RESUMO

BACKGROUND: No-show to medical appointments has significant adverse effects on healthcare systems and their clients. Using machine learning to predict no-shows allows managers to implement strategies such as overbooking and reminders targeting patients most likely to miss appointments, optimizing the use of resources. METHODS: In this study, we proposed a detailed analytical framework for predicting no-shows while addressing imbalanced datasets. The framework includes a novel use of z-fold cross-validation performed twice during the modeling process to improve model robustness and generalization. We also introduce Symbolic Regression (SR) as a classification algorithm and Instance Hardness Threshold (IHT) as a resampling technique and compared their performance with that of other classification algorithms, such as K-Nearest Neighbors (KNN) and Support Vector Machine (SVM), and resampling techniques, such as Random under Sampling (RUS), Synthetic Minority Oversampling Technique (SMOTE) and NearMiss-1. We validated the framework using two attendance datasets from Brazilian hospitals with no-show rates of 6.65% and 19.03%. RESULTS: From the academic perspective, our study is the first to propose using SR and IHT to predict the no-show of patients. Our findings indicate that SR and IHT presented superior performances compared to other techniques, particularly IHT, which excelled when combined with all classification algorithms and led to low variability in performance metrics results. Our results also outperformed sensitivity outcomes reported in the literature, with values above 0.94 for both datasets. CONCLUSION: This is the first study to use SR and IHT methods to predict patient no-shows and the first to propose performing z-fold cross-validation twice. Our study highlights the importance of avoiding relying on few validation runs for imbalanced datasets as it may lead to biased results and inadequate analysis of the generalization and stability of the models obtained during the training stage.

Assuntos

RESUMO

This study aimed to evaluate the effect of hydrolysis conditions on non-extractable phenolic compounds (NEPC) composition of grape peel and seed powder. The effect of temperature (50-90 °C), hydrochloric acid concentration (0.1-15.0 %), and time (5-20 min) were evaluated to understand their impact on NEPC release/extraction and degradation. The use of 1.0 and 8.0 % of HCl concentrations (v/v) and temperatures of 65 and 80 °C produced extracts with higher concentrations and a larger set of compounds. These conditions promoted a balance between release/extraction and degradation processes, thereby maximizing the NEPC content in the extracts. Furthermore, the results suggest that hydrolysis conditions can be set to modulate the release of specific classes. Non-extractable proanthocyanidins showed higher concentrations when intermediate values of temperature and acid concentration were applied. Hydrolysable tannins and hydroxybenzoic acids, on the other hand, were better extracted using higher acid concentrations and higher temperatures. The results suggest that the concentration and composition of NEPC are influenced by the hydrolysis conditions and the type of matrix. Hence, it is crucial to account for this compositional variation when conducting research on the biological effects of NEPC and when using this fraction as supplements or food ingredients.

Assuntos

RESUMO

The comprehensive composition of phenolic compounds (PC) from seven genotypes of guabiju were analyzed by high-performance liquid chromatography coupled to a diode array detector and mass spectrometry (HPLC-ESI-qTOF-MS/MS), and a targeted metabolomic approach was utilized to explore the PC-related similarities among the genotypes. Sixty-seven phenolic compounds were annotated and twenty-four were quantified in all genotypes of guabiju. The phenolic acids and anthocyanins were the major PC, representing more than 63% (w/w) of the total PC. Di-O-galloylquinic and tri-O-galloylquinic acids and ellagitannins were reported for the first time in guabiju. The results of hierarchical clustering and principal components analysis (PCA) suggested seven groups as suitable clusters to be formed according to phenolic composition. Eleven PC were selected as relevant for sample clustering, and six of them were highlighted as the most informative (in decreasing order of importance): epicatechin, catechin, (epi)gallocatechin gallate II, di-O-galloylquinic acid I, tri-O-galloylquinic acid and delphinidin 3-O-glucoside. To the best of our knowledge, this study contributes to the literature with the most complete phenolic profile of guabiju genotypes up to date. Moreover, guabiju susceptibility to fungal infestation related to PC composition was briefly discussed based on a parallel study using the same genotypes.

