Fuzzy linguistic model for evaluating the risk of neonatal death

The past, current, and future of neonatal intensive care units with artificial intelligence: a systematic review

Machine learning and deep learning are two subsets of artificial intelligence that involve teaching computers to learn and make decisions from any sort of data. Most recent developments in artificial intelligence are coming from deep learning, which has proven revolutionary in almost all fields, from computer vision to health sciences. The effects of deep learning in medicine have changed the conventional ways of clinical application significantly. Although some sub-fields of medicine, such as pediatrics, have been relatively slow in receiving the critical benefits of deep learning, related research in pediatrics has started to accumulate to a significant level, too. Hence, in this paper, we review recently developed machine learning and deep learning-based solutions for neonatology applications. We systematically evaluate the roles of both classical machine learning and deep learning in neonatology applications, define the methodologies, including algorithmic developments, and describe the remaining challenges in the assessment of neonatal diseases by using PRISMA 2020 guidelines. To date, the primary areas of focus in neonatology regarding AI applications have included survival analysis, neuroimaging, analysis of vital parameters and biosignals, and retinopathy of prematurity diagnosis. We have categorically summarized 106 research articles from 1996 to 2022 and discussed their pros and cons, respectively. In this systematic review, we aimed to further enhance the comprehensiveness of the study. We also discuss possible directions for new AI models and the future of neonatology with the rising power of AI, suggesting roadmaps for the integration of AI into neonatal intensive care units.

Fuzzy model to estimate the number of hospitalizations for asthma and pneumonia under the effects of air pollution

OBJECTIVE

Predict the number of hospitalizations for asthma and pneumonia associated with exposure to air pollutants in the city of São José dos Campos, São Paulo State.

METHODS

This is a computational model using fuzzy logic based on Mamdani’s inference method. For the fuzzification of the input variables of particulate matter, ozone, sulfur dioxide and apparent temperature, we considered two relevancy functions for each variable with the linguistic approach: good and bad. For the output variable number of hospitalizations for asthma and pneumonia, we considered five relevancy functions: very low, low, medium, high and very high. DATASUS was our source for the number of hospitalizations in the year 2007 and the result provided by the model was correlated with the actual data of hospitalization with lag from zero to two days. The accuracy of the model was estimated by the ROC curve for each pollutant and in those lags.

RESULTS

In the year of 2007, 1,710 hospitalizations by pneumonia and asthma were recorded in São José dos Campos, State of São Paulo, with a daily average of 4.9 hospitalizations (SD = 2.9). The model output data showed positive and significant correlation (r = 0.38) with the actual data; the accuracies evaluated for the model were higher for sulfur dioxide in lag 0 and 2 and for particulate matter in lag 1.

CONCLUSIONS

Fuzzy modeling proved accurate for the pollutant exposure effects and hospitalization for pneumonia and asthma approach.

Developing a Fuzzy Expert System to Predict the Risk of Neonatal Death

Introduction:

This study aims at developing a fuzzy expert system to predict the possibility of neonatal death.

Materials and Methods:

A questionnaire was given to Iranian neonatologists and the more important factors were identified based on their answers. Then, a computing model was designed considering the fuzziness of variables having the highest neonatal mortality risk. The inference engine used was Mamdani’s method and the output was the risk of neonatal death given as a percentage. To validate the designed system, neonates’ medical records real data at a Tehran hospital were used. MATLAB software was applied to build the model, and user interface was developed by C# programming in Visual Studio platform as bilingual (English and Farsi user interface).

Results:

According to the results, the accuracy, sensitivity, and specificity of the model were 90%, 83% and 97%, respectively.

Conclusion:

The designed fuzzy expert system for neonatal death prediction showed good accuracy as well as proper specificity, and could be utilized in general hospitals as a clinical decision support tool.

