A different way to diagnosis acute appendicitis: machine learning.
Pol Przegl Chir
; 96(2): 38-43, 2023 Oct 13.
Article
em En
| MEDLINE
| ID: mdl-38629278
ABSTRACT
<b><br>Indroduction</b> Machine learning is a branch of artificial intelligence based on the idea that systems can learn from data, identify patterns, and make decisions with minimal human intervention.</br> <b><br>Aim:
</b> Our aim is to predict acute appendicitis, which is the most common indication for emergency surgery, using machine learning algorithms with an easy and inexpensive method.</br> <b><br>Materials andmethods:
</b> Patients who were treated surgically with a prediagnosis of acute appendicitis in a single center between 2011 and 2021 were analyzed. Patients with right lower quadrant pain were selected. A total of 189 positive and 156 negative appendectomies were found. Gender and hemogram were used as features. Machine learning algorithms and data analysis were made in Python (3.7) programming language.</br> <b><br>Results:
</b> Negative appendectomies were found in 62% (n = 97) of the women and in 38% (n = 59) of the men. Positive appendectomies were present in 38% (n = 72) of the women and 62% (n = 117) of the men. The accuracy in the test data was 82.7% in logistic regression, 68.9% in support vector machines, 78.1% in k-nearest neighbors, and 83.9% in neural networks. The accuracy in the voting classifier created with logistic regression, k-nearest neighbor, support vector machines, and artificial neural networks was 86.2%. In the voting classifier, the sensitivity was 83.7% and the specificity was 88.6%.</br> <b><br>Conclusions:
</b> The results of our study show that machine learning is an effective method for diagnosing acute appendicitis. This study presents a practical, easy, fast, and inexpensive method to predict the diagnosis of acute appendicitis.</br>.Palavras-chave
Texto completo:
1
Coleções:
01-internacional
Base de dados:
MEDLINE
Assunto principal:
Apendicite
Limite:
Female
/
Humans
/
Male
Idioma:
En
Revista:
Pol Przegl Chir
Ano de publicação:
2023
Tipo de documento:
Article