Therefore, how to efficiently and cost-effectively augment existing medical image datasets has become a research hotspot. In this paper, the research progress on medical image dataset expansion methods is reviewed based on relevant literatures at home and abroad. First, the expansion methods based on geometric transformation and generative adversarial networks are compared and analyzed, and then improvement of the augmentation methods based on generative adversarial networks are emphasized. Finally, some urgent problems in the field of medical image dataset expansion are discussed and the future development trend is prospected.

Alzheimer's disease classification based on nonlinear high-order features and hypergraph convolutional neural network / 生物医学工程学杂志

Alzheimer's disease (AD) is an irreversible neurodegenerative disorder that damages patients' memory and cognitive abilities. Therefore, the diagnosis of AD holds significant importance. The interactions between regions of interest (ROIs) in the brain often involve multiple areas collaborating in a nonlinear manner. Leveraging these nonlinear higher-order interaction features to their fullest potential contributes to enhancing the accuracy of AD diagnosis. To address this, a framework combining nonlinear higher-order feature extraction and three-dimensional (3D) hypergraph neural networks is proposed for computer-assisted diagnosis of AD. First, a support vector machine regression model based on the radial basis function kernel was trained on ROI data to obtain a base estimator. Then, a recursive feature elimination algorithm based on the base estimator was applied to extract nonlinear higher-order features from functional magnetic resonance imaging (fMRI) data. These features were subsequently constructed into a hypergraph, leveraging the complex interactions captured in the data. Finally, a four-dimensional (4D) spatiotemporal hypergraph convolutional neural network model was constructed based on the fMRI data for classification. Experimental results on the Alzheimer's Disease Neuroimaging Initiative (ADNI) database demonstrated that the proposed framework outperformed the Hyper Graph Convolutional Network (HyperGCN) framework by 8% and traditional two-dimensional (2D) linear feature extraction methods by 12% in the AD/normal control (NC) classification task. In conclusion, this framework demonstrates an improvement in AD classification compared to mainstream deep learning methods, providing valuable evidence for computer-assisted diagnosis of AD.

Non-local attention and multi-task learning based lung segmentation in chest X-ray / 生物医学工程学杂志

Precise segmentation of lung field is a crucial step in chest radiographic computer-aided diagnosis system. With the development of deep learning, fully convolutional network based models for lung field segmentation have achieved great effect but are poor at accurate identification of the boundary and preserving lung field consistency. To solve this problem, this paper proposed a lung segmentation algorithm based on non-local attention and multi-task learning. Firstly, an encoder-decoder convolutional network based on residual connection was used to extract multi-scale context and predict the boundary of lung. Secondly, a non-local attention mechanism to capture the long-range dependencies between pixels in the boundary regions and global context was proposed to enrich feature of inconsistent region. Thirdly, a multi-task learning to predict lung field based on the enriched feature was conducted. Finally, experiments to evaluate this algorithm were performed on JSRT and Montgomery dataset. The maximum improvement of Dice coefficient and accuracy were 1.99% and 2.27%, respectively, comparing with other representative algorithms. Results show that by enhancing the attention of boundary, this algorithm can improve the accuracy and reduce false segmentation.

Smartphone-Based Screening for Cardiovascular Diseases: A Trend?

Abstract Cardiovascular diseases are the leading cause of death in the world. People living in vulnerable and poor places such as slums, rural areas and remote locations have difficulty in accessing medical care and diagnostic tests. In addition, given the COVID-19 pandemic, we are witnessing an increase in the use of telemedicine and non-invasive tools for monitoring vital signs. These questions motivate us to write this point of view and to describe some of the main innovations used for non-invasive screening of heart diseases. Smartphones are widely used by the population and are perfect tools for screening cardiovascular diseases. They are equipped with camera, flashlight, microphone, processor, and internet connection, which allow optical, electrical, and acoustic analysis of cardiovascular phenomena. Thus, when using signal processing and artificial intelligence approaches, smartphones may have predictive power for cardiovascular diseases. Here we present different smartphone approaches to analyze signals obtained from various methods including photoplethysmography, phonocardiograph, and electrocardiography to estimate heart rate, blood pressure, oxygen saturation (SpO2), heart murmurs and electrical conduction. Our objective is to present innovations in non-invasive diagnostics using the smartphone and to reflect on these trending approaches. These could help to improve health access and the screening of cardiovascular diseases for millions of people, particularly those living in needy areas.

