Features of the diagnostic information processing for congenital lung malformations in newborns for the automated analysis and surgical navigation systems

The article describes the processes of collecting and adapting the diagnostic information necessary for use in automated analysis of three-dimensional images and surgical navigation. The work is carried out on the basis of the NMRC Obstetrics, Gynecology And Perinatology named after V.I. Kulakov of the Ministry of Health of the Russian Federation with the financial support of the Ministry of Science and Education of the Russian Federation (Agreement dated 03.10.2016 No. 14.607.21.0162, unique identifier RFMEFI60716X0162) The work is devoted to the features of collection, segmentation and description of the results of preoperative radiological diagnostics of newborn patients with congenital malformations of the lungs, such as bronchopulmonary sequestration (BS) and congenital cystic adenomatous malformation (CCAM). The goal of the work is the development of standards for the collection, classification and segmentation of various diagnostic information of congenital lung malformations in newborns necessary to use in automated three-dimensional image analysis and surgical navigation. In order to expand the scope of application, it was decided to supplement the data bank with information from the patient’s phenotypic chart, compiled by the clinical geneticist when examining the patient. According to the developed and implemented algorithms we collected and segmented 924 series of images belonging to 148 patients with lung anomalies and 356 series of normal lung. Available text descriptions of the series are reconstructed to the original developed standard. At present, using this data bank, a subsystem of neural network analysis and reconstruction of diagnostic images of newborn patients is being developed, as well as a surgical navigation system for performing endoscopic surgical manipulations on patients for congenital malformations of the lungs.