Abstract
Objetivo
Desarrollo y entrenamiento de una red neuronal que permita clasificar a los pacientes que acuden a la consulta con síntomas de dispepsia en dos grupos: los que muy probablemente tengan enfermedad ulcerosa péptica o reflujo gastroesofágico (RGE) y los que muy probablemente tengan dispepsia funcional o idiopática. Comparación de los resultados obtenidos con la red neuronal y otros clasificadores estadísticos.
Diseño
Estudio retrospectivo.
Emplazamiento
Tres consultas de equipo de atención primaria de ámbito urbano.
Participantes
Ochenta y un pacientes con diagnóstico de dispepsia, a los que se realizó endoscopia digestiva y/o tránsito esofagogastroduodenal (EGD) documentado en la historia clínica.
Método
Entrevista personal con cuestionario predefinido sobre sintomatología y factores de riesgo de patología dispéptica. El análisis de los datos se ha realizado con clasificador determinista, clasificador estadístico y red neuronal basada en un perceptrón multicapa.
Resultados
La red neuronal clasifica correctamente a un 81% de los pacientes, con un valor predictivo negativo (VPN) del 90% y un valor predictivo positivo (VPP) del 80%.
Conclusiones
La red neuronal proporciona excelentes tasas de acierto en la clasificación de los pacientes a partir de la presencia o no de determinados síntomas. Se observa una tendencia a discriminar mejor los diagnósticos negativos (dispepsia funcional o idiopática) frente a los positivos (enfermedad ulcerosa péptica o RGE). El uso sistemático de redes neuronales en las consultas de atención primaria facilitaría al clínico su labor, aumentando la rentabilidad de cualquier decisión diagnóstica y terapéutica.
Palabras clave: Redes neuronales, Análisis discriminante, Dispepsia, Dispepsia no ulcerosa, Úlcera péptica
Abstract
Objectives
Development and training of a neurone network that enables the patients who attend the clinic with symptoms of dyspepsia to be classified into two groups: those who very probably have peptic ulcer disease or gastro-oesophageal reflux (GOR) and those more likely to have functional or idiopathic dyspepsia. Results obtained with the neurone network and with other statistical classifiers were compared.
Design
Retrospective study.
Setting
Three urban primary care clinics.
Participants
81 patients with a diagnosis of dyspepsia, who underwent a digestive tract endoscopy and/ or oesophageal-gastroduodenal meal, recorded in the clinical notes.
Method
Face-to-face interview with a set questionnaire on the symptoms and risk factors of dyspepsia pathology. Data were analysed with determinist classifier, statistical classifier and neurone network based on a multi-layer perceptron.
Results
The neurone network correctly classified 81% of patients, with negative predictor value of 90% and positive predictor value of 80%.
Conclusions
The neurone network provides very high accuracy rates in classifying patients on the basis of the presence or otherwise of determined symptoms. There was a tendency to distinguish negative diagnoses (functional or idiopathic dyspepsia) better than positive ones (peptic ulcer disease or GOR). Systematic use of neurone networks in primary care clinics would assist the doctor by increasing the accuracy of diagnostic and/or clinical decisions.
Key-words: Neurone networks, Discriminatory analysis, Dyspepsia, Non-ulcerous dyspepsia, Peptic ulcer
Bibliografía
- 1.Marble R.P., Healy J.C. A neural network approach to the diagnosis of morbidity outcomes in trauma care. Artif Intell Med. 1999;15:299–307. doi: 10.1016/s0933-3657(98)00059-1. [DOI] [PubMed] [Google Scholar]
- 2.Grigsby J., Kramer R.E., Schneiders J.L. Predicting outcome of anterior temporal lobectomy using simulated neural networks. Epilepsia. 1998;39:61–66. doi: 10.1111/j.1528-1157.1998.tb01275.x. [DOI] [PubMed] [Google Scholar]
- 3.Pantazopoulos D., Karakitsos P., Iokim-Liossi A. Back propagation neural network in the discrimination of benign from malignant lower urinary tract lesions. J Urol. 1998;159:1619–1623. doi: 10.1097/00005392-199805000-00057. [DOI] [PubMed] [Google Scholar]
- 4.Armoni A. Use of neural networks in medical diagnosis. MD Comput. 1998;15:100–104. [PubMed] [Google Scholar]
- 5.West D., West V. Improving diagnostic accuracy using a hierarchical network to model decision substasks. Int J Med Inf. 2000;57:41–55. doi: 10.1016/s1386-5056(99)00059-3. [DOI] [PubMed] [Google Scholar]
- 6.Fisher R.S., Parkman H.P. Management of nonulcer dyspepsia. N Engl J Med. 1998;339:1376–1381. doi: 10.1056/NEJM199811053391907. [DOI] [PubMed] [Google Scholar]
- 7.Ingvard W., Tone T.H., Holger U. Discriminant analysis of factors distinguishing patients with functional dyspepsia from patients with duodenal ulcer. Dig Dis Sci. 1995;40:1105–1111. doi: 10.1007/BF02064207. [DOI] [PubMed] [Google Scholar]
- 8.Dayhoff J. Neural network architectures, an introduction. Van Nostrand Reinhold. 1990 [Google Scholar]
- 9.Freedman J.A., Skapura D.M. Neural networks, algorithms, applications and programming techniques. Addison-Wesley. 1991 [Google Scholar]
- 10.Hertz J., Krogh A., Palmer R.G. Introduction to the theory of neural computation. Addison-Wesley. 1991 [Google Scholar]