Assuntos

RESUMO

Several approaches to assess the authenticity of food products have been developed, given that fraudulent products may impact consumers' confidence, affect commercial trades and lead to health risks. This paper proposes an approach to identify the chemical elements that optimally discriminate rice samples according to their producing region in the South of Brazil, the largest rice producer outside Asia. A combinatorial procedure on the concentration of 26 elements determined using inductively coupled mass spectrometry (ICP-MS) and liquid chromatography hyphenated with ICP-MS from 640 rice samples was coupled with Support Vector Machine. The assessed elements included nonmetal and metal elements of 3 types of rice collected from 5 rice-producing regions. The framework selected Mn, Fe, Cu, Zn, Ni, Mo, Cd, Cs, As, Rb, Se, and iAs as the most informative elements for tracking samples' origin. The concentration of such elements is strongly affected by fertilization procedures and soil composition.

Assuntos

RESUMO

PURPOSE: There are no criteria to establish priority for bariatric surgery candidates in the public health system in several countries. The aim of this study is to identify preoperative characteristics that allow predicting the success after bariatric surgery. MATERIALS AND METHODS: Four hundred and sixty-one patients submitted to Roux-en-Y gastric bypass were included. Success of the surgery was defined as the sum of five outcome variables, assessed at baseline and 12 months after the surgery: excess weight loss, use of continuous positive airway pressure (CPAP) or bilevel positive airway pressure (BiPAP) as a treatment for obstructive sleep apnea (OSA), daily number of antidiabetics, daily number of antihypertensive drugs, and all-cause mortality. Partial least squares (PLS) regression and multiple linear regression were performed to identify preoperative predictors. We performed a 90/10 split of the dataset in train and test sets and ran a leave-one-out cross-validation on the train set and the best PLS model was chosen based on goodness-of-fit criteria. RESULTS: The preoperative predictors of success after bariatric surgery included lower age, presence of non-alcoholic fatty liver disease and OSA, more years of CPAP/BiPAP use, negative history of cardiovascular disease, and lower number of antihypertensive drugs. The PLS model displayed a mean absolute percent error of 0.1121 in the test portion of the dataset, leading to accurate predictions of postoperative outcomes. CONCLUSION: This success index allows prioritizing patients with the best indication for the procedure and could be incorporated in the public health system as a support tool in the decision-making process.

Assuntos

RESUMO

BACKGROUND: Surgical theater (ST) operations planning is a key subject in the healthcare management literature, particularly the scheduling of procedures in operating rooms (ORs). The OR scheduling problem is usually approached using mathematical modeling and made available to ST managers through dedicated software. Regardless of the large body of knowledge on the subject, OR scheduling models rarely consider the integration of OR downstream and upstream facilities and resources or validate their propositions in real life, rather using simulated scenarios. We propose a heuristic to sequence surgeries that considers both upstream and downstream resources required to perform them, such as surgical kits, post anesthesia care unit (PACU) beds, and surgical teams (surgeons, nurses and anesthetists). METHODS: Using hybrid flow shop (HFS) techniques and the break-in-moment (BIM) concept, the goal is to find a sequence that maximizes the number of procedures assigned to the ORs while minimizing the variance of intervals between surgeries' completions, smoothing the demand for downstream resources such as PACU beds and OR sanitizing teams. There are five steps to the proposed heuristic: listing of priorities, local scheduling, global scheduling, feasibility check and identification of best scheduling. RESULTS: Our propositions were validated in a high complexity tertiary University hospital in two ways: first, applying the heuristic to historical data from five typical ST days and comparing the performance of our proposed sequences to the ones actually implemented; second, pilot testing the heuristic during ten days in the ORs, allowing a full rotation of surgical specialties. Results displayed an average increase of 37.2% in OR occupancy, allowing an average increase of 4.5 in the number of surgeries performed daily, and reducing the variance of intervals between surgeries' completions by 55.5%. A more uniform distribution of patients' arrivals at the PACU was also observed. CONCLUSIONS: Our proposed heuristic is particularly useful to plan the operation of STs in which resources are constrained, a situation that is common in hospital from developing countries. Our propositions were validated through a pilot implementation in a large hospital, contributing to the scarce literature on actual OR scheduling implementation.

