The quality and length of life of patients with cystic fibrosis (CF) are determined by a number of factors including the quality of healthcare received by patients, as well as access to drug programs dedicated to this particular disease. The purpose of this paper is to present an overview of changes in the average life expectancy and mortality rate of the CF population in Poland between 2000 and 2018. Furthermore, we would like to evaluate access to healthcare services, including the drug program, guaranteed by public healthcare system, and funded by National Health Fund (NHF). The average life expectancy of patients with CF increased in the period in question from ca. 14.5 ± 7.6–24.5 ± 8.9 years (mean ± SD, p = 0.0001). We have observed a drop in the number of deaths in paediatric age during that period. Despite the increase in life expectancy, the use of health resources in patients with CF, especially the drug program, is dramatically low. Considering the fact that in Poland there was no active countrywide CF registry, now it is possible to estimate the frequency of use of CF healthcare services in various provinces exclusively on the basis of database maintained by the Polish NHF.

Background: Cystic fibrosis is a multi-organ disease, inherited in an autosomal recessive manner, requires systematic imaging tests. The most commonly used diagnostic method due to the low cost and low radiation dose is classic X-ray imaging. For patients with cystic fibrosis, they are performed as part of the annual review, as well as in rapid diagnostics when complications are suspected. In this paper, we describe the use of machine learning in medicine to make a diagnosis with a small number of X-ray images.Methods: Our dataset contains 53 X-ray images of CF patients and 371 healthy patients. We use the Convolutional Neural Network (CNN) to classify X-ray images into two categories: CF patient / healthy one. The accuracy and loss function for training and validation were calculated as well as precision p, recall , specificity s, and Matthews correlation coefficient M_cc for prediction.