The Use of Artificial Intelligence in Esophageal Cancer Diagnosis
Code: G-1747
Authors: Soroush Rastegari ℗, Yasin Razmi, Hossein Ebrahimpour-komleh, Mohammad Shabani *
Schedule: Not Scheduled!
Tag: Cancer Diagnosis & Treatment
Download: Download Poster
Abstract:
Abstract
Background and aims: Esophageal cancer is one of the most devastating types of cancer, with a high mortality rate and poor prognosis. Early detection is crucial for improving survival rates and prognosis. This study aims to explore the application of artificial intelligence (AI) in the early diagnosis and improvement of prognosis for esophageal cancer. Method: In this study, we investigated the application of artificial intelligence models using a comprehensive clinical dataset. The dataset we obtained from Kaggle consists of 3985 records and 61 features, including demographic information and specific characteristics of esophageal cancer. Three machine learning models were developed and evaluated: a deep neural network, a decision tree, and a logistic regression model. Results: The deep neural network model, which included multiple layers with regularization and dropout, achieved 100% accuracy and an area under the curve (AUC) of 1.0, demonstrating excellent performance on the dataset. The decision tree model achieved a high accuracy of 99.7%, but its AUC was 0.52. The logistic regression model showed an accuracy of 65.7% and an AUC of 0.86, indicating its potential in predicting esophageal cancer, though it requires further improvement. Conclusion: These results highlight the potential of artificial intelligence in the early diagnosis of esophageal cancer. However, the deep neural network model demonstrated more promising performance compared to the other models.
Keywords
ArtificialIntelligence, EsophagealCancer, DeepNeuralNetwork, DecisionTree, LogisticRegression