Assuntos

RESUMO

This article aims to develop and validate a multivariate model for quantifying Robusta-Arabica coffee blends by combining near infrared spectroscopy (NIRS) and total reflection X-ray fluorescence (TXRF). For this aim, 80 coffee blends (0.0-33.0%) were formulated. NIR spectra were obtained in the wavenumber range 11100-4950 cm-1 and 14 elements were determined by TXRF. Partial least squares models were built using data fusion at low and medium levels. In addition, selection of predictive variables based on their importance indices (SVPII) improved results. The best model reduced the number of variables from 1114 to 75 and root mean square error of prediction from 4.1% to 1.7%. SVPII selected NIR regions correlated with coffee components, and the following elements were chosen: Ti, Mn, Fe, Cu, Zn, Br, Rb, Sr. The model interpretation took advantage of the data fusion between atomic and molecular spectra in order to characterize the differences between these coffee varieties.

RESUMO

AIM: Analysis of the use of ophthalmic instruments during surgical procedures in order to propose a material management method. METHOD: Mixed method study, sequential exploratory design, performed from January to June 2015, at a university hospital in southern Brazil. First, a qualitative approach was held from brainstorming and field observation. Themes were grouped into thematic categories. By connection, the quantitative stage happened through matrix arrangement and linear programming, culminating in the instrument management proposal. RESULTS: Given categories - instruments reorganization according to the time of the surgical procedure and the need surgical instruments for in each procedure - guided the definition of existing restrictions and application of mathematical models. There was an average reduction of 13.10% in the number of surgical instruments per tray and an increase of 17.88% in surgical production. FINAL CONSIDERATIONS: This proposal allowed the rationalization and optimization of ophthalmic instruments, favoring sustainability of the organization.

Assuntos

RESUMO

Several statistical-based approaches have been developed to support medical personnel in early breast cancer detection. This article presents a method for feature selection aimed at classifying cases into categories based on patients' breast tissue measures and protein microarray. The effectiveness of this feature selection strategy was evaluated against the commonly used Wisconsin Breast Cancer Database-WBCD (with several patients and fewer features) and a new protein microarray data set (with several features and fewer patients). Features were ranked according to a feature importance index that combines parameters emerging from the unsupervised method of principal component analysis and the supervised method of Bhattacharyya distance. Observations of a training set were iteratively categorized into malignant and benign cases through 3 classification techniques: k-Nearest Neighbor, linear discriminant analysis, and probabilistic neural network. After each classification, the feature with the smallest importance index was removed, and a new categorization was carried out until there was only one feature left. The subset yielding maximum accuracy was used to classify observations in the testing set. Our method yielded average 99.17% accurate classifications in the testing set while retaining average 4.61 out of 9 features in the WBCD, which is comparable to the best results reported by the literature on that data set, with the advantage of relying on simple and widely available multivariate techniques. When applied to the microarray data, the method yielded average accuracy of 98.30% while retaining average 2.17% of the original features. Our results can aid health-care professionals during early diagnosis of breast cancer.

Assuntos

RESUMO

More than one million people die or suffer non-fatal injuries annually due to road accidents around the world. Understanding the causes that give rise to different types of conflict events, as well as their characteristics, can help researchers and traffic authorities to draw up strategies aimed at mitigating collision risks. This paper proposes a framework for grouping traffic conflicts relying on similar profiles and factors that contribute to conflict occurrence using self-organizing maps (SOM). In order to improve the quality of the formed groups, we developed a novel variable importance index relying on the outputs of the nonlinear principal component analysis (NLPCA) that intends to identify the most informative variables for grouping collision events. Such index guides a backward variable selection procedure in which less relevant variables are removed one-by-one; after each removal, the clustering quality is assessed via the Davies-Bouldin (DB) index. The proposed framework was applied to a real-time dataset collected from a Brazilian highway aimed at allocating traffic conflicts into groups presenting similar profiles. The selected variables suggest that lower average speeds, which are typically verified during congestion events, contribute to conflict occurrence. Higher variability on speed (denoted by high standard deviation, and speed's coefficient of variation levels on that variable), which are also perceived in the assessed freeway near to congestion periods, also contribute to conflicts.