Design and implementation for portable ultrasound-aided breast cancer screening system / 生物医学工程学杂志

Early screening is an important means to reduce breast cancer mortality. In order to solve the problem of low breast cancer screening rates caused by limited medical resources in remote and impoverished areas, this paper designs a breast cancer screening system aided with portable ultrasound Clarius. The system automatically segments the tumor area of the B-ultrasound image on the mobile terminal and uses the ultrasound radio frequency data on the cloud server to automatically classify the benign and malignant tumors. Experimental results in this study show that the accuracy of breast tumor segmentation reaches 98%, and the accuracy of benign and malignant classification reaches 82%, and the system is accurate and reliable. The system is easy to set up and operate, which is convenient for patients in remote and poor areas to carry out early breast cancer screening. It is beneficial to objectively diagnose disease, and it is the first time for the domestic breast cancer auxiliary screening system on the mobile terminal.

Artificial intelligence-assisted colonoscopy: a narrative review of current data and clinical applications

Colonoscopy is the reference standard procedure for the prevention and diagnosis of colorectal cancer, which is a leading cause of cancer-related deaths in Singapore. Artificial intelligence systems are automated, objective and reproducible. Artificial intelligence-assisted colonoscopy has recently been introduced into clinical practice as a clinical decision support tool. This review article provides a summary of the current published data and discusses ongoing research and current clinical applications of artificial intelligence-assisted colonoscopy.

Desenvolvimento e propriedades psicométricas do Teste Informatizado de Avaliação das Funções Executivas / Development and psychometric properties of the Computerized Executive Functions Assessment Test / Desarrollo y propiedades psicométricas del Test Informatizado de Evaluación de las Funciones Ejecutivas

Funções executivas (FE) são habilidades que permitem o autocontrole comportamental e cognitivo e estão relacionadas a diversos desfechos ao longo da vida. O uso de testes informatizados para avaliar as FE pode facilitar a precisão dos registros, a padronização e a análise dos dados. Este estudo objetivou desenvolver um instrumento informatizado para avaliar FE em crianças de 4 a 10 anos e analisar características psicométricas. Foram conduzidas cinco etapas: 1) Definição teórica e metodológica; 2) Construção dos itens; 3) Estudo piloto; 4) Análise de juízes; e 5) Estudos psicométricos de validade e fidedignidade. As tarefas informatizadas mostraram-se adequadas para o público-alvo, conforme avaliação dos juízes. As diferentes tarefas de memória de trabalho, inibição e flexibilidade cognitiva apresentaram correlações significativas entre si e a maioria das medidas no teste-reteste evidenciou estabilidade na mensuração. Portanto, os resultados sugerem viabilidade para uso do instrumento no contexto brasileiro. (AU)

Executive functions (EF) are skills linked to behavioral and cognitive self-control and are related to various outcomes throughout life. The use of computerized tests to evaluate EFs can facilitate the accuracy of records, standardization and data analysis. This study aimed to develop a computerized instrument for the EF assessment of children aged 4 to 10 years, and to seek psychometric evidence. Five steps were carried out: 1) Theoretical and methodological definition; 2) Construction of the items; 3) Pilot study; 4) Analysis of experts; and 5) Psychometric studies of validity and reliability. The computerized tasks proved to be suitable for the target audience according to the expert's evaluation. The results between the different tasks of working memory, inhibition and cognitive flexibility showed significant correlations and most test-retest measures showed stability in the measurement. Therefore, the results indicate the feasibility of using the instrument in the Brazilian context. (AU)

Funciones ejecutivas (FE) son habilidades que permiten el autocontrol conductual y cognitivo y están relacionadas con diversos resultados a lo largo de la vida. El uso de tests informatizados para evaluar las FE puede facilitar la precisión de los registros, la estandarización y el análisis de datos. Este estudio tuvo como objetivo desarrollar un instrumento informatizado para la FE para niños de 4 a 10 años, y analizar evidencias psicométricas. Fueron ejecutados cinco pasos: 1) Definición teórica y metodológica; 2) Construcción de los ítems; 3) Estudio piloto; 4) Análisis de jueces; y 5) Estudios psicométricos de validez y fiabilidad. Las tareas informatizadas demostraron ser adecuadas para el público objetivo según la evaluación de los jueces. Las diferentes tareas de memoria de trabajo, inhibición y flexibilidad cognitiva mostraron correlaciones significativas entre sí y la mayoría de las medidas test-retest presentaron estabilidad en la medición. Por lo tanto, los resultados sugieren la viabilidad del instrumento para el contexto brasileño. AU)

