Autism Spectrum Disorder is a mental disorder that afflicts millions of people worldwide. It is estimated that one in 160 children has traces of autism, with five times the higher prevalence in boys. The protocols for detecting symptoms are diverse. However, the following are among the most used: the Diagnostic and Statistical Manual of Mental Disorders, 5th Edition (DSM-5), of the American Psychiatric Association; the Revised Autistic Diagnostic Observation Schedule (ADOS-R); the Autistic Diagnostic Interview (ADI); and the International Classification of Diseases, 10th edition (ICD-10), published by the World Health Organization (WHO) and adopted in Brazil by the Unified Health System (SUS). The application of machine learning models helps make the diagnostic process of Autism Spectrum Disorder more precise, reducing, in many cases, the number of criteria necessary for evaluation, denoting a form of attribute engineering (feature engineering) efficiency. This work proposes a hybrid approach based on machine learning algorithms’ composition to discover knowledge and concepts associated with the multicriteria method of decision support based on Verbal Decision Analysis to refine the results. Therefore, the study has the general objective of evaluating how the mentioned hybrid methodology proposal can make the protocol derived from ICD-10 more efficient, providing agility to diagnosing Autism Spectrum Disorder by observing a minor symptom. The study database covers thousands of cases of people who, once diagnosed, obtained government assistance in Brazil.
Considering that psychological disorders have specific causes and symptoms, the conventional way to establish diagnosis consists of analysing the behavioural reactions of a human being to events in their daily lives. The manifestation of psychological disorders varies from person to person and the degree of severity. Like ‘personality,’ ‘consciousness’ and ‘intelligence,’ the expression ‘abnormal behaviour’ is difficult to define due to sociocultural subjectivities intrinsic to the human being. Social interactions are essential factors in the search for solutions to the problems of the information society. However, neurodevelopmental disorders may affect some people from birth, hampering their healthy growth and consequently their participation in social interactions. In this context, autistic spectrum disorder is one of the primary disorders of neurodevelopment, which manifests itself before the age of three. Despite the existence of diagnostic criteria for autism spectrum disorder, there is a lack of system with models containing algorithms that help the families and health professionals in the early diagnosis of this disorder. The objective of this study is to make proactive the decision-making process for the establishment of the early diagnosis of autism spectrum disorder. For this, the research presents a hybrid model proposal composed of a specialist system structured in decision-making methodologies (Multi-Criteria Decision Analysis) and structured representations of knowledge in production rules and probabilities using Artificial Intelligence. In this context, the research aims to show one of the potentials of information technologies as a way to generate knowledge that adds value and serves as a support in medical decision making.
Predictive modelling strategies can optimise the clinical diagnostic process by identifying patterns among various symptoms and risk factors, such as those presented in cases of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), also known as coronavirus (COVID-19). In this context, the present research proposes a comparative analysis using benchmarking techniques to evaluate and validate the performance of some classification algorithms applied to the same dataset, which contains information collected from patients diagnosed with COVID-19, registered in the Influenza Epidemiological Surveillance System (SIVEP). With this approach, 30,000 cases were analysed during the training and testing phase of the prediction models. This work proposes a comparative approach of machine learning algorithms (ML), working on the knowledge discovery task to predict clinical evolution in patients diagnosed with COVID-19. Our experiments show, through appropriate metrics, that the clinical evolution classification process of patients diagnosed with COVID-19 using the Multilayer Perceptron algorithm performs well against other ML algorithms. Its use has significant consequences for vital prognosis and agility in measures used in the first consultations in hospitals.
2 March 2015 in Computational and Mathematical Methods in Medicine
Psychotics disorders, most commonly known as schizophrenia, have incapacitated professionals in different sectors of activities. Those disorders have caused damage in a microlevel to the individual and his/her family and in a macrolevel to the economic and production system of the country. The lack of early and sometimes very late diagnosis has provided reactive measures, when the professional is already showing psychological signs of incapacity to work. This study aims to help the early diagnosis of psychotics’ disorders with a hybrid proposal of an expert system that is integrated to structured methodologies in decision support (multicriteria decision analysis: MCDA) and knowledge structured representations into production rules and probabilities (artificial intelligence: AI).