Assuntos

RESUMO

In this paper, we propose a novel framework to select the most relevant X-Ray Fluorescence (XRF) energy values (i.e., features) to enhance the clustering (grouping) of counterfeit and illicit medical tablets. The framework is based on the integration of multidimensional scaling (MDS) and Procrustes analysis (PA) multivariate techniques. MDS provides a projection of the original data into a lower dimension, while PA finds a projection matrix from the original data. Such outputs give rise to a feature importance index that guides an iterative feature selection process; after each feature is inserted in the subset, an optimization procedure based on a greedy search method is carried out to maximize the clustering quality assessed through the Silhouette Index (SI). The inorganic chemical fingerprinting of 41 commercial samples (Viagra®, Cialis®, Lazar®, Libiden®, Maxfil®, Plenovit®, Potent 75®, Rigix®, V-50®, Vimax® and Pramil®) and 56 seized counterfeit samples (Viagra and Cialis) was used to validate the proposed framework. From the original 2048 data points in the full spectra, we identified a subset comprised of 41 energy values that substantially improved clustering quality; the obtained groups were assessed by visual inspection of the PCA plots.

Assuntos

RESUMO

Phenolic and nitrogenous compounds from different styles craft beers were identified by high performance liquid chromatography and mass spectrometry in order to stratify beer samples according to their style. For this, an exploratory assessment relying on Linear Discriminant Analysis was performed. Fifty-seven phenolic compounds were reported and twelve of them were found for the first time in beer: benzoic acids, 2,4-dihydroxybenzoic acid, 2,3-dihydroxybenzoic acid, dimethoxybenzoic acid; phenolic acid conjugates, 3-p-coumaroylquinic acid, 4-p-coumaroylquinic acid, 3-feruloylquinic acid, 4-feruloylquinic acid, 5-feruloylquinic acid; flavonoids, taxifolin hexoside, quercetin dihexoside, apigenin-6,8-dipentoside, and isofraxidin hexoside. Additionally, 11 nitrogenous compounds belonging to the phenolamide class were found. Two discriminant functions were generated and allowed a satisfactory separation among all beer styles. 3-Caffeoylquinic acid, 3-p-coumaroylquinic acid, 4-p-coumaroylquinic acid, 5-caffeoylquinic acid, coumaric acid, kaempferol-3-O-rutinoside, proanthocyanidin B dimer III and proanthocyanidin B dimer V were the compounds that showed the highest capacity of discriminate the beer styles (IPA, Lager and Weiss).

Assuntos

RESUMO

Erectile dysfunction medicines such as Cialis and Viagra are very popular worldwide and are between the most prevalent counterfeit medicines in Brazil. A range of analytical methods has been used to analyze Cialis and Viagra, such as ATR-FTIR, GCMS and UPLC-MS. Until now, there are no data available of DSC methods for analysis of counterfeit medicines of Cialis and Viagra. DSC is a thermal analysis that provides useful information of physico-chemical events, and however is almost not used for forensic purposes. In this study, thermal analysis of 25 counterfeit Viagra and Cialis seized by Brazilian Federal Police were performed by DSC and compared to their authentic medicines and analytical standards, along with chemometric tools. Authentic samples of Viagra displayed a similar thermal profile with the API, while Cialis were different with additional endothermic peaks, that could be related to excipients interference. Thermograms of Viagra counterfeit samples were similar to authentic samples, while Cialis showed an enlargement and displacement of endothermic peaks. Also, some Cialis counterfeit samples showed melting peaks attributed to sildenafil, the API of Viagra, instead tadalafil, confirming previous results obtained by UPLC-MS. Multivariate analysis with application of Hierarchical Cluster Analysis classified different groups of samples, including a cluster with counterfeit Cialis and Viagra, indicating the use of same API for both counterfeit medicines and possibly the same illicit production; and a cluster with authentic Viagra and counterfeit Cialis, confirming the addition of sildenafil instead tadalafil to Cialis counterfeit samples. Here for the first time we described the use of DSC for chemical profiling of Cialis and Viagra and showed that even when applied to a small group of samples, DSC along with chemometric tools can be considered as a good auxiliary method in forensic casework samples. DSC provided useful data to perform the identification of counterfeit and authentic medicines, with low cost and a simple method.