Sistema de apoio à avaliação de impactos da poluição sonora sobre a saúde pública

Resumo: O objetivo do trabalho foi desenvolver um sistema de apoio à avaliação da poluição sonora, aplicado na zona central de Rio Claro, São Paulo, Brasil. Para isso, dados foram obtidos por meio de medições sonoras e entrevistas com a população, gerando como indicadores o nível sonoro equivalente (Leq ), o índice de ruído de tráfego (LTNI ) e um diagnóstico participativo (Dp ), integrados por intermédio de um sistema de inferência fuzzy (SIF). Como resultado, o sistema proposto permitiu classificar os pontos avaliados quanto ao grau de impacto da poluição sonora sobre a saúde da população (IPS ) na área de estudo, que pode ser considerado significativo em 31,4% dos pontos e muito significativo em 62,9%. A possibilidade de adequar o SIF de acordo com as condições de estudo viabiliza a sua generalização e, desta forma, apoia a avaliação e respectiva gestão do ruído ambiental em outras regiões.

Fuzzy model approach for estimating time of hospitalization due to cardiovascular diseases

A fuzzy linguistic model based on the Mamdani method with input variables, particulate matter, sulfur dioxide, temperature and wind obtained from CETESB with two membership functions each was built to predict the average hospitalization time due to cardiovascular diseases related to exposure to air pollutants in São José dos Campos in the State of São Paulo in 2009. The output variable is the average length of hospitalization obtained from DATASUS with six membership functions. The average time given by the model was compared to actual data using lags of 0 to 4 days. This model was built using the Matlab v. 7.5 fuzzy toolbox. Its accuracy was assessed with the ROC curve. Hospitalizations with a mean time of 7.9 days (SD = 4.9) were recorded in 1119 cases. The data provided revealed a significant correlation with the actual data according to the lags of 0 to 4 days. The pollutant that showed the greatest accuracy was sulfur dioxide. This model can be used as the basis of a specialized system to assist the city health authority in assessing the risk of hospitalizations due to air pollutants.

Estimating outcomes in newborn infants using fuzzy logic

OBJECTIVE:

To build a linguistic model using the properties of fuzzy logic to estimate the risk of death of neonates admitted to a Neonatal Intensive Care Unit.

METHODS:

Computational model using fuzzy logic. The input variables of the model were birth weight, gestational age, 5^th-minute Apgar score and inspired fraction of oxygen in newborn infants admitted to a Neonatal Intensive Care Unit of Taubaté, Southeast Brazil. The output variable was the risk of death, estimated as a percentage. Three membership functions related to birth weight, gestational age and 5^th-minute Apgar score were built, as well as two functions related to the inspired fraction of oxygen; the risk presented five membership functions. The model was developed using the Mandani inference by means of Matlab^(r) software. The model values were compared with those provided by experts and their performance was estimated by ROC curve.

RESULTS:

100 newborns were included, and eight of them died. The model estimated an average possibility of death of 49.7±29.3%, and the possibility of hospital discharge was 24±17.5%. These values are different when compared by Student's t-test (p<0.001). The correlation test revealed r=0.80 and the performance of the model was 81.9%.

CONCLUSIONS:

This predictive, non-invasive and low cost model showed a good accuracy and can be applied in neonatal care, given the easiness of its use.

Fuzzy cognitive map in differential diagnosis of alterations in urinary elimination: a nursing approach

PURPOSE

To develop a decision support system to discriminate the diagnoses of alterations in urinary elimination, according to the nursing terminology of NANDA International (NANDA-I).

METHODS

A fuzzy cognitive map (FCM) was structured considering six possible diagnoses: stress urinary incontinence, reflex urinary incontinence, urge urinary incontinence, functional urinary incontinence, total urinary incontinence and urinary retention; and 39 signals associated with them. The model was implemented in Microsoft Visual C++(®) Edition 2005 and applied in 195 real cases. Its performance was evaluated through the agreement test, comparing its results with the diagnoses determined by three experts (nurses). The sensitivity and specificity of the model were calculated considering the expert's opinion as a gold standard. In order to compute the Kappa's values we considered two situations, since more than one diagnosis was possible: the overestimation of the accordance in which the case was considered as concordant when at least one diagnoses was equal; and the underestimation of the accordance, in which the case was considered as discordant when at least one diagnosis was different.