Research on computer aided diagnosis of otitis media based on faster region convolutional neural network / 生物医学工程学杂志

Otitis media is one of the common ear diseases, and its accurate diagnosis can prevent the deterioration of conductive hearing loss and avoid the overuse of antibiotics. At present, the diagnosis of otitis media mainly relies on the doctor's visual inspection based on the images fed back by the otoscope equipment. Due to the quality of otoscope equipment pictures and the doctor's diagnosis experience, this subjective examination has a relatively high rate of misdiagnosis. In response to this problem, this paper proposes the use of faster region convolutional neural networks to analyze clinically collected digital otoscope pictures. First, through image data enhancement and preprocessing, the number of samples in the clinical otoscope dataset was expanded. Then, according to the characteristics of the otoscope picture, the convolutional neural network was selected for feature extraction, and the feature pyramid network was added for multi-scale feature extraction to enhance the detection ability. Finally, a faster region convolutional neural network with anchor size optimization and hyperparameter adjustment was used for identification, and the effectiveness of the method was tested through a randomly selected test set. The results showed that the overall recognition accuracy of otoscope pictures in the test samples reached 91.43%. The above studies show that the proposed method effectively improves the accuracy of otoscope picture classification, and is expected to assist clinical diagnosis.

Application of multiple empirical kernel mapping ensemble classifier based on self-paced learning in ultrasound-based computer-aided diagnosis for breast cancer / 生物医学工程学杂志

Both feature representation and classifier performance are important factors that determine the performance of computer-aided diagnosis (CAD) systems. In order to improve the performance of ultrasound-based CAD for breast cancers, a novel multiple empirical kernel mapping (MEKM) exclusivity regularized machine (ERM) ensemble classifier algorithm based on self-paced learning (SPL) is proposed, which simultaneously promotes the performance of both feature representation and the classifier. The proposed algorithm first generates multiple groups of features by MEKM to enhance the ability of feature representation, which also work as the kernel transform in multiple support vector machines embedded in ERM. The SPL strategy is then adopted to adaptively select samples from easy to hard so as to gradually train the ERM classifier model with improved performance. This algorithm is verified on a B-mode ultrasound dataset and an elastography ultrasound dataset, respectively. The results show that the classification accuracy, sensitivity and specificity on B-mode ultrasound are (86.36±6.45)%, (88.15±7.12)%, and (84.52±9.38)%, respectively, and the classification accuracy, sensitivity and specificity on elastography ultrasound are (85.97±3.75)%, (85.93±6.09)%, and (86.03±5.88)%, respectively. It indicates that the proposed algorithm can effectively improve the performance of ultrasound-based CAD for breast cancers with the potential for application.

Herramienta para la detección automática de nódulos pulmonares solitarios en series de imágenes de tomografía computarizada multicorte / A tool for automated detection of solitary pulmonary nodules in series of multicut computerized tomography images

Introducción: el nódulo pulmonar solitario es uno de los problemas más frecuentes en la práctica del radiólogo, que constituye un hallazgo incidental habitual en los estudios torácicos realizados durante el ejercicio clínico diario. Objetivo: implementar un sistema de diagnóstico asistido por computadora que facilite la detección del nódulo pulmonar solitario en las series de imágenes de tomografía computarizada multicorte. Métodos: se utilizó Matlab para el desarrollo y evaluación de un conjunto de algoritmos que constituyen elementos necesarios de un sistema de diagnóstico asistido por computadora. En orden: un algoritmo para la extracción de las regiones de interés, algoritmo para la extracción de características y un algoritmo de detección de nódulo pulmonar solitario para el cual se probaron varios clasificadores. La evaluación de los algoritmos fue efectuada en base a las anotaciones realizada por especialistas a la colección de imágenes LIDC-IDRI (Lung Image Database Consortium). Resultados: el método de segmentación empleado para extracción de las regiones de interés permitió generar la adecuada división de las imágenes originales en regiones significativas. El algoritmo utilizado en la detección mostró para el conjunto de prueba además de buena exactitud (de 96,4 por ciento), un buen balance de sensibilidad (91,5 por ciento) para una tasa de 0,84 falsos positivos por imagen. Conclusiones: el trabajo de investigación y la implementación realizada se reflejan en la construcción de una interfaz gráfica en Matlab como prototipo del sistema de diagnóstico asistido por computadora, con el que se puede contribuir a detectar más fácilmente el NPS(AU)