Assuntos

RESUMO

BACKGROUND: In this article, we propose a method that integrates systematic layout planning techniques to lean health care practices aided by multicriteria decision analysis that could be applied to reformulate the layout of health care facilities. METHODS: We analyze a high-variety sterilization unit of a large public hospital located in Brazil. The unit is currently implementing lean practices, and layout changes are required to provide more efficient materials and information flows. RESULTS: Traditional design of health care facilities is not aligned with lean implementation and its underlying practices and principles. We propose the integration of such approaches to enhance their benefits. To rank and select the best layout alternative, a multicriteria decision analysis method (analytic hierarchy process) is adopted. CONCLUSIONS: There are 3 contributions here: the integration of lean principles into traditional health care facility design practices, the use of multicriteria decision analysis to refine the determination of the best layout solution, and the application of our propositions in a real case study.

Assuntos

RESUMO

Objetivo: Relatar a experiência de desenvolver uma sistemática para racionalização de instrumentais em bandejas cirúrgicas. Método: Estudo de desenvolvimento de sistemática para racionalização de instrumentais, realizado em 2015, a partir do método qualitativo, em um centro de materiais e esterilização (CME) de um hospital universitário federal de Porto Alegre, Brasil. Resultados: Houve redução média do quantitativo de instrumentais em bandejas institucionais em 10,92%; diminuição de bandejas de propriedade das equipes médicas, sendo 84,06% pertencentes à equipe da otorrinolaringologia; e inativação definitiva de 369 instrumentais da cirurgia ortopédica, o que significou 72,84% do total dos instrumentais inativados. Além disso, houve condução de melhorias no gerenciamento de instrumentais, otimização do tempo de preparo e redução da esterilização por expiração do prazo de utilização. Conclusão: A realocação de instrumentais e o acréscimo de peças em bandejas específicas permitiu a reavaliação das solicitações de compras de instrumentais e a melhoria das relações entre as equipes. Essa sistemática contribuiu significativamente para o gerenciamento de instrumentais, otimizando processos e envolvendo as equipes cirúrgicas no trabalho do CME e evidenciou que pode ser aplicada em outras instituições.

Objective: To report the experience of developing a systematic approach for the rationalization of instruments in surgical trays. Method: Study of the development of a systematic approach for the rationalization of instruments, carried out in 2015, using a qualitative method, in the Central Sterile Supply Department (CSSD) of a federal university hospital in Porto Alegre, Brazil. Results: There was a 10.92% average reduction in the number of instruments in institutional trays, a reduction in the number of trays owned by medical teams ­ 84.06% belonged to the otorhinolaryngology team ­ and a definitive inactivation of 369 orthopedic surgery instruments, which represented 72.84% of the total number of inactivated instruments. In addition, improvements were made to the management of instruments, the optimization of preparation time and the reduction of sterilization by expiration date. Conclusion: The relocation of instruments and the addition of items in specific trays allowed for the reappraisal of requests for purchase of instruments and the improvement of relationships between the teams. This systematic approach contributed significantly to the management of instruments, the optimizing processes and the involvement of the surgical teams in the work of the CSSD, thus demonstrating that it can be applied in other institutions.

