Computational Intelligence in Hernia Surgery with Classification of Repair Types

Problems of the ventral hernia are very common and their evaluation by computational methods can help with the selection of the most appropriate treatment. This study collected data from the records of the set of more than 2800 patients from different European countries observed during last 11 years (2012-2022) and were collected by specialists in hernia surgery. The majority of patients were treated by the standard surgery approach, with the increasing trend towards robotic surgery. This paper is devoted to a statistical evaluation of the treatment method related to the age of the patients, their body mass index (BMI), and the type of repair. Appropriate mathematical methods are used to extract and classify the selected features.The proposed methodology focuses on computational and machine learning methods. Mesh hernia surgery is classified into two classes: groin hernia repairs (GHR) and primary ventral, incisional ventral and parastomal hernia repairs (PVHR, IVHR and PHR).This paper presents statistics of surgery hernia treatments related to the age of patients and repair technologies with different kinds of meshes. The main conclusions point to classification of repair technologies related to the BMI and the age of the patients. Separation accuracy of GHR surgery from other types of repairs reached 69.2 % and 70 % for the Bayesian and SVM methods, respectively. The proposed methodology suggests a very close interdisciplinary approach and the use of computational intelligence in hernia surgery, with its possible use in the clinical environment.