RESUMEN
Breast cancer analysis implies that radiologists inspect mammograms to detect suspicious breast lesions and identify mass tumors. Artificial intelligence techniques offer automatic systems for breast mass segmentation to assist radiologists in their diagnosis. With the rapid development of deep learning and its application to medical imaging challenges, UNet and its variations is one of the state-of-the-art models for medical image segmentation that showed promising performance on mammography. In this paper, we propose an architecture, called Connected-UNets, which connects two UNets using additional modified skip connections. We integrate Atrous Spatial Pyramid Pooling (ASPP) in the two standard UNets to emphasize the contextual information within the encoder-decoder network architecture. We also apply the proposed architecture on the Attention UNet (AUNet) and the Residual UNet (ResUNet). We evaluated the proposed architectures on two publically available datasets, the Curated Breast Imaging Subset of Digital Database for Screening Mammography (CBIS-DDSM) and INbreast, and additionally on a private dataset. Experiments were also conducted using additional synthetic data using the cycle-consistent Generative Adversarial Network (CycleGAN) model between two unpaired datasets to augment and enhance the images. Qualitative and quantitative results show that the proposed architecture can achieve better automatic mass segmentation with a high Dice score of 89.52%, 95.28%, and 95.88% and Intersection over Union (IoU) score of 80.02%, 91.03%, and 92.27%, respectively, on CBIS-DDSM, INbreast, and the private dataset.
RESUMEN
Thoracic pain is a shared symptom among gastrointestinal diseases, muscle pain, emotional disorders, and the most deadly: Cardiovascular diseases. Due to the limited space in the emergency department, it is important to identify when thoracic pain is of cardiac origin, since being a symptom of CVD (Cardiovascular Disease), the attention to the patient must be immediate to prevent irreversible injuries or even death. Artificial intelligence contributes to the early detection of pathologies, such as chest pain. In this study, the machine learning techniques were used, performing an analysis of 27 variables provided by a database with information from 258 geriatric patients with 60 years old average age from Medical Norte Hospital in Tijuana, Baja California, Mexico. The objective of this analysis is to determine which variables are correlated with thoracic pain of cardiac origin and use the results as secondary parameters to evaluate the thoracic pain in the emergency rooms, and determine if its origin comes from a CVD or not. For this, two machine learning techniques were used: Tree classification and cross-validation. As a result, the Logistic Regression model, using the characteristics proposed as second factors to consider as variables, obtained an average accuracy (µ) of 96.4% with a standard deviation (σ) of 2.4924, while for F1 a mean (µ) of 91.2% and a standard deviation (σ) of 6.5640. This analysis suggests that among the main factors related to cardiac thoracic pain are: Dyslipidemia, diabetes, chronic kidney failure, hypertension, smoking habits, and troponin levels at the time of admission, which is when the pain occurs. Considering dyslipidemia and diabetes as the main variables due to similar results with machine learning techniques and statistical methods, where 61.95% of the patients who suffer an Acute Myocardial Infarction (AMI) have diabetes, and the 71.73% have dyslipidemia.
Asunto(s)
Inteligencia Artificial , Dolor en el Pecho , Anciano , Dolor en el Pecho/epidemiología , Dolor en el Pecho/etiología , Servicio de Urgencia en Hospital , Humanos , Aprendizaje Automático , México/epidemiología , Persona de Mediana EdadRESUMEN
The article presents a study based on timeline data analysis of the level of sarcopenia in older patients in Baja California, Mexico. Information was examined at the beginning of the study (first event), three months later (second event), and six months later (third event). Sarcopenia is defined as the loss of muscle mass quality and strength. The study was conducted with 166 patients. A total of 65% were women and 35% were men. The mean age of the enrolled patients was 77.24 years. The research included 99 variables that consider medical history, pharmacology, psychological tests, comorbidity (Charlson), functional capacity (Barthel and Lawton), undernourishment (mini nutritional assessment (MNA) validated test), as well as biochemical and socio-demographic data. Our aim was to evaluate the prevalence of the level of sarcopenia in a population of chronically ill patients assessed at the Tijuana General Hospital. We used machine learning techniques to assess and identify the determining variables to focus on the patients' evolution. The following classifiers were used: Support Vector Machines, Linear Support Vector Machines, Radial Basis Function, Gaussian process, Decision Tree, Random Forest, multilayer perceptron, AdaBoost, Gaussian Naive Bayes, and Quadratic Discriminant Analysis. In order of importance, we found that the following variables determine the level of sarcopenia: Age, Systolic arterial hypertension, mini nutritional assessment (MNA), Number of chronic diseases, and Sodium. They are therefore considered relevant in the decision-making process of choosing treatment or prevention. Analysis of the relationship between the presence of the variables and the classifiers used to measure sarcopenia revealed that the Decision Tree classifier, with the Age, Systolic arterial hypertension, MNA, Number of chronic diseases, and Sodium variables, showed a precision of 0.864, accuracy of 0.831, and an F1 score of 0.900 in the first and second events. Precision of 0.867, accuracy of 0.825, and an F1 score of 0.867 were obtained in event three with the same variables. We can therefore conclude that the Decision Tree classifier yields the best results for the assessment of the determining variables and suggests that the study population's sarcopenia did not change from moderate to severe.
Asunto(s)
Sarcopenia , Anciano , Teorema de Bayes , Femenino , Fuerza de la Mano , Humanos , Aprendizaje Automático , Masculino , México/epidemiología , Prevalencia , Sarcopenia/diagnóstico , Sarcopenia/epidemiologíaRESUMEN
This paper presents a study based on data analysis of the sarcopenia level in older adults. Sarcopenia is a prevalent pathology in adults of around 50 years of age, whereby the muscle mass decreases by 1 to 2% a year, and muscle strength experiences an annual decrease of 1.5% between 50 and 60 years of age, subsequently increasing by 3% each year. The World Health Organisation estimates that 5-13% of individuals of between 60 and 70 years of age and 11-50% of persons of 80 years of age or over have sarcopenia. This study was conducted with 166 patients and 99 variables. Demographic data was compiled including age, gender, place of residence, schooling, marital status, level of education, income, profession, and financial support from the State of Baja California, and biochemical parameters such as glycemia, cholesterolemia, and triglyceridemia were determined. A total of 166 patients took part in the study, with an average age of 77.24 years. The purpose of the study was to provide an automatic classifier of sarcopenia level in older adults using artificial intelligence in addition to identifying the weight of each variable used in the study. We used machine learning techniques in this work, in which 10 classifiers were employed to assess the variables and determine which would provide the best results, namely, Nearest Neighbors (3), Linear SVM (Support Vector Machines) (C = 0.025), RBF (Radial Basis Function) SVM (gamma = 2, C = 1), Gaussian Process (RBF (1.0)), Decision Tree (max_depth = 3), Random Forest (max_depth=3, n_estimators = 10), MPL (Multilayer Perceptron) (alpha = 1), AdaBoost, Gaussian Naive Bayes, and QDA (Quadratic Discriminant Analysis). Feature selection determined by the mean for the variable ranking suggests that Age, Systolic Arterial Hypertension (HAS), Mini Nutritional Assessment (MNA), Number of chronic diseases (ECNumber), and Sodium are the five most important variables in determining the sarcopenia level, and are thus of great importance prior to establishing any treatment or preventive measure. Analysis of the relationships existing between the presence of the variables and classifiers used in moderate and severe sarcopenia revealed that the sarcopenia level using the RBF SVM classifier with Age, HAS, MNA, ECNumber, and Sodium variables has 82'5 accuracy, a 90'2 F1, and 82'8 precision.