Objetivo: Relatar la experiencia de desarrollar una sistemática para racionalización de instrumentales en bandejas quirúrgicas. Método: Estudio de desarrollo de sistemática para racionalización de instrumentales, realizado en 2015, desde el método cualitativo, en un centro de materiales y esterilización (CSSD) de un hospital universitario federal de Porto Alegre, Brasil. Resultados: Hubo reducción media del cuantitativo de instrumentales en bandejas institucionales en el 10,92%; disminución de bandejas de propiedad de los equipos médicos, siendo el 84,06% pertenecientes al equipo de la otorrinolaringología; e inactivación definitiva de 369 instrumentales de la cirugía ortopédica, lo que significó el 72,84% del total de los instrumentales inactivados. Además, hubo conducción de mejoras en el gerenciamiento de instrumentales, optimización del tiempo de preparo y reducción de la esterilización por expiración del plazo de utilización. Conclusión: La reubicación de instrumentales y el incremento de piezas en bandejas específicas permitió la reevaluación de las solicitaciones de compras de instrumentales y la mejora de las relaciones entre los equipos. Esa sistemática contribuyó significativamente para el gerenciamiento de instrumentales, perfeccionando procesos e involucrando a los equipos quirúrgicos en el trabajo de CSSD y evidenció que puede aplicarse en otras instituciones.

Assuntos

RESUMO

In forensic and pharmaceutical scenarios, the application of chemometrics and optimization techniques has unveiled common and peculiar features of seized medicine and drug samples, helping investigative forces to track illegal operations. This paper proposes a novel framework aimed at identifying relevant subsets of attenuated total reflectance Fourier transform infrared (ATR-FTIR) wavelengths for classifying samples into two classes, for example authentic or forged categories in case of medicines, or salt or base form in cocaine analysis. In the first step of the framework, the ATR-FTIR spectra were partitioned into equidistant intervals and the k-nearest neighbour (KNN) classification technique was applied to each interval to insert samples into proper classes. In the next step, selected intervals were refined through the genetic algorithm (GA) by identifying a limited number of wavelengths from the intervals previously selected aimed at maximizing classification accuracy. When applied to Cialis®, Viagra®, and cocaine ATR-FTIR datasets, the proposed method substantially decreased the number of wavelengths needed to categorize, and increased the classification accuracy. From a practical perspective, the proposed method provides investigative forces with valuable information towards monitoring illegal production of drugs and medicines. In addition, focusing on a reduced subset of wavelengths allows the development of portable devices capable of testing the authenticity of samples during police checking events, avoiding the need for later laboratorial analyses and reducing equipment expenses. Theoretically, the proposed GA-based approach yields more refined solutions than the current methods relying on interval approaches, which tend to insert irrelevant wavelengths in the retained intervals. Copyright © 2016 John Wiley & Sons, Ltd.

Assuntos

RESUMO

Real-time collision risk prediction models relying on traffic data can be useful in dynamic management systems seeking at improving traffic safety. Models have been proposed to predict crash occurrence and collision risk in order to proactively improve safety. This paper presents a multivariate-based framework for selecting variables for a conflict prediction model on the Brazilian BR-290/RS freeway. The Bhattacharyya Distance (BD) and Principal Component Analysis (PCA) are applied to a dataset comprised of variables that potentially help to explain occurrence of traffic conflicts; the parameters yielded by such multivariate techniques give rise to a variable importance index that guides variables removal for later selection. Next, the selected variables are inserted into a Linear Discriminant Analysis (LDA) model to estimate conflict occurrence. A matched control-case technique is applied using traffic data processed from surveillance cameras at a segment of a Brazilian freeway. Results indicate that the variables that significantly impacted on the model are associated to total flow, difference between standard deviation of lanes' occupancy, and the speed's coefficient of variation. The model allowed to asses a characteristic behavior of major Brazilian's freeways, by identifying the Brazilian typical heterogeneity of traffic pattern among lanes, which leads to aggressive maneuvers. Results also indicate that the developed LDA-PCA model outperforms the LDA-BD model. The LDA-PCA model yields average 76% classification accuracy, and average 87% sensitivity (which measures the rate of conflicts correctly predicted).