Introduction: solitary pulmonary nodules are one of the most frequent problems in radiographic practice. They are a common incidental finding in chest studies conducted during routine clinical work. Objective: implement a computer-assisted diagnostic system facilitating detection of solitary pulmonary nodules in multicut computerized tomography image series. Methods: Matlab was used to develop and evaluate a set of algorithms constituting necessary components of a computer-assisted diagnostic system. The order was the following: an algorithm to extract regions of interest, another to extract characteristics, and another to detect solitary pulmonary nodules, for which several classifiers were tested. Evaluation of the algorithms was based on notes taken by specialists on the LIDC-IDRI (Lung Image Database Consortium) image collection. Results: the segmentation method used for extraction of regions of interest made it possible to create a suitable division of the original images into significant regions. The algorithm used for detection found that the test set exhibited good accuracy (96.4%), a good sensitivity balance (91.5%), and a 0.84 rate of false positives per image. Conclusions: the research and implementation work done is reflected in the construction of a Matlab graphic interface serving as a prototype for a computer-assisted diagnostic system which may facilitate detection of SPNs.

Research status and trend of artificial intelligence in the diagnosis of urinary diseases / 生物医学工程学杂志

Recently, artificial intelligence (AI) has been widely applied in the diagnosis and treatment of urinary diseases with the development of data storage, image processing, pattern recognition and machine learning technologies. Based on the massive biomedical big data of imaging and histopathology, many urinary system diseases (such as urinary tumor, urological calculi, urinary infection, voiding dysfunction and erectile dysfunction) will be diagnosed more accurately and will be treated more individualizedly. However, most of the current AI diagnosis and treatment are in the pre-clinical research stage, and there are still some difficulties in the wide application of AI. This review mainly summarizes the recent advances of AI in the diagnosis of prostate cancer, bladder cancer, kidney cancer, urological calculi, frequent micturition and erectile dysfunction, and discusses the future potential and existing problems.

Study on classification and identification of depressed patients and healthy people among adolescents based on optimization of brain characteristics of network / 生物医学工程学杂志

To enhance the accuracy of computer-aided diagnosis of adolescent depression based on electroencephalogram signals, this study collected signals of 32 female adolescents (16 depressed and 16 healthy, age: 16.3 ± 1.3) with eyes colsed for 4 min in a resting state. First, based on the phase synchronization between the signals, the phase-locked value (PLV) method was used to calculate brain functional connectivity in the θ and α frequency bands, respectively. Then based on the graph theory method, the network parameters, such as strength of the weighted network, average characteristic path length, and average clustering coefficient, were calculated separately (

Development of Clinical Information Navigation System Based on 3D Human Model / 中国医疗器械杂志

A clinical information navigation system based on 3D human body model is designed. The system extracts the key information of diagnosis and treatment of patients by searching the historical medical records, and stores the focus information in a predefined structured patient instance. In addition, the rule mapping is established between the patient instance and the three-dimensional human body model, the focus information is visualized on the three-dimensional human body model, and the trend curve can be drawn according to the change of the focus, meanwhile, the key diagnosis and treatment information and the original report reference function are provided. The system can support the analysis, storage and visualization of various types of reports, improve the efficiency of doctors' retrieval of patient information, and reduce the treatment time.