RESULTS

The overestimation of the accordance showed an excellent agreement (kappa=0.92, p<0.0001); and the underestimation provided a moderate agreement (kappa=0.42, p<0.0001). In general the FCM model showed high sensitivity and specificity, of 0.95 and 0.92, respectively, but provided a low specificity value in determining the diagnosis of urge urinary incontinence (0.43) and a low sensitivity value to total urinary incontinence (0.42).

CONCLUSIONS

The decision support system developed presented a good performance compared to other types of expert systems for differential diagnosis of alterations in urinary elimination. Since there are few similar studies in the literature, we are convinced of the importance of investing in this kind of modeling, both from the theoretical and from the health applied points of view.

LIMITATIONS

In spite of the good results, the FCM should be improved to identify the diagnoses of urge urinary incontinence and total urinary incontinence.

Establishing the risk of neonatal mortality using a fuzzy predictive model

The objective of this study was to develop a fuzzy model to estimate the possibility of neonatal mortality. A computing model was built, based on the fuzziness of the following variables: newborn birth weight, gestational age at delivery, Apgar score, and previous report of stillbirth. The inference used was Mamdani's method and the output was the risk of neonatal death given as a percentage. 24 rules were created according to the inputs. The validation model used a real data file with records from a Brazilian city. The receiver operating characteristic (ROC) curve was used to estimate the accuracy of the model, while average risks were compared using the Student t test. MATLAB 6.5 software was used to build the model. The average risks were smaller in survivor newborn (p < 0.001). The accuracy of the model was 0.90. The higher accuracy occurred with risk below 25%, corresponding to 0.70 in respect to sensitivity, 0.98 specificity, 0.99 negative predictive value and 0.22 positive predictive value. The model showed a good accuracy, as well as a good negative predictive value and could be used in general hospitals.

Model for differential nursing diagnosis of alterations in urinary elimination based on fuzzy logic

Nursing diagnoses associated with alterations of urinary elimination require different interventions. Nurses, who are not specialists, require support to diagnose and manage patients with disturbances of urine elimination. The aim of this study was to present a model based on fuzzy logic for differential diagnosis of alterations in urinary elimination, considering nursing diagnosis approved by the North American Nursing Diagnosis Association, 2001-2002. Fuzzy relations and the maximum-minimum composition approach were used to develop the system. The model performance was evaluated with 195 cases from the database of a previous study, resulting in 79.0% of total concordance and 19.5% of partial concordance, when compared with the panel of experts. Total discordance was observed in only three cases (1.5%). The agreement between model and experts was excellent (kappa = 0.98, P < .0001) or substantial (kappa = 0.69, P < .0001) when considering the overestimative accordance (accordance was considered when at least one diagnosis was equal) and the underestimative discordance (discordance was considered when at least one diagnosis was different), respectively. The model herein presented showed good performance and a simple theoretical structure, therefore demanding few computational resources.

The use of expert systems on the differential diagnosis of urinary incontinence

The differential diagnosis of urinary incontinence classes is sometimes difficult to establish. As a rule, only the results of urodynamic testing allow an accurate diagnosis. However, this exam is not always feasible, because it requires special equipment, and also trained personnel to lead and interpret the exam. Some expert systems have been developed to assist health professionals in this field. Therefore, the aims of this paper are to present the definition of Artificial Intelligence; to explain what expert system and system for decision support are and its application in the field of health and to discuss some expert systems for differential diagnosis of urinary incontinence. It is concluded that expert systems may be useful not only for teaching purposes, but also as decision support in daily clinical practice. Despite this, for several reasons, health professionals usually hesitate to use the computer expert system to support their decision making process.

