The use of the artificial neural network for three-month prognosis in diabetic foot syndrome

Abstract

AimsMaking a reliable prognosis in new patients with diabetic foot syndrome (DFS) is challenging. We used the artificial neural network (ANN) to identify the patients who didn't heal in three months. We provided data for an application which helps predict the course of healing in DFS. Methods175 in-patients (213 limbs) with DFS ulcerations were enrolled in this prospective observational study and were followed up for three months. Thirty-five clinical variables were included in the statistical analysis. ResultsSix significant variables predicting the outcome of DFS treatment were identified: probe-to-bone test, presence of blood flow in Doppler probe, prior amputation within the foot, erythrocyte sedimentation rate, the area and duration of the ulceration. ANN was created with nine input neurons, six hidden nodes and two output neurons. The area under the ROC curve was 0.85. The total accuracy was 82.21 %, sensitivity 91.6 %, specificity 66.18 %. ConclusionsANN as a new prognosis method in DFS ulcers can be reliably used in the prediction, helping physicians and patients predict the course and outcome of the treatment. The algorithm can be particularly useful in identifying individuals who fail to be healed.