Automatic segmentation of brain tumors in magnetic resonance imaging / Segmentação automática de tumores cerebrais em imagens de ressonância magnética

ABSTRACT Objective To develop a computational algorithm applied to magnetic resonance imaging for automatic segmentation of brain tumors. Methods A total of 130 magnetic resonance images were used in the T1c, T2 and FSPRG T1C sequences and in the axial, sagittal and coronal planes of patients with brain cancer. The algorithms employed contrast correction, histogram normalization and binarization techniques to disconnect adjacent structures from the brain and enhance the region of interest. Automatic segmentation was performed through detection by coordinates and arithmetic mean of the area. Morphological operators were used to eliminate undesirable elements and reconstruct the shape and texture of the tumor. The results were compared with manual segmentations by two radiologists to determine the efficacy of the algorithms implemented. Results The correlated correspondence between the segmentation obtained and the gold standard was 89.23%. Conclusion It is possible to locate and define the tumor region automatically with no the need for user interaction, based on two innovative methods to detect brain extreme sites and exclude non-tumor tissues on magnetic resonance images.

RESUMO Objetivo Desenvolver um algoritmo computacional aplicado a imagens de ressonância magnética, para segmentação automática de tumores cerebrais. Métodos Foram utilizadas 130 imagens de ressonância magnética nas sequências T1c, T2 e FSPRG T1c e nos planos axial, sagital e coronal de pacientes acometidos com câncer cerebral. Os algoritmos empregaram técnicas de correção de contraste, normalização de histograma e binarização, para desconectar estruturas adjacentes do cérebro e realçar a região de interesse. A segmentação automática foi realizada por meio da detecção por coordenadas e por média aritmética da área. Operadores morfológicos foram utilizados para eliminar elementos indesejáveis e reconstruir a forma e a textura do tumor. Os resultados foram comparados com as segmentações manuais de dois médicos radiologistas, para determinar a eficácia dos algoritmos implementados. Resultados Os acertos foram de 89,23% na correspondência entre a segmentação obtida e o padrão-ouro. Conclusão É possível localizar e delimitar a região tumoral de forma automática, sem necessidade de interação com o usuário baseado em dois métodos inovadores de detecção dos extremos do cérebro e de exclusão dos tecidos não tumorais em imagens de ressonância magnética.

Machine learning techniques for computer-aided classification of active inflammatory sacroiliitis in magnetic resonance imaging

Abstract Background: Currently, magnetic resonance imaging (MRI) is used to evaluate active inflammatory sacroiliitis related to axial spondyloarthritis (axSpA). The qualitative and semiquantitative diagnosis performed by expert radiologists and rheumatologists remains subject to significant intrapersonal and interpersonal variation. This encouraged us to use machine-learning methods for this task. Methods: In this retrospective study including 56 sacroiliac joint MRI exams, 24 patients had positive and 32 had negative findings for inflammatory sacroiliitis according to the ASAS group criteria. The dataset was randomly split with ∼ 80% (46 samples, 20 positive and 26 negative) as training and ∼ 20% as external test (10 samples, 4 positive and 6 negative). After manual segmentation of the images by a musculoskeletal radiologist, multiple features were extracted. The classifiers used were the Support Vector Machine, the Multilayer Perceptron (MLP), and the Instance-Based Algorithm, combined with the Relief and Wrapper methods for feature selection. Results: Based on 10-fold cross-validation using the training dataset, the MLP classifier obtained the best performance with sensitivity = 100%, specificity = 95.6% and accuracy = 84.7%, using 6 features selected by the Wrapper method. Using the test dataset (external validation) the same MLP classifier obtained sensitivity = 100%, specificity = 66.7% and accuracy = 80%. Conclusions: Our results show the potential of machine learning methods to identify SIJ subchondral bone marrow edema in axSpA patients and are promising to aid in the detection of active inflammatory sacroiliitis on MRI STIR sequences. Multilayer Perceptron (MLP) achieved the best results.(AU)

Evaluation of the color vision acuity pattern of undergraduates of health courses in a Brazilian university / Avaliação do padrão de acuidade visual para cores de acadêmicos da área de saúde em Universidade brasileira

ABSTRACT Objective: The goal of the study is to analyze the color vision acuity pattern in undergraduates of health courses and to discuss the impact of these diseases in this population. Color deficiencies interfere significantly in the daily routine of professionals in the health area who need to discern different color hues in several situations of their everyday practice. Methods: Sixty-four volunteers, undergraduates of health courses of the Federal University of Alfenas (UNIFAL-MG), participated in the study. One man was excluded because he did not fit the inclusion criteria. Two groups were analyzed according to sex with the Farnsworth Munsell 100-Hue test. Results: There were no significant differences between the eyes and between the groups analyzed. The color vision acuity pattern is between 35 and 40, according to the Total Error Score. The gender issue does not influence the general pattern of the color vision acuity of the health courses undergraduates when those with color vision disorders are removed. Conclusion: Screenings and guidance should be given to undergraduates of health courses so that, aware of their condition of presenting some type of color disorder, they shall make the appropriate decision on which career to follow so that such limitation does not interfere with the quality of their daily life.