Fatores de risco para óbito em Unidade de Terapia Intensiva Neonatal

OBJETIVO:Estimar fatores de risco para óbito durante internação em uma Unidade de Terapia Intensiva Neonatal (UTIN) por modelo logístico hierarquizado. MÉTODOS: Trata-se de estudo observacional, analítico e longitudinal com recém-nascidos internados na UTIN de um hospital universitário, no período de janeiro/2000 a dezembro/2003. A variável dependente foi óbito intra-hospitalar e as independentes foram variáveis antenatais, perinatais e pós-natais. Criou-se um modelo hierarquizado em três níveis. Realizada a análise bivariada, foram incluídas no modelo as que apresentavam p<0,20 e mantidas se p<0,05. O procedimento utilizou o programa SPSS 10.0 para a análise e estimativa da acurácia, adotando-se nível de significância de 5%. RESULTADOS: Foram incluídos no estudo 367 recém-nascidos, tendo ocorrido 69 (18,8%) óbitos no período. As variáveis com significância estatística que compuseram o modelo final foram: relato de natimorto anterior, ordem de nascimento, Apgar de quinto minuto inferior a 7, idade gestacional inferior a 37 semanas e ventilação mecânica durante a internação. O modelo apresentou acurácia de 86,9%. CONCLUSÕES: O modelo obtido neste estudo contém variáveis dos três níveis hierárquicos e poderá ser utilizado em Unidades de Terapia Intensiva Neonatal que apresentem comportamento semelhante à unidade na qual se realizou este estudo.

Aplicação do escore CRIB como preditor de óbito em unidade de terapia intensiva neonatal: uma abordagem ampliada

OBJETIVOS: avaliar o uso do escore CRIB (Clinical Risk Index for Babies) em todos os recém-nascidos internados em Unidade de Terapia Intensiva Neonatal (UTIN) e comparar seus resultados com peso ao nascer e idade gestacional. MÉTODOS: estudo observacional, envolvendo todos os recém-nascidos internados na UTIN do Hospital Universitário de Taubaté. As variáveis foram escore CRIB, peso ao nascer, idade gestacional, uso de surfactante, cateterização umbilical, asfixia neonatal e óbito. Foram comparadas as médias do escore CRIB, peso ao nascer e idade gestacional segundo óbito. Foram estimados os valores da sensibilidade, especificidade, valores preditivos positivo e negativo e risco relativo e criadas curvas Receiver Operating Characteristic (ROC) para CRIB, peso ao nascer e idade gestacional. Utilizou-se da técnica t de Student e qui-quadrado de tendência linear. A significância estatística foi alfa = 5%. RESULTADOS: óbito esteve associado a maiores valores do CRIB; houve tendência de mais casos com asfixia, uso de surfactante, cateterização umbilical e óbitos com as classes maiores do CRIB. A curva ROC relativa ao CRIB foi maior que as relativas ao peso ao nascer e idade gestacional. CONCLUSÕES: o escore CRIB foi bom preditor do óbito quando aplicado em todos os RN.

Fuzzy expert system in the prediction of neonatal resuscitation

In view of the importance of anticipating the occurrence of critical situations in medicine, we propose the use of a fuzzy expert system to predict the need for advanced neonatal resuscitation efforts in the delivery room. This system relates the maternal medical, obstetric and neonatal characteristics to the clinical conditions of the newborn, providing a risk measurement of need of advanced neonatal resuscitation measures. It is structured as a fuzzy composition developed on the basis of the subjective perception of danger of nine neonatologists facing 61 antenatal and intrapartum clinical situations which provide a degree of association with the risk of occurrence of perinatal asphyxia. The resulting relational matrix describes the association between clinical factors and risk of perinatal asphyxia. Analyzing the inputs of the presence or absence of all 61 clinical factors, the system returns the rate of risk of perinatal asphyxia as output. A prospectively collected series of 304 cases of perinatal care was analyzed to ascertain system performance. The fuzzy expert system presented a sensitivity of 76.5% and specificity of 94.8% in the identification of the need for advanced neonatal resuscitation measures, considering a cut-off value of 5 on a scale ranging from 0 to 10. The area under the receiver operating characteristic curve was 0.93. The identification of risk situations plays an important role in the planning of health care. These preliminary results encourage us to develop further studies and to refine this model, which is intended to implement an auxiliary system able to help health care staff to make decisions in perinatal care.