Assuntos

RESUMO

In the majority of countries, breast cancer among women is highly prevalent. If diagnosed in the early stages, there is a high probability of a cure. Several statistical-based approaches have been developed to assist in early breast cancer detection. This paper presents a method for selection of variables for the classification of cases into two classes, benign or malignant, based on cytopathological analysis of breast cell samples of patients. The variables are ranked according to a new index of importance of variables that combines the weighting importance of Principal Component Analysis and the explained variance based on each retained component. Observations from the test sample are categorized into two classes using the k-Nearest Neighbor algorithm and Discriminant Analysis, followed by elimination of the variable with the index of lowest importance. The subset with the highest accuracy is used to classify observations in the test sample. When applied to the Wisconsin Breast Cancer Database, the proposed method led to average of 97.77% in classification accuracy while retaining an average of 5.8 variables.

Assuntos

RESUMO

Na maioria dos países, o câncer de mama entre as mulheres é predominante. Se diagnosticado precocemente, apresenta alta probabilidade de cura. Diversas abordagens baseadas em Estatística foram desenvolvidas para auxiliar na sua detecção precoce. Este artigo apresenta um método para a seleção de variáveis para classificação dos casos em duas classes de resultado, benigno ou maligno, baseado na análise citopatológica de amostras de célula da mama de pacientes. As variáveis são ordenadas de acordo com um novo índice de importância de variáveis que combina os pesos de importância da Análise de Componentes Principais e a variância explicada a partir de cada componente retido. Observações da amostra de treino são categorizadas em duas classes através das ferramentas k-vizinhos mais próximos e Análise Discriminante, seguida pela eliminação da variável com o menor índice de importância. Usa-se o subconjunto com a máxima acurácia para classificar as observações na amostra de teste. Aplicando ao Wisconsin Breast Cancer Database, o método proposto apresentou uma média de 97,77% de acurácia de classificação, retendo uma média de 5,8 variáveis.

In the majority of countries, breast cancer among women is highly prevalent. If diagnosed in the early stages, there is a high probability of a cure. Several statistical-based approaches have been developed to assist in early breast cancer detection. This paper presents a method for selection of variables for the classification of cases into two classes, benign or malignant, based on cytopathological analysis of breast cell samples of patients. The variables are ranked according to a new index of importance of variables that combines the weighting importance of Principal Component Analysis and the explained variance based on each retained component. Observations from the test sample are categorized into two classes using the k-Nearest Neighbor algorithm and Discriminant Analysis, followed by elimination of the variable with the index of lowest importance. The subset with the highest accuracy is used to classify observations in the test sample. When applied to the Wisconsin Breast Cancer Database, the proposed method led to average of 97.77% in classification accuracy while retaining an average of 5.8 variables.

Assuntos

RESUMO

Attenuated total reflectance (ATR), a sampling technique by Fourier transform infrared (FTIR) spectroscopy, has been adopted as an analytical tool for detecting fraudulent medicines. The spectrum generated by FTIR-ATR typically relies on hundreds of equally spaced wavenumbers which may reduce the performance of techniques tailored to classify samples into classes, i.e., authentic or fraudulent. This paper proposes a novel method for selecting subsets of wavenumbers (variables) that better classify samples into such classes. For that matter, principal components analysis (PCA) is integrated to the k-nearest neighbor (KNN) classification technique. PCA is applied to FTIR-ATR data, and a variable importance index is built on the PCA outputs. An iterative backward variable elimination is started guided by that index; after each variable removal, samples are categorized into authentic or fraudulent classes using KNN, and the classification accuracy is measured. The wavenumber subset compromising high accuracy and reduced percent of retained variables is chosen. When applied to Cialis FTIR-ATR data, the proposed approach retained only average 1.84% of the original variables and increased the classification accuracy average 2.1%, to 0.9897 from 0.9689; as for Viagra data, the method increased average classification accuracy 1.56%, from 0.9135 to 0.9278, using only 7.72% of the original variables.

Assuntos