RESUMO Objetivo: O objetivo do estudo é analisar a acuidade visual média para cores de estudantes da área de saúde e discutir o impacto das doenças que a afetam nessa população. Deficiências cromáticas interferem de forma significativa no dia a dia de profissionais da área da saúde que necessitam de discernir diferentes matizes em diversas situações de sua prática profissional. Métodos: Participaram da pesquisa 64 voluntários, estudantes de cursos da área de saúde da Universidade Federal de Alfenas, sendo que 1 homem foi excluído por não se adequar aos critérios de inclusão. Dois grupos foram analisados, de acordo com o sexo, com o teste de Farnsworth Munsell 100-Hue. Resultados: Não houve diferenças significativas entre os olhos e entre os grupos analisados. O padrão de visão de cores encontra-se entre 35 e 40, de acordo com a Pontuação do Erro Total. A questão de gênero não influencia no padrão geral da qualidade de visão de cores de estudantes da área de saúde, quando retirados aqueles que apresentam distúrbios da visão cromática. Conclusão: Devem ser realizadas triagens e orientação para estudantes de cursos da área de saúde para que, cientes da sua condição de apresentar algum tipo de distúrbio cromático, possam tomar a decisão adequada sobre qual carreira seguir para que tal limitação não interfira na qualidade de sua vida diária.

Overview of Deep Learning in Gastrointestinal Endoscopy

Artificial intelligence is likely to perform several roles currently performed by humans, and the adoption of artificial intelligence-based medicine in gastroenterology practice is expected in the near future. Medical image-based diagnoses, such as pathology, radiology, and endoscopy, are expected to be the first in the medical field to be affected by artificial intelligence. A convolutional neural network, a kind of deep-learning method with multilayer perceptrons designed to use minimal preprocessing, was recently reported as being highly beneficial in the field of endoscopy, including esophagogastroduodenoscopy, colonoscopy, and capsule endoscopy. A convolutional neural network-based diagnostic program was challenged to recognize anatomical locations in esophagogastroduodenoscopy images, Helicobacter pylori infection, and gastric cancer for esophagogastroduodenoscopy; to detect and classify colorectal polyps; to recognize celiac disease and hookworm; and to perform small intestine motility characterization of capsule endoscopy images. Artificial intelligence is expected to help endoscopists provide a more accurate diagnosis by automatically detecting and classifying lesions; therefore, it is essential that endoscopists focus on this novel technology. In this review, we describe the effects of artificial intelligence on gastroenterology with a special focus on automatic diagnosis, based on endoscopic findings.

Establishment of a deep feature-based classification model for distinguishing benign and malignant breast tumors on full-filed digital mammography / 南方医科大学学报

OBJECTIVE@#To develop a deep features-based model to classify benign and malignant breast lesions on full- filed digital mammography.@*METHODS@#The data of full-filed digital mammography in both craniocaudal view and mediolateral oblique view from 106 patients with breast neoplasms were analyzed. Twenty-three handcrafted features (HCF) were extracted from the images of the breast tumors and a suitable feature set of HCF was selected using -test. The deep features (DF) were extracted from the 3 pre-trained deep learning models, namely AlexNet, VGG16 and GoogLeNet. With abundant breast tumor information from the craniocaudal view and mediolateral oblique view, we combined the two extracted features (DF and HCF) as the two-view features. A multi-classifier model was finally constructed based on the combined HCF and DF sets. The classification ability of different deep learning networks was evaluated.@*RESULTS@#Quantitative evaluation results showed that the proposed HCF+DF model outperformed HCF model, and AlexNet produced the best performances among the 3 deep learning models.@*CONCLUSIONS@#The proposed model that combines DF and HCF sets of breast tumors can effectively distinguish benign and malignant breast lesions on full-filed digital mammography.